PDA

View Full Version : Boost Pressure Testing


thebitmaster
06-02-2016, 04:30 AM
So I got the GDE Eco Tune and the swirl valve problem is gone, which closes out this thread:

47465

The vehicle drives *much* better, but unfortunately the turbo pressure problem still exists, though it takes a lot longer to manifest. So I still need to find and correct the leak. I'm fairly sure the turbo is a bit noisier and slightly more grindy-sounding than before this problem occurred. This makes me suspect the resonator or maybe (hopefully not) damage to the blades. That noise stops abruptly when the ECM disengages the turbo, which it now does after about ten minutes of driving instead of nearly immediately before the GDE tune. It's still getting a Turbo Pressure Low code.

So here's what I did to try to find the leak before I got the GDE tune.

1) Create plug to fit into air intake hose for turbo. I did this with a PVC plug and a compressor quick connect, which has some built-in thread seal. I had to buzz down the edges of the PVC to get it to fit. I added about four layers of duct tape around the outside to help smooth out the connection.

2) Plug this device into the air intake hose leading to the turbo, tightening down the hose clamp thoroughly

3) Pressurize my shop compressor to 5 PSI

4) Open the oil fill cap to allow pressure to escape at far end of system

5) Connect compressor to plug, starting airflow

At this point, air eventually comes up out of the oil fill hole. The hose doesn't really inflate. there doesn't seem to be much pressure.

6) Spray all hoses and connectors with soapy water to look for leaks, and listen for hissing anywhere but at the input and the oil fill hole.

I tried this at both 5 and 10 PSI, and didn't find any obvious leaks. At 5 PSI, I tried closing the oil fill cover briefly. This lead to a noticeable inflation of the air tube the compressor was hooked to and minor leakage around where my plug was installed. This leads me to believe other hoses need to be blocked to prevent the air from going through the engine and bypassing the turbo. Also, within about 10 seconds, oil starting to squirt out of the dipstick pillar. I disconnected the compressor to relieve the pressure.

In some of the (non Sprinter) videos I watched to learn this technique, they had to plug additional lines to get the turbo to pressurize correctly. I did not plug any additional lines, and was wondering if the air might be going through those lines directly to the engine, bypassing the turbo.

So my questions are:

1) Which additional lines do I need to plug to do the test properly? I have the GDE tune. Doesn't that close the EGR valve?

2) Is it necessary to disconnect the final line where it goes into the engine and plug that, too, or is it ok to let the throttle body be the plug? I've gotten mixed stories about this online. It looks like the hose at the throttle body end is much more complicated to remove, so I was hoping to avoid that if possible.

I'm already using Feedly to track my fuel economy, and would like to produce better data about the GDE tune's effects, but I can't get clean data until this turbo problem is fixed.

Thanks!

Scott

lindenengineering
06-02-2016, 06:00 AM
Question
Why not simply put a good live data scanner on it and do a function test?
:thinking:
Dennis

thebitmaster
06-02-2016, 08:40 PM
Question
Why not simply put a good live data scanner on it and do a function test?
:thinking:
Dennis

I have an iCarSoft MBII, which can read some live data. Which type of test were you considering? I've watched the boost position sensor, the boost pressure vs. incoming pressure and exhaust pressure, etc. As you'd expect, the boost position sensor starts in the high 90% range and goes between 70-100% when the boost is working. I've seen boost pressures as high as 1600, but most of the time they don't go over 1300. Once the low boost pressure code gets set, you can see it kill the turbo by setting the boost position motor to 5.4%, where it stays until the system is reset. Naturally boost falls off after that. It seems to me like that most likely means there's a leak, and the next step is to find it by pressure testing the system and using whistling noises and soapy water to find it. Right now that's not working because I think additional hoses have to be blocked to prevent air from going into the engine by another path. I don't know the engine well enough yet to know which hoses those might be, though the EGR is obviously one of them.

Is there some way a scanner can isolate the location of the leak? I have the ability to pull live data, but I can't record it. I've got a big car tablet holder I use to hold my iCarsoft MBII where I can see it when I'm in the driver's seat, GPS-style.

Right now my main question are:

1) Which additional hoses do I need to block to pressure test the turbo, resonator, and hoses or
2) What other method of finding the leak might work?

I am gonna try changing the air filter, since that's simple and the oil leaking from the EGR has caused a lot of smoke. Since the GDE tune is in place, that shouldn't happen anymore.

Thanks!

thebitmaster
06-07-2016, 04:41 AM
I've read that a bad backflow pressure sensor could cause a turbo limp mode problem.

Is there somewhere I could find normal values for the Input Pressure, Boost Pressure, Exhaust Back Pressure, etc? I can get those values from my iCarSoft MBII, but I don't know what to expect. My backflow pressure is *definitely* reading substantially higher than my boost pressure pretty much all the time. Is that normal or the sign of a bad sensor? Are there charts I could look at somewhere that show the relationships at common loads? I could also post my values on here for opinions if that would help.

Thanks!

Aqua Puttana
06-07-2016, 11:24 AM
I can't answer your questions.

When I did a pressure test on my 2004 OM647 engine I isolated the charge air system by disconnecting the hoses from the turbo end and the intake manifold end. I then used DIY plugs and compressed air to pressure the system. 15 psi will likely reveal any leaks. I wouldn't exceed 20 psi.

After removal the hoses can be inspected visually. My concern was more for a leak in the intercooler or any plastic bits. Fortunately my system was tight. My LHM problems traced to my O2 sensor.

vic

thebitmaster
06-08-2016, 01:48 AM
So my friend Lee and I did a some snooping while revving the engine in gear with both brakes on, and there are definitely bubbles coming out from the join between the turbo and the metal hose that leads to the resonator. The FSM isn't very helpful as to how, exactly, this piece comes off, but there's at least one screw I can see that has it bolted on to a bracket. Is there some kind of gasket in between those two pieces?

https://c3.staticflickr.com/8/7754/27498081866_976b705450_z.jpg

Any clues how that comes off, or which parts I need to fix it?

Also, the metal shell around the turbo looks...distressed? It's hard to imagine it coming from the factory this beat-up and bent-looking:

https://c3.staticflickr.com/8/7411/27433319082_de3a2bc7ef_z.jpg

Thanks!

Scott

Jimdawdy
06-08-2016, 05:23 AM
My guess is the green o ring between the pipes is damaged and causing the bubbles.

(This tread has good info on removing the pipe and resonator.)
http://sprinter-source.com/forum/showthread.php?t=33956&highlight=glow+plug+replacement

The pipe is retained with a single screw.The resonator is retained by three screws and a clip retained by a screw.



Jim

thebitmaster
06-08-2016, 07:30 PM
Jim, thanks for the link! That's all the info I needed to order a new one and to replace it when the part arrives. Fingers crossed that's the only leak!

The part seems to be "Turbocharger O Ring Seal-3.0L V6 2007-2016". I ordered two from Europarts SD. Will feed back to the list once I've replaced the seal and some some driving!

Scott

My guess is the green o ring between the pipes is damaged and causing the bubbles.

(This tread has good info on removing the pipe and resonator.)
http://sprinter-source.com/forum/showthread.php?t=33956&highlight=glow+plug+replacement

The pipe is retained with a single screw.The resonator is retained by three screws and a clip retained by a screw.



Jim

autostaretx
06-11-2016, 04:26 AM
The Sprinter Store sells an internal re-enforcing piece for the hose ends.
https://sprinterstore.com/product/single-turbo-hose-adapter/

--dick

thebitmaster
06-11-2016, 04:38 AM
So I got the gasket and took the pipe off. The old gasket looked ok, but I switched it anyway. Here's the original before I wiped it off.

http://c2.staticflickr.com/8/7391/27592453585_e991c408c3_n.jpg

I tried to figure out if I should use any lube, but couldn't get a definitive answer, so I used some of the small amount of oil that was inside the tube to lube the new gasket.

After removing the one screw on the pipe and four screws and the clip on the resonator, I dropped my extension and my only torx socket somewhere into the engine. I swore a lot and got bit by mosquitos.

When I put everything back together, I once again noticed how the bracket for the tube did absolutely nothing to hold any tension on the join between it and the turbo where the new gaskets was. This is a picture of that join before I took it apart. Note how much of the inner tube is showing. That entire part shiny part should be covered by the tubes.

http://c7.staticflickr.com/8/7319/26847212190_aba3e45144_n.jpg

Does anyone know why this connection is so weak, only held in by a single screw? Doesn't it make more sense to use a clamp or flange screws? The screw doesn't even increase the tension of the join. Is is a like a safety valve, some kind of pressure release that fails first to save the rest of the system? Is there something I should be doing before tightening the final screw besides pushing the pipe in as hard as I can by hand? Maybe with a ratchet strap?

So when was ready to put the last screw in, the bracket and the screw hole were *way* out of alignment when the joint was pushed together as far as it would go. This picture doesn't fully do it justice.

http://c5.staticflickr.com/8/7294/27558362876_bcbbbe42a8.jpg

It was off on both axes by almost 1/2". I tried adjusting some of the screws on the resonator to give some play, but there was absolutely no way to get the pipe both snugged into the turbo and also line up with the bracket.

The only possible way to get the bracket to be in the right place was to bend it. I gently tapped on it with a hammer and steel rod until it was in what seemed to be the correct position. This mean that the screw was in the middle of the oval hole when the pipes were pushed together, allowing the compression I was adding by pushing it into the turbo. I then tightened it down and put everything else back together.

I did a 20 minute test drive and the turbo worked well. I was getting turbo pressure approximately twice intake pressure on hard acceleration. However, I went back into limp mode shortly thereafter with a current and stored low turbo pressure code. Resetting it only helped for about 15 feet before both code would be reset.

I did a pressure test after I got back, and, sure enough, it was leaking at that seam again. The pipes still look pretty snug, there was no gap, at least on top where I could see. I'm wondering if maybe the two hoses aren't coming together at exactly the right angle, and it's more open at the bottom. But I was seeing the bubbles on the top of the tube.

I went back to work with the hammer so that the bracket was even closer to the turbo. This gives me more ability to compress the joint hard before putting the screw back on. I might try to pull the join together with a ratchet strap before tightening the screw this time. Then I'll do another test drive and report back. If the test drive fails again, I'll take it all apart again in the morning.

Is there something magical I'm missing here? This seems like it should have just been a simple gasket replacement.

Scott

thebitmaster
06-11-2016, 04:43 AM
Dick,

Thanks for letting me know! Those are for the hoses going to and from the intercooler, though, correct? They seem like a worthwhile upgrade if I find a leak in those hoses. Right now, though, I'm dealing with a leak in the join of the the little aluminum pipe that goes from the turbo to the resonator.

Scott

The Sprinter Store sells an internal re-enforcing piece for the hose ends.
https://sprinterstore.com/product/single-turbo-hose-adapter/

--dick

thebitmaster
06-11-2016, 06:19 AM
The Bad News is that when I took the junction apart again, the new gasket was *gone*. There's only one place it could have gone. !!#@#@!!@$ I'm guessing it either got clipped in half somehow or fell out place while I was adjusting the bracket. Would not have thought it possible.

The Good News is that I was a lot more careful on the second assembly, and left all the resonator bolts loose until I had the bracket and pipe in exactly the right place and tightened up. This still required some prying and pounding on the bracket, but not as much. I then tightened all the resonator bolts and went for a test drive.

I actually had two LHM events on first trip out, but I think I might have incorrectly cleared the codes from the DTC due to lateness. However, I decided to give it one more try. I got all the way to Webberville and back, about 13 miles, with no problems. I had the turbo pressure up to 31 PSI with intake at about 13 PSI, tried a variety of speeds and gears, and did one Unnecessary Display of Acceleration from a stop on the way back. Everything works, no codes, and no noticeable effect from the engine having snarfed an o-ring.

It's obviously too early to declare a Total Victory, but if I had a Miller, I'd be drinking it right now.

Thanks for your help!

thebitmaster
06-12-2016, 12:41 AM
Oops! Spoke too soon!

Vehicle went back to fairly quickly losing boost and setting low boost code with driving today. Resetting codes didn't help for more than 20 feet most of the time.

I repeated the stall test and looked for bubbles at the previous problem site, the join between the turbo outlet and the resonator tube. This time I saw no bubbles after two tests. I think maybe that leak is fixed.

Because I'm too terrified to get under the vehicle when it's in gear, I'm exploring this post about pressure testing the lower turbo pipes and the intercooler with a compressor. I realize this post is for a T1N and not my NCV3, but the general idea should be the same.

http://sprinter-source.com/forum/showpost.php?p=161278&postcount=162

One thing I'm noticing during the stall tests is that there is some black smoke coming from the passenger side of the engine, noticeable only at higher load. I'm going to look at night to see if I can narrow down the location. Could that be an intake manifold leak? I'm assuming such a leak would cause loss of turbo pressure, no?

It would be deeply ironic to have gotten the GDE tune to solve the swirl motor problem and have to get the manifolds done anyway!

thebitmaster
06-12-2016, 04:56 AM
So I finally discovered the crankcase ventilation hose on the turbo intake. Back at the beginning of this thread, I tried to pressure test the system by putting air into a plug in the intake tube, but it wasn't working because air was escaping through the engine. I asked "which other hoses do I have to block?" and was unable to get a definitive answer. I think the answer is the EGR hose, which should already be permanently blocked by the GDE tune, and this crankcase ventilation hose. The hose has a *bunch* of sticky crud on it where it joins to the turbo.

It not immediately clear how it is held on. How is this safely removed? I don't see any obvious clamps, and I see zero in the FSM about removing this part. If I can't find out, I'm going to try to gently squeeze the middle shut with a cushioned wood clamp or vice grips. I'm hoping this will allow me to pressurize the entire turbo path from the input to the throttle body without running the engine. Stall tests work well for looking at stuff on the top, but not so well when one has to be in front of or under the vehicle. If it works I'll report back with the steps I used.

72chevy4x4
06-12-2016, 01:09 PM
I don't have any answers for you, but live nearby in RR if you want an extra set of eyes to help out.

as far as a pressurized leak test; if looking for a leak in the IC/hoses, I would first clean the engine with an engine cleaner and hose it off, then try to drive for a bit under boost. My theory is that the pressurized air leaking out during driving conditions will allow some oil to escape if there is a leak.

from experience, lining up the resonator and tube leading to the turbo's exducer can be confusing, it doesn't seem to be tight enough.

I would love to check out your GDE tune, I've been considering something for my (work) van for sometime. A little more pep on some hill country hills would sure help.

thebitmaster
06-12-2016, 07:34 PM
It would be great to make a local connection for diagnostic buddy-up! I'll PM you.

The cleaning and driving idea sounds like a good diagnosis technique. I'd like to get this engine spotless so I can see leaks right away.

At the moment it looks like my Home Depot-plug pressurization system is probably going to work. The main issue I've having right now is air escaping through the crankcase ventilation hose. The near end just pops off, but to block it well I need to close the far end. There seems to be some kind of metal clip or retainer there, but it's not clear how it comes off. I can't find any mention of that hose in the FSM, this forum, or online anywhere.

How does the crankcase side of the breather hose come loose?

Scott

I don't have any answers for you, but live nearby in RR if you want an extra set of eyes to help out.

as far as a pressurized leak test; if looking for a leak in the IC/hoses, I would first clean the engine with an engine cleaner and hose it off, then try to drive for a bit under boost. My theory is that the pressurized air leaking out during driving conditions will allow some oil to escape if there is a leak.

from experience, lining up the resonator and tube leading to the turbo's exducer can be confusing, it doesn't seem to be tight enough.

I would love to check out your GDE tune, I've been considering something for my (work) van for sometime. A little more pep on some hill country hills would sure help.

thebitmaster
06-13-2016, 04:07 AM
How does one disconnect the crankcase end of the crankcase ventilation hose?

The near end is held on by a barbed nipple, but I can't tell exactly how the far and connects and don't want to hurt it.

Instead I put a thin piece of wood on either side of the middle of the tube, then clamped it down with a wood clamp. I'm not certain it was 100% closed, but it was close. I'd want to take off the far end so I can fold it over to clamp it.

I put my plug back into the intake tube for the turbo and proceeded to pressurize it. I started at 5 PSI on the regulator and then went up. The plug popped out a few times. After I took a little tape off and added some bailing wire to hold it in, I got it to stay, though it was leaking just a little. Under this configuration I was able to put some pressure into the system. There was definitely some air coming out of the open oil fill hole, but I couldn't be sure if it was coming through the incompletely closed crankcase hose, the throttle body at the far end, or the EGR valve, which should be closed by the GDE tune.

I did a spritz test with soapy water over the entire pressure path to the throttle body except the intercooler itself. I didn't find any leaks, which was discouraging. It's very hard to see on top of the intercooler hoses. However, I didn't have a pressure gauge after the compressor, so I was never 100% sure of system pressure. Tomorrow I will get an inline gauge. Since I had several places where air was escaping the system, the PSI might not have been high enough to trigger the leak. Also, I did all this with the engine cold. If I can't find any leaks cold at 15 PSI tomorrow, I'll put things back together, warm up the engine, then try again.

Is there any way to detect a leak in the intercooler itself besides listening for hissing? It looks like I'd have to pull the bumper and headlights to get soap water on it in a way I could see bubbles. I know a smoke machine doesn't make enough pressure for turbo leaks. Is there any way to safely generate smoke and inject it into the compressor line?

thebitmaster
06-13-2016, 04:15 AM
When I was replacing the gasket between the turbo and the resonator, I discovered that there was a *very* large gap between the two pipes that I'm pretty certain was the cause of the original leak there. When I put it back together, I tapped the pipe bracket closer to the turbo to close the gap. At this point the ring on the pipe butts directly into the flanges on the turbo with no gap. Is this correct? It doesn't leak there anymore when I do a stall test. After losing one new gasket, I put the old one back because it looked like it was in perfect shape, identical to the new one.

This is how it looked before I opened it up the first time:

https://c7.staticflickr.com/8/7319/26847212190_aba3e45144.jpg

Now that shiny piece of tube is 100% gone, with the outer diameters meeting with no gap. Is that correct?

Thanks!

72chevy4x4
06-14-2016, 11:15 AM
did you locate the other missing (new) gasket?

thebitmaster
06-14-2016, 06:08 PM
I have not yet located the gasket that got blown into the piping. One thing that occurred to me is that I might be able to use the pressure testing to blow it back out if it's still in the turbo intake piping. There is a *very* small chance it might have fallen out when I took the resonator pipe off, but I was specifically looking for it as I pulled the pieces apart, so I don't think that's likely.

My plan is to start with the resonator and take off all the pipes to see if I can find it. I can do a closer visual inspection of each one. I'm not 100% clear on how the pipe end that goes into the throttle body comes off, but that info is probably around somewhere. The FSM has been shockingly unhelpful throughout this entire affair so far.

I got a test plug for the intercooler hose that goes into the throttle body, and once I put everything back together, I'm gonna block that and repeat the pressure test.

Does anyone know how to remove the far end of the crankcase ventilation pipe that starts at the turbo intake and goes to the crankcase?

thebitmaster
06-15-2016, 05:18 AM
I wanted to test the intercooler hoses, and sometimes with a problem like this if you take everything apart, clean it, and put it back together, the problem Goes Away.

Took off the inlet hose, the resonator and pipe, and both intercooler hoses. Turbo itself looks pretty kosher at this point:

https://c7.staticflickr.com/8/7452/27645148766_511abe6d4f.jpg

After cleaning, bending, and visual examination, I pressure tested both intercooler hoses at 15 PSI using this setup. No leaks found. Note that this method is dangerous, those plugs can come out fast enough to cause serious injuries and an amusing *pop*ping noise, which is less funny when you're bleeding.

https://c4.staticflickr.com/8/7756/27680330635_4399f52418.jpg

The gasket on one of the hoses at the metal end was definitely a little looser than the other. In place they looked normal, but if I pinched each side and tried to slide them sideways with the same amount of force, the looser one stuck out about twice as far as the other. These might be a good candidate for replacement.

I discovered some damage on the driver side pipe coming down from the resonator. It's *very* hard to tell if the deformation goes far enough into the pipe to effect the seal:

https://c7.staticflickr.com/8/7792/27067809134_baa14f5775.jpg

It *looks* like the seal is well past the damage, and even the flare at the base of the hose is above it:

https://c3.staticflickr.com/8/7377/27067809034_d45b9e9b1d.jpg

The place where the gasket meets the cone is *way* above this join point. It seems to mate well when assembled and clipped. The good news is that pipe is only $80 new, which for a piece of Mercedes engine metal is pretty cheap.

There was no sign of the lost gasket. I was going to try to look into the intercooler inlet with a dental mirror and flashlight, but got distracted by family health issues. I replaced the gasket on the turbo-to-resonator pipe with a new one, removing the still-looks-fine old one. When I ordered 2 replacement gaskets, they sent me one green and one black. They were the same dimensions, in the same bag, and had the same part number.

I lubed all joins with fresh motor oil before re-assembly.

I verified there were no codes with my iCarSoft MB II before the first test drive. I got a low boost code stored (but not pending) less than two miles from the farm. I did a reset at a gas station, and was then able to drive another 12 miles under a variety of conditions, RPMs, and levels of driver responsibility and everything worked. This is exactly what happened after I did the hammer job on the resonator pipe bracket, and the next day I was back to low boost codes after a few miles. I'm going to go visit my Dad in the hospital tomorrow, which will be a longer test drive.

Since I've verified that the two intercooler hoses are *not* leaking, I'd like to pressure test the intercooler with them hooked up. Hopefully this will allow me to hear hissing from the intercooler *without* having to take the bumper off. If I can find a way to make it *pop* from the intake end, I might get my gasket back!

The problem is that the test plug that works really well for blocking the metal end of the intercooler pipe would be really hard to put a compressor nipple on, unlike the PCV that fits snuggly into the rubber end. There's probably a tool that combines both functions already. If the boost problem continues, I may get one.

Any creative ideas how to get a nipple plug into either the metal end of the intercooler pipe or the intercooler input itself? Maybe a short extension hose with a clamped plug?

Aqua Puttana
06-15-2016, 12:36 PM
I can't answer your questions.

It is good to warn people about the potential missiles if the parts let go. 15 psi doesn't seem like much, but when applied against a decent surface area it can let go with great force.

... Turbo itself looks pretty kosher at this point:

...
I'm not so certain. As a practical matter, it certainly isn't damaged to a point of affecting power/operation, but your blade tips show some erosion. That would hint that there was some dirty air getting in to the intake.

But... it's tough to tell from pictures. Here's another one.

77274

There are more pictures/info here.

http://sprinter-source.com/forum/showthread.php?p=302804#post302804

Any creative ideas how to get a nipple plug into either the metal end of the intercooler pipe or the intercooler input itself? Maybe a short extension hose with a clamped plug?
Put the system back together enough with the already tested good parts and put pressure on the intercooler from there. :idunno:

From my recollection, you have found and corrected some air leaks. The EGR has some history of causing LHM and not throwing EGR codes. Have you checked the turbo actuator linkage for proper operation, or at least lubed it?

vic

thebitmaster
06-15-2016, 05:41 PM
Vic! Honored to have you on my thread, sir. Your posts about pressure testing your T1N were a major resource when trying to figure out how to test my NCV3.

I can't answer your questions.


Ha! I don't believe you!


It is good to warn people about the potential missiles if the parts let go. 15 psi doesn't seem like much, but when applied against a decent surface area it can let go with great force.


Agreed. In my original message I said:


After cleaning, bending, and visual examination, I pressure tested both intercooler hoses at 15 PSI using this setup. No leaks found. Note that this method is dangerous, those plugs can come out fast enough to cause serious injuries and an amusing *pop*ping noise, which is less funny when you're bleeding.


I'm not so certain. As a practical matter, it certainly isn't damaged to a point of affecting power/operation, but your blade tips show some erosion. That would hint that there was some dirty air getting in to the intake.

But... it's tough to tell from pictures. Here's another one.

77274

There are more pictures/info here.

http://sprinter-source.com/forum/showthread.php?p=302804#post302804



I'll check into that thread on turbo wear. I did just change the air filter a few days ago.


Put the system back together enough with the already tested good parts and put pressure on the intercooler from there. :idunno:


That was my intention. The question is, how? I've tried two approaches so far:

1) Pressurize from the main turbo air intake hose. Problem one is that it's hard to seal well, and hard to get my PVC plug out after. I still get some hissing and pop-out. The larger problem is that the little hose that Ts into this one. As long as it's not plugged, air goes into the engine and decent pressure can't be achieved. That's why I want to know how to pull off the far end, because that's the easiest way to plug it. I tried using a wood clamp with a piece of wood on either side of the center, but air was still getting into the engine somehow. I didn't have the far end plugged, I don't know how much air gets past the throttle body into the engine from the end. And despite extensive research, I've been unable to tell exactly what the EGR path is. Despite it being closed by the GDE tune, it might be dirty as you said, and I'd like to be able to isolate that path for testing. If the problem continues, I'm definitely going to try cleaning the EGR valve.

2) test the intercooler by pressurizing from the far end of its now validated hoses. The problem here is that I don't know how to make a pressure nipple plug that will fit into the metal side of those hoses. The 2" test plug does a dandy job of blocking one end, but there isn't enough wiggle room inside the mechanism to drill a hole and put a nipple in there. Maybe a long enough piece of PVC to get past the metal part, and use a hose clamp below the metal part to seal?


From my recollection, you have found and corrected some air leaks. The EGR has some history of causing LHM and not throwing EGR codes.


Gonna clean the EGR if the problem continues, and probably even if it doesn't. With the GDE tune it shouldn't get any dirtier, but as you said it might be causing problems now.


Have you checked the turbo actuator linkage for proper operation, or at least lubed it?

vic

I haven't done a visual inspection or lubed it. I've been watching its reported behavior through my iCarsoft MB II, and it seems to behave as expected. It starts at 95% or so, moves between there and 60% according to driving conditions, and goes to 4.54% and stays there once the ECM disables the turbo. Resetting the system puts it back at 95% until next time. Lubing it sounds like a good idea. Is there info on how to do that here? How would I test the linkage without a scanner than can order the positioner around?

Gonna go visit my Dad in the hospital, which should be a good shakedown run. Thanks for the help! I'm confident will get this soon. A week ago I had no idea what was going no, now I've got a pretty good feeling for most of the turbo system.

thebitmaster
06-16-2016, 12:04 AM
Got intermittent de-activation of turbo several times driving around today. Cycling the ignition makes it go away for a while. Still a low-boost code. Oddly it's only a stored code, and not a stored-and-pending code. It definitely goes longer before this happens, but we're not out of the woods yet. It seems to happen more often at steady speeds than during heavy acceleration.

Next plan is to check and clean the EGR valve and see if I can pressure test the intercooler. I'd like lube the turbo positioner motor but am not sure exactly where to put the lube or which kind to use. If none of that fixes it, I'll get new gaskets for the intercooler hoses and that damaged downpipe on the driver's side.

If that doesn't fix it I might sell the van and start working on fixing the head gasket on the Metro. :)

Scott

Aqua Puttana
06-16-2016, 10:51 AM
Vic! Honored to have you on my thread, sir. Your posts about pressure testing your T1N were a major resource when trying to figure out how to test my NCV3.
That's kind, but you may be disappointed with what I can provide.


I did just change the air filter a few days ago.
I was more thinking a little bit of grit being sucked in through a poor seal.


2) test the intercooler by pressurizing from the far end of its now validated hoses. The problem here is that I don't know how to make a pressure nipple plug that will fit into the metal side of those hoses. The 2" test plug does a dandy job of blocking one end, but there isn't enough wiggle room inside the mechanism to drill a hole and put a nipple in there. Maybe a long enough piece of PVC to get past the metal part, and use a hose clamp below the metal part to seal?
There are various sized test and blocking rubber plugs with bolts and thumb nuts to expand them. One source is the sizes that are used for blocking pipes/closing pools for the winter.

An all thread electrical nipple like used on table lamps and fixtures is NPT National Pipe Thread. They are seamed and not intended for pressure, but for a 15 psi test one could be used to replace the bolt/thumb nut. An added air fitting could then pass air through the expanded plug.

... Lubing it [vane actuator] sounds like a good idea. Is there info on how to do that here? How would I test the linkage without a scanner than can order the positioner around?
The linkage can be popped off so the parts can be moved manually to check for smooth operation. Reports are that it doesn't take much of a "catch" in the linkage to send a Sprinter into LHM.

I just use heavy duty wheel bearing grease applied with an acid brush on my T1N linkage.

Good luck.
vic

thebitmaster
06-17-2016, 05:27 AM
Today I cleaned the EGR. I put some minor scrapes in the mating surfaces despite great care in prying it apart. I had to use a hammer through thin plastic to get it rotate first. Pliers wouldn't budge it, and were marking the exterior. I then carefully pried the edges back and forth on each side, doing my best to avoid the metal gasket. I though about using some PB blaster or similar but wasn't sure if it would hurt something in the valve.

https://c5.staticflickr.com/8/7388/27619560492_a46219d7a4.jpg

The damage wasn't bad, but it's a little worse than the picture shows. I hope I didn't undo the entire point of the exercise by messing up the mating surface too badly.

The gasket had lost its coating in some places from age. I don't know how much that might effect the seal. The GDE tune keeps this valve closed, so if it fails a pressure test later, I might plug the applicable hoses. The tune needs the valve to be there to talk to it, but it the vehicle will never open the valve while operating.

https://c1.staticflickr.com/8/7342/27108180824_bca8d46ea3.jpg

I left the coolant hoses attached. The valve was very dirty. I didn't try the action pre-cleaning to see if it was non-smooth. I cleaned it with brake cleaner and an electric toothbrush. After the cleaning the mechanism moved smoothly and self-closed smartly when let go, though I was surprised by how much force it took to open the valve. With the GDE tune it should be closed at all times. Unless I undo the tune, I can't test the motor by turning the key. If that's worth trying, I'm willing to do it. Vic said that the EGR can trigger a limp without setting a code. I wasn't getting any EGR codes before the tune, just the swirl motor open code.

This is the "after" picture.

https://c4.staticflickr.com/8/7376/27720471315_083cfd81aa.jpg

I also cleaned the sensor visible when the EGR valve is out and the part of the pipe I could reach. It was also covered with soot. I then re-assembled everything and carefully creeped the valve back down by doing a cross pattern on the screws until it was fully seated.

If I wanted to pressure test the EGR circuit, where would I disconnect it? Is the small hose that Ts into the airbox-to-turbo hose the EGR hose? That's the one I was (mistakenly?) calling the crankcase ventilation hose. Since it's before the turbo, why would and EGR leak with no code set affect boost pressure?

As I was taking one of the bolts out it fell down into the labyrinth beneath the EGR. I'm learning how important fishing is a maintenance skill. And here I thought it was something mechanics did to escape their families!

I didn't make that same mistake removing the clip on the turbo actuator. Once I saw how impossible small and springy it was, I knew instantly I would never see it again if I didn't have it tied to something before I popped it off. Behold!

https://c7.staticflickr.com/8/7395/27442505270_2334905a12.jpg

I wrapped a very thin piece of twist-tie wire around the loop before popping it off, and it worked beautifully.

The turbo actuator arm was quite lose and moved up and down smoothly, not even a tiny bit of resistance at any point. I was surprised at how relatively lose it was. I didn't take the hose off to verify the vane movement as it was already dark. The scanner data I'm getting while driving and the turbo response when it's working seem to confirm it's doing its thing. I didn't lubricate it because I couldn't see well enough to see where to put the lube, and the action was already as smooth as a baby's bottom.

After all this, I did another test drive. I cleared the codes, restarted the vehicle, and headed out. I got a loss of turbo and a low boost code after less than a mile during mild acceleration. Oddly, I get turbo loss most often not at high boost levels, but low ones just above intake pressure. I'm watching intake pressure, boost positioner location and boost pressure as I drive on my scanner. I stopped and cleared it, then drove another 6 miles with full boost. I was able to get it up to 22 PSI over intake several times, down and upshifted, sped up and slowed down, etc. I checked for codes when I stopped, and there were no DTCs. However, within a mile of starting on the trip back, I lost the turbo again and had to finish driving back without it.

It's definitely odd that the vehicle can perform perfectly for long periods and then suddenly get the low boost code at low pressure.

One other thing I wanted to mention, none of my tire sensors work and so I get the blinking and solid lamp for those frequently. Is there any chance that that might be killing the turbo, and the low boost codes are just a result of the fact that it's been cut? I'm not clear if the ECM is smart enough *not* to set a low boost code if it was the ECM that turned off the boost for another reason. I don't know if clearing the codes for the engine clears those as well.

I'm feeling pretty frustrated right now. :cry:

I haven't tried to clean the MAF sensor yet. What are normal values of that for various speeds/RPMS/boosts? What's a typical failure mode? I'll see if I can find out where it is. More full-system pressure testing seems to be in order.

Thanks to Vic for more ideas about how to make better test plugs!

thebitmaster
06-17-2016, 10:46 PM
I also want to note that I had good evidence last year that the Y cable on this vehicle was having fusible link problems. Low voltage on the system, excessive voltage drop on start, had the vehicle die once (and only once) in the middle of a busy intersection. How likely is it that could trigger this kind of problem? Random voltage drop maybe causing ECM to kill the turbo? The problem seems worse is bad/wet weather, I'm guessing water and grit are getting into the fusible link.

Also, see a *little* bit of discussion on the list about nominal exhaust back-pressure values for various driving conditions, but nothing really comprehensive yet. I'm wondering if excessive back pressure might be causing the ECM to kill the turbo? No such code is being set. How might I investigate this?

Thanks!

sailquik
06-18-2016, 12:03 AM
After looking at the bunged up out of round turbo fitting in Post 21, that's a highly likely leak path.
Did you replace that part with a nice round smooth part the o'ring can actually seal inside?
Sorry, but o'rings require smooth round surfaces to seal correctly.
You cannot beat something back "sort of into shape" with a hammer and expect an o'ring to provide a
positive seal against ~22 PSIG Boost pressure (36 PSIA MAP pressure @ sea level).
Roger

thebitmaster
06-18-2016, 05:26 AM
After looking at the bunged up out of round turbo fitting in Post 21, that's a highly likely leak path.
Did you replace that part with a nice round smooth part the o'ring can actually seal inside?
Sorry, but o'rings require smooth round surfaces to seal correctly.
You cannot beat something back "sort of into shape" with a hammer and expect an o'ring to provide a
positive seal against ~22 PSIG Boost pressure (36 PSIA MAP pressure @ sea level).
Roger

Hi Roger! Thanks for checking in on me here.

I ordered that downpipe along with replacements and spares for most of the various turbo gaskets and the EGR valve gasket while it's still pretty easy to get it out.

The pipe bracket I was banging on was the turbo-to-resonator one. The original gap between pieces was almost 1/4", which was clearly too much and failed your stall test. Now there is no gap.

Will defer further hunting and pressure testing until the parts arrive, then report back. I'll replace that pipe and the two gaskets on the intercooler hoses first, then try again.

Thanks for all the help!

thebitmaster
06-21-2016, 02:46 AM
Today I replaced the aluminum downtube from the resonator to the intercooler input hose. That lower screw with like 3/4" of clearance on the near side of the suspension was *very* hard to get off, but otherwise things went well. While I was in there, I replaced the turbo resonator output gasket and both of the intercooler hose gaskets. I lubed everything with fresh engine oil.

For the first time, Junior made the run from where to Webberville, a round trip of 12 miles, with no codes or problems! Naturally it's too early to tell if it's a permanent fix, but things are looking good!

Thanks to Roger, Vic, and everyone else who helped me through my first semi-major Sprinter repair.

This will allow me to do a decent amount of driving before posting my review of the GDE tune and to start ordering RVing parts for my trip to Portland. Time to go look at solar panels!

Yehaw!

thebitmaster
06-22-2016, 03:27 AM
I once again spoke too soon. I got about 35 miles on the system with no codes or problems, but after a stop and resumption, the low turbo code popped up again. After that reseting the codes or restarting the vehicle would only restore boost for a mile or so before the turbo cut out again with a low boost code. I crawled under the vehicle to double-check the intercooler hoses, since one of those clips popping off would explain the situation, but they both looked secure. It's very odd that it worked for so long after I replaced the downpipe, then went back to being routinely broken again.

The turbo generally sounds "breathier" than it used to, and the cessation of that sound is very noticeable when the ECM kills it on the low boost code. We've been investigating the input side of the equation, but at this point I'm wondering if there might be a problem with the exhaust side.

After I get done with some Heavy Drinking, I think the next step is to pressure test the entire turbo input path from the air input hose to the throttle body to see if I can find a leak. If I can't, pressure testing the exhaust is probably next. I'm also wondering if I made the join between the turbo and the resonator pipe too tight by modifying the clip. Maybe it works when it's cold, but doesn't have enough expansion room when it heats up?

There's also some damage on the turbo-to-resonator input pipe. It's not nearly as bad as on the resonator-to-intercooler one, it's just a an edge that's been ground down that's nowhere near the join or the gasket. I'm not willing to replace it until I see it leak on a pressure test.

What is the small hose that T-s into the turbo input hose right at the turbo input? Is that the EGR line? How should I handle the EGR for pressure testing? Should I plug it? The GDE tune will keep it closed, but I don't know if it's meant to be 100% pressure tight when closed, or just to block most of the flow? I've been trying to figure out how the EGR hoses connect to the engine, but haven't had much success either on here or in the FSM.

thebitmaster
06-22-2016, 04:45 AM
Also, although this hadn't happened on previous resonator removals, one of the brass screw mounts fell off the resonator when I was putting it back on this time. It was badly distorted like it had been overtightened in the past. Luckily I don't think it goes through to the inside of the resonator. Pressure testing will tell for sure.

https://c4.staticflickr.com/8/7405/27830271075_aa12a8c3d4.jpg

https://c7.staticflickr.com/8/7457/27218121454_f0708466f5.jpg

I put it back in there mainly for alignment. It was too distorted to fully repair, otherwise I would have epoxied it back in the hole. If it turns out to be contributing to the leak, I'll probably epoxy in a similarly sized t-nut.

thebitmaster
06-23-2016, 06:30 PM
After Many Frustrations trying to get a test plug to stay in the turbo inlet hose, I removed the hose and constructed this little gem:

https://c2.staticflickr.com/8/7290/27784846721_3932be7933.jpg

It's made out of a 2" to 2" steel-reinforced hose-clamp plumbing adapter, a 2" SCH40 end cap, about 3" of 2" PVC pipe, and a pressure nipple with some teflon thread tape on it.

Here it is hooked up:

https://c7.staticflickr.com/8/7119/27826504406_c4b0c78f1f.jpg

It slips right over the turbo input. Tighten both hose clamps, and you can apply pressure directly to the input. I managed to get to a steady pressure at 15 PSI after plugging the metal end of the charge air cooler output hose with a 2" test plug, allowing me to pressure test the entire turbo path. I then re-hooked the charge air cooler hose and tested the system, and the air came out of the engine *after* going through the throttle body, still at a decent pressure. Although there is still some danger of blow-out, there is less than other setups I've seen, as the connection is *very* secure and there is hardware in the flight path by virtue
of positioning. Thanks to Vic and others who've posted similar setups for inspiration.

I found that the original leak where the resonator pipe joins the turbo had come back. I was able to "repair" it by loosening the bolt and wiggling it around until the hissing and bubbling stopped. I didn't have to adjust the bracket again, which makes me suspicious. The edge of the metal resonator pipe where it joins the turbo has a small bit of it taken off by some kind of grinding tool.

https://c2.staticflickr.com/8/7346/27249249713_d2c26713db.jpg

It barely cuts into the circular inner diameter, but it does a *bit* and there's definitely a stronger tendency for it to leak on that side. I'm skeptical that it's causing the leak, but if it keeps coming back in that location and further bracket "adjustments" fail to fix the problem, I may order a new one.

There was one other instance where I was wiggling the charge air output hose, which had the test plug in it (pointed away from my face) under pressure, and I got some leakage. I could make it stop by doing certain kinds of wiggling. However, I've tested the hose itself in isolation *twice* at 15 PSI with lots of bending and palpitation with no leakage, once before and once after, and I've also tested it again fully assembled and was unable to reproduce the problem. It most likely had to do with my test setup and not an actual leak.

After this "repair", I made the 12-mile Kessel Run with full turbo, no codes, no limp, and no strange hissing sounds. I'm confident that if I need to pressure test the system again, I have the know-how and tools to do it.

thebitmaster
07-03-2016, 01:32 AM
I'm pretty frustrated at this point. I'm still getting consistent low boost pressure codes despite generally better boost pressure after having:

Replaced the turbo-to-resonator pipe gasket (at least twice, lost one gasket into the system) after a stall test revealed a leak at that junction. I also adjusted the bracket that holds it on with a mallet to make it line up better to overcome what seemed to be a big (> 1/3") gap.

Replaced the turbo-to-resonator pipe itself (it had a ground-down edge that got into the outer diameter)

One of the brass nuts in the resonator came loose. It's not causing a leak, I just put it back into the hole for alignment purposes. I could epoxy it in, but haven't done so yet.

Replaced the resonator-to-charge-air-input pipe, as it had some serious damage on the lower end

Replaced both charge air input gaskets

Pressure tested both charge air hoses independently to 15 PSI with bending and palpitation with no leakage

Pressure tested the system to 15 PSI from the turbo air input to the charge air output hose, which was plugged with a test plug, and also with the charge air output hose in place, thereby pressurizing the engine. I found a second leak in between the turbo output and the resonator pipe, which I fixed with a little more mallet work. I then found a leak at the junction between the charge air output hose and the intercooler at the hose clamp. Neither the hose nor the intercooler was damaged, it was actually the clamped connection that was leaking. It seemed like there was a lack of roundness due to a minor kink at the clamp near the nut. It I moved the hose around under pressure, it would sometimes start hissing there. If I over tightened it, the hissing would stop, or at least decrease enough that I wasn't sure I could hear it anymore. With all the cicadas around right now there's no such thing as silence. If I then (carefully) slowly loosened the connection under pressure, it would start leaking right next to the nut. I took the clamp off and examined it, and it seems like the part where the nut is attached to the clamp wasn't as round anymore. Instead of providing even pressure all the way around, that little bit would become less round for a while as it was tightened, until it got *really* tight. I removed that clamp and replaced it, for testing purposes, with one from a 2" rubber plumbing fixture of a similar size and dimensions. Having done *that*, I no longer detect any leaks.

There is a big difference in the behavior of the hose clamp intercooler connections when loosened under pressure. If I slowly loosen the input hose clamp, at a certain point the hose pops of with a *POP* all at once. However, when I do the same on the output side, the one that had the distorted clamp, but even with the stand-in clamp, it starts to hiss and loose pressure *long* before it pops off.

So having done all that, I can pressurize the system to 15 PSI and find no leaks either through soapy water (which worked on the top leak, but never showed on the bottom one) or obvious hissing noises. With the charge air output hose blocked with a test plug at the metal end, pressure is maintained, the hoses are slightly bulging, my compressor doesn't cycle, and it seems like the seal is leak-free.

When driving, I'm definitely seeing higher boost pressures on average than before. The hissing sound that accompanies the throttle is much less, though it can still be heard when I goose it a bit then immediately let off. The vehicle performs *much* better, and I can get the boost pressure up to around 21 PSI with hard acceleration. I can frequently make a 12 mile country highway trip with no codes, but am still getting low turbo pressure codes, frequently within a mile of starting out. The engine always has the 1/4 mile of dirt road to warm up before I get onto the highway.

The codes almost *always* occur at RPMs below 2000 at 40-50 MPH with very little extra boost happening. They almost *never* occur during highway speeds or hard acceleration. The most reliable way to cause it to happen is to putter down the road just below the point where it shifts up and hold it there for a minute or two. I captured the freeze-frame data for four failures today, and will post those if anyone thinks they would help.

Since the leak isn't showing during pressure testing, I'm assuming it's either temperature or motion dependent. I'm very frustrated, am hundreds of dollars into this repair, and have a trip to the West Coast planned for the next few weeks. I'm trying to decide what to try next.

I don't know of any safe way to check the underside for leaks with the system actually running. I'm not doing a stall test while under the vehicle. Although I've done extensive pressure testing on the intercooler hoses, it's still possible the infamous metal-hose-end problem could be the root cause here. It might be leak-free at room temperature, but pinholing under heat and motion. I've pressure tested those hoses at least twice at 15 PSI by themselves, but I suppose it's still possible there's a leak that either only shows when it's hot, or is blocked by the presence of the test plug.

At the moment I'm on the verge of ordering the RioDanco hose-end adapters. It's almost worth $160 just to never have to pull those freaking clips off again. Does anyone else have additional suggestions? I'm going to do another stall test when my friend gets back to see if maybe the topside leak comes back when the engine is running, but other than replacing the resonator just on principle I'm not sure what else to try.

The most recent problem I was seeing was with the hose clamps on the intercooler output. When I loosened it under pressure it always started leaking right near that bend at the nut long before the rest let go, telling me it's not providing even pressure like the one on the other side. I think that hose clamp may actually be too damaged to use, but I don't see a way to order just that clamp online. Also, I'm not clear on the relative positioning of the parts on that clamp. I may have taken the clamp off to clean the hoses. Is the extra metal re-enforcing strip supposed to go under the rubber loop on the hose, or somewhere else? If it's in the wrong place, it might be making it be more un-round.

Also, is it possible that the hoses themselves have simply gotten too tired to hold the seal at the clamp when host, even if they have no cracks or holes? They certainly balloon out quite a bit at 15 PSI. I'd hate to have to get both the adapters *and* a new set of hoses.

I've had the system apart more than 10 times now, and it will frequently work fine for a day or a trip, only to fail again. I'm at my wits end! :bow:

Aqua Puttana
07-03-2016, 12:37 PM
I appreciate the frustration. The trigger for LHM can be difficult to track down. I also appreciate the detail you have provided.

You have been concentrating on the charge air system integrity. It appears that you have found some leaks and corrected them.

I wouldn't ignore other possibilities.

What DTC's have you seen with your scan tool? Have you recorded all DTC's before clearing/reset? There may be a clue in the history. Once cleared DTC's don't necessarily always return.

You mentioned swirl valves being corrected by the GDE tune. Have you asked GDE whether the swirl valves can still trigger an LHM condition? Have searched the forum to see whether swirl valves have triggered LHM for others?

The charge air system flow is a calculated and closely monitored parameter. There are many sensor inputs which need to be in proper range for the ECM aka ECU to remain happy.

If you have live monitor capability a couple first things to check is the O2 sensor and the EGR operation providing expected in range results.

Sorry I can't offer more.

http://sprinter-source.com/forum/showthread.php?p=161775#post161775

vic

thebitmaster
07-03-2016, 06:36 PM
Hi Vic! Thanks for the help.

It's also worth noting here that the *only* thing that this LHM is doing is killing the turbo. I still have all gears and the normal top speed. It just drives like a pig at low speeds and requires a lot of manual shifting to offset the bow wave of hate building up behind me on single-lane roads.

I appreciate the frustration. The trigger for LHM can be difficult to track down. I also appreciate the detail you have provided.

You have been concentrating on the charge air system integrity. It appears that you have found some leaks and corrected them.

I wouldn't ignore other possibilities.

What DTC's have you seen with your scan tool? Have you recorded all DTC's before clearing/reset? There may be a clue in the history. Once cleared DTC's don't necessarily always return.


From the Sprinter-specific side of my iCarSoft MB2, I'm getting

2359-001, Charge Pressure Control - Too Low Boost Pressure, current and stored.

https://c7.staticflickr.com/8/7466/26825499750_f62f304759.jpg

When I see the Charge Position jump from its normal operating range to 4.53% and I hear the hissing from the turbo stop, and, I stop, pull the freeze frame data, turn off the engine, put the key in the pre-ignition position, clear the code, and verify it's gone before starting again. So it's consistently coming back, and the timing of the freeze-frame is consistent with the low pressure code being the source of the turbo disable. This is the *only* code I'm getting besides the ones for the tire sensors being dead and one for the license plate light being off.

While I'm driving, I'm using the EGR monitor setting on the MB2. It shows me the charge positioner, the ambient pressure, the intake pressure, and the boost pressure. It's this screen with that one extra parameter:

https://c2.staticflickr.com/8/7247/26495755593_808eb25916.jpg

So I can see those values in real time as I'm driving. They seem to change as I would expect, understanding that I have no idea what nominal values are beyond a maximum of about 21 PSI above input pressure on the boost.

When I'm idling at stop the boost pressure is usually within a few percent of the other two, which are *exactly* the same. Sometimes it's a little higher or lower than ambient. Naturally the boost pressure sensor is a suspect here, but if it's malfunctioning, it's doing so in a subtle way. The boost pressure starts out low, climbs when I hit the gas and load the engine, goes up to around 21 PSI over input (2400 hPa with input at about 990 hPa) on hard maneuvers. It almost *never* sets the code at pressures above 4 PSI over input. I can drive 12 miles on the highway at higher pressures and never see the code. I can consistently force it to set the code by puttering along at about 35-45 MPH just below the next shift point at around 1600-1800 RPM. At that point the boost pressure is usually only about 4 PSI over input. From a new reset, it sets the code and kills the turbo within less than a minute.

As I've removed leaks from the pressure path, I've seen the level of boost at low speeds go up noticeably from about 1100 hPa to 1500-1600 hPa. It's not consistent over multiple repairs, but seems to be fairly consistent within a given extended test drive. There is also an accompanying change in the hissy-ness of the turbo, it's less when the pressure is higher.

This slightly grindy hissy noise is one of the main reasons I think there's still a leak. It's been so long since the system has functioned properly that I don't know the system is supposed to sound like. Getting my hands on another similar van in the area just to see how it sounds would really help tell me if the sound is unusual. My memory tells me it is. It changes in proportion to throttle, and is most noticeable when I gun the engine and then immediately let off. The "normal" engine noise dies down a little faster, with the hissing sliding off and dying just a little bit afterward. At idle I don't hear it at all.

So the data I'm getting from my scanner, plus the hissing sound, seems consistent with a small leak that mainly manifests at lower pressure.



You mentioned swirl valves being corrected by the GDE tune. Have you asked GDE whether the swirl valves can still trigger an LHM condition? Have searched the forum to see whether swirl valves have triggered LHM for others?


One major point of the GDE tune is to prevent codes from being set and LHM from the swirl valves misbehaving. That code went away as soon as I installed the tune. If I specifically go an check the positioner for the one that's not correct, I can see that it's different from the one that is. But the computer has been told to keep those valves open and ignore misposition warnings. My original code said that one was stuck *open*, but I never verified that by visual inspection because it looked too complex. The GDE tune also permanently closes the EGR. I didn't mention in the above list, but earlier in the thread I took apart and cleaned the EGR. It was dirty, but the valve motion seemed pretty clean after I was done. I did put some nicks in the polished mating surface where the seal goes, and I also re-used the worn-looking seal, so it's possible I introduced a small leak there. I've since ordered a fresh gasket, but haven't installed it. My inclination long-term is to get block plates and cut the EGR entirely to remove that variable. The part still needs to be there for the ECM to talk to, but it no longer plays a role in the physical operation of the vehicle.

If this persists, one thing I intend to try it to reverse the tune and see if I still get the low boost pressure code. The swirl code will eventually put me in LHM, but if it takes longer than the boost code to show up, I could verify if the tune is introducing a low boost code when the factory firmware is not. If that's true, I will contact GDE with that data and see what they say. If the tune is involved, my guess would be that something else, like, say the MAF or the boost sensor is *slightly* off its normal range, and when that is combined with the changes in engine behavior and shift points from the tune, it's creating a situation the ECM doesn't like that usually doesn't happen with the tune. I still owe GDE a review in the next month or so to get my discount, and they did ask for an honest one. :)


The charge air system flow is a calculated and closely monitored parameter. There are many sensor inputs which need to be in proper range for the ECM aka ECU to remain happy.

If you have live monitor capability a couple first things to check is the O2 sensor and the EGR operation providing expected in range results.

Sorry I can't offer more.

http://sprinter-source.com/forum/showthread.php?p=161775#post161775

vic

Thank you for the link. I've ready many of those threads, but maybe not all of them.

I can monitor all sensors in real time with the MB2, though the way they're grouped is hardwired, at least under the Sprinter-specific things. Under the ODBII branch, I can pick and choose among a much smaller subset of sensors. And I can't record data in real time unless I point a camera at the screen, which is probably my biggest functional disappointment with the iCarsoft MB2. One thing I *really* want is example data for all the sensors, including the O2, MAF, boost pressure, exhaust back pressure, etc. Is there a good source for those numbers and ranges? I have the Windows FSM setup with the little RAM Crystler Jeep logo in the upper left corner, but don't see values for those. Maybe I'm looking in the wrong place? Where do I find typical ranges?

Assuming it's a sensor problem, I was going to check/clean/replace-if-out-of-spec the MAF and the boost pressure ones. I'll add O2 to that list. I looked at the MAF, because I've heard it's a suspect in these situations. The plastic housing looked clean, but I couldn't see the wires inside and don't know that normal values are. I'm guessing if it were way off, though, the ECM would be setting other codes that depend on it, not just the low-turbo code.

Another path I haven't taken yet is to pressure test the exhaust. A leak there would explain both the hissing sound and why addressing the boost side isn't completely fixing the problem. Anyone know how to make an exhaust pressure adapter? That goofy anime-shaped hole in tailpipe looks hard to plug.

I broke down and ordered the charge air hose adapters. I'm telling myself it's a "long term investment that probably would have to be done eventually anyway", but my dwindling bank account is getting pretty tired of that rationalization. :cry:

thebitmaster
07-03-2016, 06:51 PM
Also going to try cleaning the MAF, since it's cheap and easy and might help.

72chevy4x4
07-04-2016, 04:58 AM
think I asked before, but what gives with the "at least twice, lost one gasket into the system". Are you sure there are no other objects in the IC piping?

Pressurizing the system wouldn't determine if their is blockage as it would be a flow issue vs a pressure issue (unless near 100% blockage).

I don't know for a fact, but would anticipate the ecu would not be too happy about a restricted airflow situation and shut down with a LHM error. If this were an issue, would seem the pressure observed on a monitor would rise more quickly as compared to a typical system.

Have you tried rolling into the boost where you try not to build up too quickly and not maximum boost pressure? maybe my starting on a longer downhill grade would test this theory.

keep trying!

thebitmaster
07-04-2016, 08:44 AM
think I asked before, but what gives with the "at least twice, lost one gasket into the system". Are you sure there are no other objects in the IC piping?


I'm fairly certain one of the turbo-to-resonator pipe o-rings got sucked into the system. When I first took that junction apart, the original gasket slipped off the pipe before I saw it. So when I looked to see where it was, it was inside the resonator pipe right where the flare is instead of on the turbo end over the pipe where it was supposed to be. Because I'd never had the system apart before, I thought that the place the old gasket occupied was the correct location. So when I put it back together, that's where I put the new one. There's not really much preventing it from getting blown into the intercooler from there, and I think that's what happened. There is a *tiny* possibility that it fell out on top of the engine when I took it apart next time, but I was being *very* careful because I'd detected a leak with a stall test there (surprise surprise!) I was specifically watching for the o-ring. When the pipe came apart, it was gone. I haven't been able to recover it yet. I know it's not in the intercooler hoses, so it's probably in the intercooler. It might be partially blocking pressure in weird ways there. I thought about pressurizing the system, then popping the input hose off suddenly to see if I could blow it out, but I haven't tried that. Isn't it possible to clean out the intercooler, maybe with water? If so, that might also be worth trying.


Pressurizing the system wouldn't determine if their is blockage as it would be a flow issue vs a pressure issue (unless near 100% blockage).

I don't know for a fact, but would anticipate the ecu would not be too happy about a restricted airflow situation and shut down with a LHM error. If this were an issue, would seem the pressure observed on a monitor would rise more quickly as compared to a typical system.

Have you tried rolling into the boost where you try not to build up too quickly and not maximum boost pressure? maybe my starting on a longer downhill grade would test this theory.

keep trying!

I'm about to post a Very Long Post describing the seven test runs I did with data, but the short version is that the error almost *never* occurs at high speeds and boost loads. It almost *always* occurs when I slowly accelerate up to about 45 MPH and then keep it there, just below the shift point at about 1800 RPM with the boost at 1200 hPa with intake at 990 hPa. I can reliably reproduce the code within 1/2-1 mile of doing this. However, if I drive it like I stole it and keep the boost up in the 1500-2200 hPa range, I can drive six miles or more with no codes. So it *seems* like I've already done the test you're describing, unless I'm misunderstanding. It is almost 4 AM.

This situation might jive with the lost-gasket-in-the-intercooler theory. At lower pressures the gasket might block more and create back pressure, reducing boost. At higher pressures it might bend and not block airflow as much. It might also explain the buzz-hiss noise if part of it was flapping, although I think that noise pre-dated losing it. On the other hand, there were leaks then, which would have produced a similar sound.

What about bypassing the intercooler with a piece of PVC just for short-term testing purposes? If that fixes the problem, I know I have go fishing.

thebitmaster
07-04-2016, 09:14 AM
I'm going to put all my data into a spreadsheet tomorrow, but here's the Executive Summary.

I cleaned the MAF with MAF cleaner as best I could. It's a Bosch, but there are no Torx security screws to remove the naughty bits.

48654

I tried to squirt it into all the holes I could find into the middle part. I also used the cleaner on the connectors for the MAF and the other sensor next to it. I let everything dry.

I then did 3 runs with the MAF re-installed. I was able to reproduce the low boost as mentioned above at 45 MPH and 1200ish hPa reliably. I captured the freeze frame data, which includes the MAF values, for each run. At some point that wasn't the beginning, I also started photographing the Charge Positioner %, the ambient and intake pressures, and the boost pressure right at the moment the turbo cut out. After each time, I reset all codes, cleared them, and verified no DTCs.

I then disconnected the MAF plug after having heard a dirty rumor that the system, at least on T1Ns (which I don't have) that the ECM would provide "reasonable values" with it unplugged, and that sometimes MAF problems could be diagnosed by disabling it. I repeated the tests. The vehicle drove weirdly, sometimes waited until 3500 RPM to shift up, but I was able to drive at *least* 3-5 times as far with the same method of slowly accelerating to 45 MPH then holding it until the ECM killed the turbo. The boost pressures were about the same. And it was a different code related to the MAF being disconnected, *not* a low boost code.

2011-001 Charge air system - Sensitivity drift - The air mass is too large.

So disconnecting the MAF sensor prevents the boost code from happening under the same conditions. I repeated the test three times and captured all the freeze frames for that too.

Finally, after another DTC reset and re-connection of the MAF, I drive it like I stole it back from Webberville 6 miles with no codes, which is what I expected. The boost pressure was 1500-2200 hPa with intake at 990 hPa. I then managed to reproduce the low boost code at 35 MPH on my trip home within less than a mile.

It seems like further testing of the MAF to under conditions where bad values would obvious is in order. It's in the freeze frame data, so maybe it will jump out at someone once they see it all together in one place. I wish I had another to swap out and try. Obviously I need more NCV3 friends here in Austin. But this doesn't *necessarily* mean that the MAF is the problem. The vehicle drivers very weirdly with it disconnected, and the internal state of the system and its sensor might be so different from when it's still plugged in that the witches brew of circumstance and readings that's causing the code might not be happening.

What else can I do to check if the MAF values are where they are supposed to be?

I may also repeat this test after undoing the GDE tune tomorrow night. I know the Swirl Valve code will eventually put me into LHM, but if it takes a lot longer than the turbo low boost code to set, I'll have some idea if the tune is contributing to the problem.

Aqua Puttana
07-04-2016, 12:54 PM
First let me say that I'm just some guy posting on the internet. My 2004 T1N has really been very reliable so my actual hands on experience with engine/drive train operating problems is limited. Most of my repairs have been mechanical in nature related to wear parts or rusting problems.

For many of my posts I am just a parrot of the forum. Forum advice automatically sets the scenario that I have not even seen the vehicle under discussion. Other than a few exceptions where the symptoms are clear, I'm not smart enough and I don't have enough experience to troubleshoot from a distance. Most of my posts are intended to provide information, not a diagnosis.

So many times Sprinter problems do trace to basic electrical or mechanical issues, not complex computer control system problems as many seem to expect.

... I can drive 12 miles on the highway at higher pressures and never see the code. I can consistently force it to set the code by puttering along at about 35-45 MPH just below the next shift point at around 1600-1800 RPM. At that point the boost pressure is usually only about 4 PSI over input. From a new reset, it sets the code and kills the turbo within less than a minute.
Have you checked for smooth operation of the turbo vane controller? There have been threads where LHM problems have been tracked to a slight catch in the mechanical linkages. Corrosion of the OM647 linkages is a known problem. I lube my linkages occasionally. The linkage can be popped loose to check for smooth operation before lubing.

... Maybe I'm looking in the wrong place? Where do I find typical ranges?

...
As I said above, I can only provide limited input. Were I in your situation I would need to search for that specific information.

I do have some information for the T1N. I assume that similar information for the NCV3 should be out there somewhere.

77736

77737

77738

77739

77740

Sorry I cant provide more.

vic

Aqua Puttana
07-04-2016, 12:55 PM
cont.

77741

77742

done

thebitmaster
07-04-2016, 04:37 PM
First let me say that I'm just some guy posting on the internet.

So many times Sprinter problems do trace to basic electrical or mechanical issues, not complex computer control system problems as many seem to expect.


Noted.


Have you checked for smooth operation of the turbo vane controller? There have been threads where LHM problems have been tracked to a slight catch in the mechanical linkages. Corrosion of the OM647 linkages is a known problem. I lube my linkages occasionally. The linkage can be popped loose to check for smooth operation before lubing.


I popped off the clip and checked the motion after you first suggested it. It was super loose, it felt like it wasn't attached to anything until it hit the limits of its motion. I was surprised at how much front-to-back play there was in it. It was jiggling a multi-piece fishing lure. I didn't have the acid brush and lube at the time, but I do now. It wasn't entirely clear how to see what I was doing or get in there, as on the NCV3 it's hard to see the linkage directly. I had to use a dental mirror to figure out how to get the clip off.

I'm forgetting the formatting for a specific post in an thread at the moment, but I describe that in #26 of this thread.

Lubing it sounds like a good idea. It wouldn't take long and would remove another variable.


As I said above, I can only provide limited input. Were I in your situation I would need to search for that specific information.


Thanks for posting that!

thebitmaster
07-04-2016, 06:19 PM
So I just ran across this service bulletin:

48642

...and while the idea that I'm going to replace the turbo-to-airbox seal EVERY TIME I TAKE IT OFF, which I've naturally had to do at least 15 times now in diagnosing this problem, is completely insane, it *does* appear that mine is installed backwards! It says that the larger lip edge is supposed to face toward the front of the vehicle. I replaced a damaged one earlier in the process, and since the old one had the lip at the *back* of the vehicle (incorrect), that's how I put on the new one. That would not only explain why there's dust wear on the turbo vanes, but it might also be causing the low boost code! I will immediately flip it around, make sure the clamp is tightened to spec, and do some test drives.

thebitmaster
07-04-2016, 07:06 PM
Nope, I'm wrong. It seems to be installed correctly, with the lip at the front and on the inner side of the gasket, which is how the old one was, *and* how the new one came to me from the factory.

https://c2.staticflickr.com/8/7300/28009325361_e004a5c399.jpg

Time to go get my torque screwdriver and check that clamp tension! I think I will reverse the GDE tune and see if that changes my behavior for the trip.

Scott

So I just ran across this service bulletin:

48642

...and while the idea that I'm going to replace the turbo-to-airbox seal EVERY TIME I TAKE IT OFF, which I've naturally had to do at least 15 times now in diagnosing this problem, is completely insane, it *does* appear that mine is installed backwards! It says that the larger lip edge is supposed to face toward the front of the vehicle. I replaced a damaged one earlier in the process, and since the old one had the lip at the *back* of the vehicle (incorrect), that's how I put on the new one. That would not only explain why there's dust wear on the turbo vanes, but it might also be causing the low boost code! I will immediately flip it around, make sure the clamp is tightened to spec, and do some test drives.

thebitmaster
07-05-2016, 03:47 AM
So it *seems* like I've tested the intercooler output hose both by itself and "in system." This is evidently the most common cause of this problem. However, it's possible that the failure mode where the rubber peels off the metal might actually get blocked by a test plug when it's tested in isolation. If the leak is in the zone sealed off or is re-enforced by the test plug, it would pass individual pressure testing. And the only way to test it "in system" is to have it plugged into the engine, at which point some air is flowing through the system. A small leak might not get detected there. Since I won't get under there to do a stall test, I can't be sure it's not leaking at the most common fail point.

I've already ordered the aluminum end adapters, so the point is *probably* moot. However, if I *switched* the input and output hoses, I could block at the end of the input hose (which would then be on the output side) with the test plug, and then pressurize the system. If the output hose (now on the input side and hooked up at both ends with no test plug) had a leak, it would show up under those conditions.

thebitmaster
07-05-2016, 09:36 PM
The Good News is that I don't have to clean the Charge Air Cooler or go fishing for that resonator pipe gasket I lost into the system.

The Bad News is that's because the Charge Air Cooler has a leak. Removing it in the Texas heat with no shade but my drover hat was an unfun experience.

I'd pressure tested the whole system, and never heard any hissing from there. I blame the cidadas. I didn't do much soapy water testing on the cooler, because I couldn't reach the whole thing. However, I should have spayed what I could get to:

https://c6.staticflickr.com/8/7457/27497880653_45f26e898f.jpg

That big greasy, grimy spot in the corner is a classic tell-tale of a turbo pressure leak.

Most of the process of getting it off seemed to be covered by the FSM, but I could not for the life of me find the part that talked about the huge plastic fender that mates with the radiator, etc. I finally had to wing it.

https://c6.staticflickr.com/8/7342/28112709845_a8fc27cb24.jpg

My inherent laziness realized that if there was actually a leak, cleaning and playing Fishing for Gaskets would be a waste of time, so I decided to pressure test first, not realizing it would literally be the path of least resistance. I used the same plug and adapter I use on the entire system:

https://c4.staticflickr.com/8/7636/28035301051_c021fd0f81.jpg

Once I was 6 hours into the project, that greasy spot finally caught my attention and I sprayed it with soapy water.

https://c6.staticflickr.com/8/7336/28112699925_919db6a8a5.jpg

...and got bubbles at the grease stain.

Just ordered a new one. I can't stop myself from being excited I found another stubborn leak, but I'm no longer making the reckless assumption this means I'll never see another low boost code again.

Wish me luck!

thebitmaster
07-05-2016, 10:03 PM
Since I've found the likely source of the problem with the charge air cooler leak, I'm not going to take the time to post those 7 test runs with their freeze-frame data I mentioned earlier. However, I will mention that, after trying those test runs with the cleaned MAF sensor and the disconnected MAF sensor, I did do another three runs after de-activating the GDE tune. After restoring to stock, I got the low-boost code much faster, almost immediately. I also got the swirl valve positioner code on one of the three runs, but not the other two. I got the low boost code every time. So for sure the GDE tune wasn't responsible, and it actually allowed me to drive with boost a lot longer than the stock firmware. I'm guessing that's because of the changes in the engine parameters that enable the higher performance with the tune. The stock firmware probably sets up the conditions for the leak sooner due to different compressions, RPMs, shift points, etc.

Aqua Puttana
07-06-2016, 09:51 AM
Great news! Not that you need to buy a new cooler, but that you located the problem.

You are not the first person that it took some time to track down an inter-cooler leak. Fortunately the coolers are generally reliable. The leak must be fairly small and stable to not have triggered a DTC and LHM while under hard acceleration. The typical scenario is for the higher system acceleration pressure to open the leak more which triggers LHM.

A 15 psi pressure smoke test may have revealed the leak, but I have no idea how that could be done DIY.

The Good News is that I don't have to clean the Charge Air Cooler or go fishing for that resonator pipe gasket I lost into the system.

The Bad News is that's because the Charge Air Cooler has a leak. Removing it in the Texas heat with no shade but my drover hat was an unfun experience.

...
Good one. :thumbup:

:cheers: vic

thebitmaster
07-06-2016, 04:31 PM
Great news! Not that you need to buy a new cooler, but that you located the problem.

You are not the first person that it took some time to track down an inter-cooler leak. Fortunately the coolers are generally reliable. The leak must be fairly small and stable to not have triggered a DTC and LHM while under hard acceleration. The typical scenario is for the higher system acceleration pressure to open the leak more which triggers LHM.

A 15 psi pressure smoke test may have revealed the leak, but I have no idea how that could be done DIY.


Good one. :thumbup:

:cheers: vic

Thanks, Vic! :cheers:

I've been thinking about how to make a high-pressure smoke test work for a while. If the smoke could be put into the de-pressurized system *first*, with the higher pressure added *after*, that might work. I'm assuming the smoke chemical they use is safe for the turbo pressure path. Using something that fluoresces and using a blacklight to search would also work. It probably actually wouldn't be that hard to make a high-pressure smoke adapter for a standard compressor setup. It might require having a valve on the far side of the system that could be opened instead of a test plug to allow the smoke to fill the system before full-pressure testing occurred. Or you could just fill it up on low pressure with the test plug out, then put it in and increase pressure.

thebitmaster
07-06-2016, 04:56 PM
I also wanted to mention this bulletin again from Daimler:

http://sprinter-source.com/forum/showthread.php?t=48642

which basically says you're supposed to replace the (on my van, orange) input-hose-to-turbo-gasket *EVERY SINGLE TIME IT'S REMOVED FOR ANY REASON*, and the clamp should be tightened to 3 nM. I used a torque screwdriver commonly used for gun maintenance.

Not doing so could void the warranty if you still have one. I mention this because I had to take that gasket off every single time I wanted to switch from using the van to pressure testing the system, which in the course of this repair was at least 10 times. Sure enough, due to wear and over tightening, my first one ripped. I repaired it with RTV silicone, but I would only use it again in an emergency.

Following Daimer's instructions to the letter is the best way to go, and that might mean admitting you're not going to use your van until you find the leak so you only have to replace it once. A more balanced approach would be to re-use the existing gasket until you're done taking it on and off, then replacing it with a brand new one when you're finished. A final possibility would be to make an adapter that fits on the input end of the hose near the airbox. I completely failed to make this work because getting an airtight seal on the other end proved too difficult. My adapters kept leaking or popping out. However, it occurs to me that if one took out the MAF sensor on the airbox side so it was still attached to the hose, then used a similar rubber double-hose-clamp setup (maybe even the same exact hardware I used) on the far end, a good seal could be achieved more easily. There would be a risk of the MAF sensor getting damaged if there was a high-pressure pop-off, but it would mean being able to test the system without taking it apart nearly as much.

So if you use my system for pressure testing, be aware of that bulletin. My turbo is showing signs of dust-wear, and that might be the reason. At 200k+ I'm long out of warranty, but if you're not good about putting on an angel face in front of your favorite service manager, playing by the rules is the safer bet.

Be careful out there. :smilewink:

thebitmaster
07-06-2016, 05:09 PM
It does seem like this particular kind of leak is the hardest to find. I have no doubt that a qualified dealer mechanic would have found it sooner, especially because of the visual inspection aspect and access to more sophisticated tools (not to mention air conditioned, scorpion-free workspace!), but for the most part I think I did things in the right order. The result, though, is that I will have spent over $600 on the system and probably 30-40 hours of work by the time I'm done. Believe it or not, I got a Sprinter to *reduce* my cost of living! *Most* of those things were wear parts that have to be replaced anyway, and the ones that weren't, the aluminum pipes, were badly damaged, but that's a lot more money than I thought I'd spend when it looked like just a minor hose leak. It's also the way that used buyers get borked. It had the low boost code when I bought it used a little over a year ago, but it would only limp very occasionally, and the "qualified Sprinter mechanic" told me that the fixes for that *and* the infamous Swirl Valve code would be only $350 in parts. With the GDE tune to "fix" the swirl problem, I'm well over $1000 on the vehicle, and haven't even replaced the tires or their troublesome sensors yet, much less started the RV conversion. And while I don't think the intercooler hoses are actually leaking, I did order the aluminum end adapters because they seemed like a good long-term investment. Hopefully this means my turbo will be good for another 200k!

Froggy
07-06-2016, 07:18 PM
I really appreciate your optimism and I hope it goes well for you! I have owned numerous cars, trucks and vans over the past 50 years (more than 600,000 miles driven in them). But my 2007 Sprinter easily wins the distinction of being the most expensive to maintain and temperamental in terms of running. I dread seeing the next but inevitable CEL. Wife says I need vanity tag that says, "Whatswrongnow?" LOL

thebitmaster
07-06-2016, 11:11 PM
I really appreciate your optimism and I hope it goes well for you! I have owned numerous cars, trucks and vans over the past 50 years (more than 600,000 miles driven in them). But my 2007 Sprinter easily wins the distinction of being the most expensive to maintain and temperamental in terms of running. I dread seeing the next but inevitable CEL. Wife says I need vanity tag that says, "Whatswrongnow?" LOL

Oh no! :cry:

At least I can do a substantial amount of the work myself and save the Big Money for things that really require it.

Jfragola76
07-07-2016, 04:52 AM
Very interesting post. I have similar issues on my 2007, right down to the oil soot on the intercooler. Pretty sure that's one of the sources of my low boost code but I haven't had the time to really get into it. Please continue to post your findings with the new intercooler. Good luck !

thebitmaster
07-07-2016, 05:27 AM
Very interesting post. I have similar issues on my 2007, right down to the oil soot on the intercooler. Pretty sure that's one of the sources of my low boost code but I haven't had the time to really get into it. Please continue to post your findings with the new intercooler. Good luck !

Thanks! Glad to hear this super-long thread is helpful to someone. Intercooler should be here this weekend. The easiest thing to check yourself is taking off the intercooler hoses and shining a flashlight from the outside while looking at the inside in a dark room. Getting the clips back on is a pain, but otherwise it's pretty easy to do. Make sure not to over tighten the hose clamps on the intercooler side of the hoses. I'm told the driver's side one is the one most prone to leaks.

If you have access to a shop compressor, making the adapter and test plug I used for pressure testing is easy and cheap. Refer to earlier in the thread for how. I may do a full writeup of the technique later. If you have the greasy spots on the intercooler, it might be worth spraying them with soapy water at idle to see if there's any bubbling even though there's not much pressure in the system at idle.

I know I should know this by now, but does gunning the engine in park put any pressure in the turbo path, or only doing a stall test where it's in gear? It is, in my opinion, too dangerous to be under the vehicle during a stall test, which is the easiest way to pressure test the system. One other idea I had, though, was to use a camera phone to focus on the problem area from an angle where it won't get dripped on to take a video, or maybe use Facetime or Skype to see a live video feed using another camera or computer, then spray that greasy spot (take care not to get your phone wet) with soapy water, move yourself out of the way, and do the stall test (which is putting the vehicle in Drive, holding down the brake pedal firmly, then smoothly flooring the gas all the way down and back up to pressurize the turbo system. It sounds horrible but I've been told it's ok as long as you don't do it long and wait a few minutes for things to cool down in between.) The video will never be as good as looking at it with your eyes, but the bubbles I got from my intercooler leak were more than obvious to see even at low resolution. I also found an earlier leak at the junction between the turbo and the resonator pipe with a stall test and soapy water. That was safe to do because I could see the top of the turbo path without standing in front of the vehicle's path of (potential) motion.

Is there a mechanic's GoPro yet?

moondawg14
07-07-2016, 12:54 PM
Is there a mechanic's GoPro yet?

Mostly!

https://www.amazon.com/Accessory-Magnetic-Safety-compatible-cameras/dp/B00GFEXMY6

thebitmaster
07-09-2016, 07:30 PM
You can imagine how exited I was when, at 2 AM, after installing the new intercooler, I got a low boost code within a mile using my standard 45 mph test. The new one arrived with a dent, which make me very, very angry, but I wasn't willing to wait to deal with the shipping company or source to put it in. I pressure tested it, paying particular attention to the dented area, and found no leaks.

https://c4.staticflickr.com/9/8573/28198629435_08ac7bb795.jpg

I then installed the aluminum hose end adapters and put everything together. The clamps on the hose adapters are horrible, and basically impossible to tighten when actually on the vehicle because they use nylock nuts on a thread instead of the normal hose clamp configuration. Tightening them all the way requires an extra-deep 11 mm socket, which I couldn't find. The best policy when installing them is to put them on very lightly so they can still rotate, line the lightly assembled hose up with the ends and rotate the new metal end so that the keys lines up with the socket. Once you've got that rotation right, take the hose back off and tighten the clamp off the vehicle. The *first* low boost code I got after putting things back together was because of a leak in the new passenger-side adapter where it had slipped out of alignment and not been fully tightened because of how hard that is to do when it's actually plugged. Given more time and effort, I might replace their clamps with the same one Daimler uses on the other end.

Before:

https://c2.staticflickr.com/9/8653/28198627545_f87e289a17.jpg

After:

https://c2.staticflickr.com/9/8745/28198633265_24854df2f5.jpg

So much for dreams of being able to use a "simple hose clamp" to remove these hoses instead of the Clip of Doom and Frustration.

Does anybody know the proper torque for the hose clamps on the intercooler lines? After all this uncertainly, I find I don't trust myself to gauge the torque by feel. I couldn't find it in the FSM.

After resolving the leak in the new passenger-side and re-assembling and pressure-testing both hoses, I proceeded with a system pressure test with the intercooler output hose plugged into the engine, and found no leaks. At this point I have replaced *everything* in the entire turbo pressure path except the resonator, the resonator gasket, and the rubber hoses, which have been modded with the aluminum end adapters and pressure tested repeatedly.

So after all that, I still get the code, but it takes longer. I can drive maybe twice the distance or more at 45 MPH before I lose the turbo. At this point I'm beginning to suspect a Major Deity might be angry with me.

After I got my friend helped me do another stall test and I sprayed the upper turbo path with soapy water. No bubbles. Remember I pressure tested the system (though without plugging the end) also.

Overall, pressures seem to be a bit higher and more stable than before. As before, the code doesn't get set when I drive the vehicle as highways speeds or higher boost pressures. The test I'm using is somewhat artificial, in that most people don't cruise straight and level at 45 MPH for multiple miles very often. That speed is usually a transition between higher and lower speeds.

Here is the freeze frame data from the last two runs. I reset DTCs in between. Does anything look wildly incorrect? The thing I notice the most is that the system voltage is rarely much over 13 V. I already have good reason to suspect I have a bad fusable link on the Y cable, and I'll be verifying that today. I know that can cause otherwise bogus codes. Having found like four different leaks in my turbo pressure path, I can't believe it's magical coincidence, but it's still Another Thing to Check. I'll be charging RV house batteries from that alternator soon, which is another reason it needs to be fixed.

The values for the Ambient and Boost Pressure aren't from the freeze-frame but from the live data feed. Do those MAF values look ok? I don't get the low boost code when I disconnected the MAF sensor, but the vehicle drives so differently and still eventually throws a different boost code because of No MAF that I can't draw any conclusions.

Keep in mind I have the GDE tune so the EGR should always be closed.

Run 1
Mercedes DTC 2359-001 Charge Pressure Control - Too Low Boost Pressure
Generic DTC P0299 Turbocharger/Supercharger Underboost
Calculated LOAD Value 67.10%
Engine Coolant Temperature 186 o F
Intake Maniforld Absolute Pressure 42.3 inHg
Engine RPM 1809
Vehicle Speed 44 mph
Intake Air Temperature 102 o F
Intake Pressure 995h hPa, 14.43 PSI (estimated)
Boost Pressure 1258 hPa, 18.24 PSI (estimated)
Air Flow Rate from Mass Air Flow Sensor 7.9 lb/min
Absolute Throttle Position 89.80%
Time Since Engine Start 164 sec
Fuel Rail Pressure 7676.3 psi
Commanded EGR 4.30%
EGR Error 0.00%
Fuel Level Input 32.90%
Barometric Pressure 29.4 inHg
Control module voltage 13.19V
Relative Throttle Position 2.00%
Ambient Air Temperature 91 o F
Accelerator Pedal Position D 29.80%
Accelerator Pedal Position E 29.80%
Commanded Throttle Actuator Control 96.90%

Run 2
Mercededs DTC 2359-001
Generic DTC P0299
Calculated LOAD Value 52.50%
Engine Coolant Temperature 194 o F
Intake Maniforld Absolute Pressure 41.4 inHg
Engine RPM 1714
Vehicle Speed 37 mph
Intake Air Temperature 100 o F
Intake Pressure 995h hPa, 14.43 PSI (estimated)
Boost Pressure 1258 hPa, 18.24 PSI (estimated)
Air Flow Rate from Mass Air Flow Sensor 7.7 lb/min
Absolute Throttle Position 91.00%
Time Since Engine Start 246 sec
Fuel Rail Pressure 6807.8 psi
Commanded EGR 4.30%
EGR Error 0.00%
Fuel Level Input 32.90%
Barometric Pressure 29.4 inHg
Control module voltage 13.39V
Relative Throttle Position 0.40%
Ambient Air Temperature 91 o F
Accelerator Pedal Position D 27.80%
Accelerator Pedal Position E 27.80%
Commanded Throttle Actuator Control 98.40%

At this point I'm about ready to just drive the thing off a cliff with me in it. Instead I'll probably look into pressure testing the exhaust and verify and/or fix the fusable link. I may try the just-add-solder technique, at least temporarily. I may also try to bypass the link with another cable for testing purposes.

I have a 100 minute round trip today to pick up solar gear for my RV conversion, so we'll have a better real-world test case.

thebitmaster
07-10-2016, 06:38 AM
I may have found the last problem.

At the beginning of the thread, I modified the turbo resonator pipe bracket to close the 1/4" gap between it and the turbo. In the process, it pulled the other end of the pipe further out of the resonator. Closing the gap at the turbo end meant that this end was angled, with quite a bit of gasket showing. It never leaked on pressure tests that I could see or hear, but it's exactly the sort of situation where the addition of heat and the unpredictable pulsing of turbo pressure could maybe open it.

I pressure tested the resonator today, and in the process wanted to pressure test that join. Naturally it was nearly impossible to hold the resonator and the pipe together with my hands even at under 5 PSI. But the thing that I discovered is that that join can hold pressure with an enormous amount of the gasket pushed out of the aluminum socket in the pipe. However, how much it can stick out is hugely influence by how straight it is. I used this same setup to test that combination, and the resonator in isolation. There were no leaks:

https://c6.staticflickr.com/9/8745/28131610981_30206c1df9.jpg

My hammer job on the bracket may have brought those pipes snug, but at the cost of the resonator gasket being straight and pushed in further. Suspecting a rat, I reversed the process, ensuring that as much of the resonator gasket was inside the pipe as possible, and that it was straight instead of elbowed. That meant that 1/4" bit of shiny metal between the other end of the resonator pipe and the turbo came back. I suspect, though, that some version of this is how it's supposed to be. The resonator end gets priority for tightness, and the turbo end is ok to have some gap. The gap once again looks pretty much like this:

https://c7.staticflickr.com/8/7319/26847212190_aba3e45144.jpg

This makes me uncomfortable, but I have nothing to compare it to at the moment. How does that gap compare on anyone else's NCV3?

However, the other end, which I don't have a picture of, was much more straight and only sticking out about a 1/4" itself, which is much tighter than before.

Naturally, as soon as I did a first test at resetting codes, I got a code within a mile. This was done to Destroy my Soul and prevent whatever shining future awaits me in Portland in a few weeks. Mostly undaunted, I grabbed the freeze frame, reset the codes again, and got back on the road. After that initial code, I was able to drive 13 miles at 43 mph, right at the conditions that used to set the code, and *no code*. I did one brief unnecessary display of acceleration just to watch the boost pressure over a big range, then immediately came back to 43 mph.

I've *definitely* never driven that far at that speed since this problem happened. The hissing noise I've been complaining about was almost gone, and that's the first time I can honestly say I'm sure I heard a substantial difference. Previously I could reliably reproduce the low boost code in less than mile, sometimes a few hundred feet.

Obviously it's too early to be sure, and I *did* get that initial code leaving the property. What I'm guessing, though, is that that leak was probably at the turbo-to-resonator-pipe junction, and went away once the system was sufficiently warmed up. It's possible that I can adjust the bracket again to close that join *just* a little more and only cause a little more gap on the resonator gasket end. As long as it's straight, that join can take it. That's what I'll try if I get another code. I'll also do another stall test on both joins tomorrow.

I think we're on the home stretch, which is fabulous, because I just picked up my $1200 worth of solar gear today for $700.

Also, after seeing the OBD2 voltage almost always in the low 13s VDC, did the voltage drop test from the battery positive terminal to the alternator positive terminal with the AC and headlights on, and it was over 1 VDC. I checked out the fused link with a flashlight, and it looks pretty fried. I'm gonna replace the cable, but I need to figure out which alternator I have first. Any reason to get the Mercedes cable when the Dodge one is cheaper? I'm kind of tempted to build it myself. I've got a friend who works on electric cars that would have all the gear.

Jimdawdy
07-10-2016, 06:54 PM
I had almost a 1volt drop from the alt. to the starter and battery. I used a Dodge harness for under $70. http://www.ebay.com/itm/281829377140?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT

My scan gauge voltage is now 14.1 volts.

Jim

thebitmaster
07-12-2016, 06:20 AM
The Executive Summary:

Through a Long Story (see below) I discovered that the boost code goes away almost 100% of the time if I disconnect the electrical connector on the small hose that Ts into the airbox-to-turbo hose at the turbo end. I've been trying to get a definitive answer to what this is, and haven't really gotten one yet. It's the one at the bottom of the picture:

https://c2.staticflickr.com/9/8780/27641330193_9263dd471f.jpg

I think maybe this is the crankcase ventilation hose, and the electrical thing is a little heater with something to do with condensation. If that's the case, here's my Hairbrained Theory about what's going on.

The vehicle has a bad positive Y-cable, with a 1 VDC drop from battery+ to alternator+ under heavy load. A replacement is already on order. So the system voltage reported over OBD2 is rarely more than 13.2 VDC. Having a bad Y-cable is know to cause a variety of sensor glitches. I think this is one.

Heating takes a lot of power, and I think when that little heater is plugged in, it's drawing power away from some other nearby sensor, modifying its output just enough to cause the code. The MAF sensor is a good candidate, though it might be something else. I could be completely wrong about that hose and its electrical doodad, but some variation of this is probably what's happening. When it's connected, its pulling power from an already starved sensor, causing the code.

For the moment, I'm leaving the electrical connector disconnected. The cable should be here in a few days. Hopefully turning it off won't cause serious problems before then. Knowing what it is would obviously help.

The Long Story: Get out your paper and pencil boys and girls!

After nearly losing my mind after replacing the intercooler didn't banish the code forever, I was grasping at straws. It seemed like I'd tried everything. I've replaced everything in the turbo path except the resonator, the big resonator gasket, and the rubber part of the two hoses. And I do mean *everything*. Each of those old pieces had been pressure tested to 15 PSI both in isolation and in system with no leaks, sometimes more than once. I'm fairly certain I fixed another leak by un-doing my hammer job on the turbo-to-resonator pipe bracket, which meant a 1/4" gap on the turbo side, but much less and almost none on the resonator side. Things got a lot better after that, I got almost a full day of driving with no more codes, and the system sounded better. But then the code started popping up again, and I could reproduce it with depressing reliability just as before.

So this morning I did another full-system pressure test, both with the intercooler output connected to the engine, and with it plugged. There were no bubbles or hisses anywhere. This is the second comprehensive pressure test I've done since I started, and while there is a *very* small chance that there's some kind of bizarre leak that only shows up under heat and changing pressure, I *really* doubt it. One thing I hadn't done on all previous test was spray the turbo shell and actuator with soapy water, so I did that too. I saw no bubbles.

As usual I was having a bear of a time getting the metal end of the intercooler output hose into the engine. Anybody have any tips for dealing with those clips? Mamma Mia! However, I saw the little sensor that's just above where the intercooler output goes into the engine. I figured it was probably some kind of boost sensor, and it looked pretty black and greasy. Having almost certainly eliminated input-side pressure problems from the system, sensor failure is definitely the next most likely candidate. So I pulled it out, got out my MAF cleaner, and hosed it off. It looked much better after I let it dry. I had a revelation, then, that this sensor was IN THE PRESSURE PATH and had a small rubber gasket that holds the pressure in. I had *never* explicitly checked this sensor for leaks, although I'm fairly certain it caught a lot of soap water overspray when I was hosing down the place where the metal and of the intercooler output hose mates with the engine. I cleaned the gasket and lubed it with fresh engine oil, taking care not to foul the sensor, and put it back. By this time I'd already removed the pressure setup, or I would have tested again and explicitly hosed down the outside of the sensor seal.

As I was re-assembling things, I investigated the electrical connector on the little hose that Ts into the turbo input shown above. It looked pretty dirty, too, so I used the last of the MAF cleaner to hose it off. It was pretty wet with cleaner, so I left it to dry while I finished reassembly of the rest of the system in prep for a test run.

After I got everything back together, I went for a 45 MPH test run. And it was *really* different than before! Not only were boost pressures higher, but the noise was mostly gone, and I got *no* codes for miles and miles. It was a fun drive! On my scanner, the most noticeable difference was that the Charge Pressure Positioner values were mostly in the low 50s, which I had never seen before at those speeds. Generally they tend to say in the high 70s and 90s unless I really floor it. I took several pictures.

So at this point I had two theories. One was that the problem, besides all the leaks, had been caused by that sensor I cleaned, which was probably the boost pressure sensor. Now that it was providing more accurate data, everything was working properly, and the different Charge Pressure Positioner values I was seeing was where they were *supposed* to be. Again besides the leaks, the extra hissy-ness might have been the turbo vanes being incorrectly set by bad inputs. The second theory was that maybe Vic had been right and the turbo linkage had a small catch in it that I didn't detect when I checked it before. It had felt as loose a bunch of paperclips strung together. I had gotten the lube and some acid brushes, but never went back to lube the linkages because it seemed like it wasn't a problem. However, I had just inundated that area with soap water, which is, at least temporarily, very slippery. It's also possible I got the contacts wet and they were giving bad data. I'm fairly certain there's some value to the second theory, because although I drove most of the rest of the day with no boost codes, part but not all of which happened after I *did* lube the linkages (at least as best I could with a headlamp and dental mirror, and without removing either the motor or the turbo heat shield. It's hard to be sure I got everything) I never saw the Charge Pressure Positioner values that low on average again. They returned to higher average values except on hard acceleration, and the turbo was a bit hissy-er. However, I drove over twenty miles under conditions where I'd previously been able to reproduce the code within less than a mile. I was slowly failing in my epic battle to not be happy and feel better.

Then I made the mistake of singing.

I had decided that the code had been gone long enough that it was time to put the engine cover back on. In the process of doing that, though, guess what I found?

I had never plugged the t-hose back into the turbo inlet hose, *or* re-connected its electrical connector. Remember when I left them off to dry? So then I realize I've been driving all day with a dime-size hole in my post-airbox turbo air inlet path, and felt Very Bad. I immediately re-connect them, put the engine cover back on, and got on the road...

...and almost immediately the turbo shut off with a low boost code. !@##%@#%#@!!

I reset the codes twice, and they came back twice in a few miles. I was back where I started AGAIN.

I got out and re-disconnected the hose from the turbo input and reset the codes again. I drove all the rest of the way with *no* code. I don't remember if I only disconnected the hose or both it and the sensor. At any rate, when I re-connected them again, BINGO, the code came back!

At first I was convinced that the easier airflow caused by removing the pipe and letting a small amount of unfiltered air in to supplement the airbox air might be correcting a bad sensor, the MAF being the most likely. I did a bunch of tests where I drove straight and level at 45 MPH with different combinations while watching the Load Value, Speed, RPMs, and MAF values. It did appear that the MAF values were about 1 lb/min lower when the hose was disconnected, which made sense. Was my MAF sensor slightly off, and this was compensating? Was it too hard to pull air in through the airbox? I had just gotten a new filter a few weeks ago. However, it was a FRAM filter and not a Mercedes or Dodge one. Could the other filter restrict airflow more than a factory filter?

However, at some point it occurred to me that maybe I should try with the hose plugged in, but the electrical connector disconnected. And when I did that, *there was still no code*. I then proceeded to go back and forth several times to prove that the code was caused by the connector being on. It was very, though not 100%, reliable. I had one extended trip with everything together and no code, and once and only once when I got a code with the thing unplugged. But overall, this was the most reliable fix I've seen so far.

So that's the story, and I'm sticking to it until I get the new Y cable. If I'm right, that will fix the problem and I can plug the connector back in. If that doesn't fix it, then at least I know that connector is involved and go from there.

Anyone who got this far in reading gets a cookie.

What a day!

Aqua Puttana
07-12-2016, 12:24 PM
Good job!! :thumbup:

The part is the Crankcase Breather Heater.

Assuming the NCV3 design hasn't changed much from my OM647 T1N...

77910

77911

The MAF sensor does have a 12 volt supply as does the Crankcase Breather Heater.

That gives credibility to your theory.

Certainly replace the Wye cable and maybe that will cure your problems completely.

But is it the voltage problem alone that is causing your problem?

It would be interesting to check the load on the crankcase breather heater and compare it to expected.

It may be a combination of an electrical connection/harness issue with the MAF being affected by the heater load. Or maybe your MAF is on the edge of being wonky and the voltage just tips the balance.

Not intending to rain on your parade, just suggesting that there may be a combination of issues.

:cheers: vic

P.S. - "Forgot to reconnect the part." There's been times when something like that has worked for me too. No amount of planning will ever replace dumb luck.

thebitmaster
07-12-2016, 10:45 PM
Good job!! :thumbup:

The part is the Crankcase Breather Heater.

Assuming the NCV3 design hasn't changed much from my OM647 T1N...

77910

77911

The MAF sensor does have a 12 volt supply as does the Crankcase Breather Heater.

That gives credibility to your theory.

Certainly replace the Wye cable and maybe that will cure your problems completely.

But is it the voltage problem alone that is causing your problem?

It would be interesting to check the load on the crankcase breather heater and compare it to expected.

It may be a combination of an electrical connection/harness issue with the MAF being affected by the heater load. Or maybe your MAF is on the edge of being wonky and the voltage just tips the balance.

Not intending to rain on your parade, just suggesting that there may be a combination of issues.

:cheers: vic

P.S. - "Forgot to reconnect the part." There's been times when something like that has worked for me too. No amount of planning will ever replace dumb luck.

Thanks! I'll have to remember that quote!

I would really like a way to test the MAF sensor to see if it's in spec, as well as the crankcase hose heater. When I did the tests with and without the crankcase breather heater plugged in, I recorded the MAF values along with the engine load, RPM, and speed to see if I could see a difference in the values. They were about the same. They were about 1 lb/min different when the crankcase breather hose was unplugged, but that's what I'd expect when some of the air is bypassing the MAF and going into that hose. I really wish I had another know-good MAF to swap it with, and if necessary I'll find one. Of course it could be some other sensor, like the O2 one, which I haven't been able to find yet. Knowing which sensors the ECM uses for the turbo boost sensing would be super helpful.

Vic, when you lubed your turbo linkages, specifically which parts did you lube? I mainly hit the places where the linkages join, like the clip, the next elbow, and the pin where the linkage disappears into the turbo. I didn't remove the turbo heat shield to get inside, or do any lubing on the electrical motor (which the NCV3 uses instead of a vacuum actuator.) Should I have done more? I'm really wondering about those boost actuator %s in the 50s. I've never seen that before, but I got a smoother and quieter operation of the turbo when they happened. That was after I sprayed down the turbo and the actuator with soapy water. Two hours later, the values went back to the more common and noisy 75-95% range at 45 MPH. It seems most likely that happened because of the soap water. I had also just cleaned that sensor between the intercooler and the butterfly valve, which is probably the boost sensor. It's possible that it was clean for a short period of time before getting somewhat fouled again. There is definitely black oil in the turbo pressure path. There should be less now that the EGR is permanently closed.

I'll be doing a heavily loaded storage run today and that will be a good further test drive to see if the heater being disconnected keeps the turbo on. I've got the wiring harness on order from Group 1 Auto, but they haven't shipped it yet.

Aqua Puttana
07-12-2016, 11:50 PM
I've had no turbo problems. I just brushed around wheel bearing grease on the exposed parts as a preemptive strike.

If you think that the soap solution might have provided temporary lubrication my suggestion would be to spray some WD40 around. That will provide some similar, but slightly longer lasting lubrication. WD40 shouldn't have any negative consequences.

vic

thebitmaster
07-16-2016, 03:47 PM
...and I just lost my five page post. :bash:

Replaced Y cable. Everything looks correct there. 14.2 v, 0.07 v drop from alternator+ to battery+, etc.

Low boost code continues. I tried:

3 runs with the Crankcase Breather Hose plugged back in. Same low boost code at low speeds within 1/2 mile

1 run with the electrical connector on the CBH rotated as far away from the Boost Pressure Regulator to prevent crosstalk. Still have the code

3 runs with the CBH electrical connector disconnected. Previously this almost completely eliminated low boost codes. Not anymore.

Finally I plugged that back in and tried pulling the hose out of the turbo air inlet pipe. At this point something more interesting happened:

I got three new codes:

2978-002 : Y27/13 (Exhaust gas recirculation cooling solenoid valve) - Short circuit to ground

2527-002 Y27/11 (Exhaust gas recirculation positioner) - Short circuit to ground

2526-004 Y77/1 (Boost pressure regulator) - Signal line is interrupted

When I was connecting and disconnecting the Crankcase Breather Hose Heater, I was having to pull on the wires harder than I was comfortable with to get it back on. Once it was on there wasn't as much tension, but the process of getting it on or off felt like too much pulling. That cable physically overlaps with the Boost Pressure Regular cable. I'm wondering if there might be some kind of short there, and if that short might be a Root Cause in this situation.

These codes popped up after I did the Y cable thing, but *not* right away. I was able to do another 6 or 7 test drives without them, and the turbo was doing the same thing it usually does. The Boost Pressure Regulator would start at 95%, them move between there and about 70% as I was driving. Then, when the ECM kills the turbo and sets the low boost pressure code, it would jump to 5.43% and stay there.

The EGR codes are unclearable, i.e. they survive a DTC reset. The new boost code pops up again while the vehicle is in park, which means the Boost Pressure Actuator starts disabled at 5.43%.

So although it's nice to have 14.2 V and I bet most of the system is happier, I'm worse off than I was before. :yell:

One thing I noticed when re-connecting the negative battery cable after the Y-cable replacement was that I didn't need to pull the red tab out in order to pull the cable off. Moderate pressure popped it off *without* having to pull the red tab out. When I clip it back on, it definitely *snaps* into place. This ends up with the red tab even with the edge of the black box. From this position, I can pull the red tab out as you would expect to remove it, but I don't *have* to. I can just pull on the box and it pops off even with the red tab in what *looks* like the locked position. I tried fiddling with it to see if I could push the red tab in further to lock it down better, but was unable to get the bottom of the red tab closer to the cable connection than the bottom edge of the black casing. I'm assuming there's an equivalent to the Y-positive cable load test for the ground wire? I'll also check to make sure I didn't loosen the negative connection in the battery box while upgrading the positive Y cable.

I cleaned the contacts on the Boost Pressure Regular connector and the EGR valve connector with B12 chem tool, and will probably get some actual sensor cleaner today. I let them dry overnight and will see if there's any improvement. If not I'm going to have to start doing circuit testing on those sensors and see if I can find a short or disconnected wire.

I'm open to suggestions. I'm leaving for Portland in less than 10 days and I don't think I'll make it over those mountains without the turbo. :bow:

rpmckee
07-16-2016, 04:00 PM
Hello, I spent the last half hour looking through the post and I still have a question. The hose on the driver side has play in it and the one on the passenger side has none, the aluminum connection. I take it this is not right.

https://goo.gl/photos/ZVb8u8i3LPMQDSuNA, here is a video of hose.

Thanks in advance, I haven't had time to try some of the other creative test.

thebitmaster
07-16-2016, 05:00 PM
Hello, I spent the last half hour looking through the post and I still have a question. The hose on the driver side has play in it and the one on the passenger side has none, the aluminum connection. I take it this is not right.

https://goo.gl/photos/ZVb8u8i3LPMQDSuNA, here is a video of hose.

Thanks in advance, I haven't had time to try some of the other creative test.

I watched the video, and it looks to me like the clip on that hose is not on correctly. When the clip in place, the *only* part that should be visible from a end-on viewpoint (looking straight into the hole) is the little tab that pulls out. The part of the clip closest to the camera looks out of place. The clip slides into a groove that goes all the way around the pipe. When installed correctly with the hose *off* and looking into the hole, you should *only* see the clip on the outside in the middle where the little tab sticks out, but you should also see the clip from the *inside* of the pipe on the left and right sides where the tabs on the metal end of the hose go. The clip portions visible inside the pipe are what locks into the little grooves in the two tabs on the metal end of the hose and holds the entire join together. Your clip looks out of place.

I've found the clips are a huge pain to work with. I've had the most success using two flathead screwdrivers to pry the outsides of the clip near the tabs outward while sliding the metal end of the hose in with the third arm I had added to improve my ski-boxing. When it's all together, the clip should be inside the same groove all the way around the end of the pipe, the metal end of the hose should be 100% flush with the pipe end, and there should be *no* wiggle.

thebitmaster
07-16-2016, 05:03 PM
I'm also worried about my ECM now. It hasn't seen 14.2 V in a very, very long time. After I get some proper electrical contact cleaner I'll clean the contacts on it as well.

appltech
07-18-2016, 03:24 AM
I've been fighting a similar problem also. My clean air hose kept coming loose form turbo. It a fairly new hose and gasket (6months) New pcv hose assembly. But the lastcouple weeks its lhm on and off . Got hose to stay on but codes keep coming back . So buddy I feel for you.

thebitmaster
07-18-2016, 03:40 AM
Finally decoded FSM enough to trace the codes back to a common fuse, #8
, a 20A below the driver's console. Sure enough, it was blown. I replaced it and added the sensors in question back one at a time, checking the fuse after each. Eventually all codes were corrected but the crankcase vent one. I didn't have a way to verify the heater itself, so i ordered a new one. I added a new boost temp and pressure sensor for good measure, since this latest storm of sensor craziness was a distraction from my main problem. Of course, if there's been a problem with the crankcase ventilator for a while, it could have been fritzing the turbo actuator from day one. It's possible it's the connector and not the heater itself, but the 12V+ was a solid 14.2v and there was a good connection to the ECM on the signal wire. From one to the other with the engine idling was 10.2v. I'll leave that wire unplugged until parts arrive.

thebitmaster
07-18-2016, 10:06 PM
I've been fighting a similar problem also. My clean air hose kept coming loose form turbo. It a fairly new hose and gasket (6months) New pcv hose assembly. But the lastcouple weeks its lhm on and off . Got hose to stay on but codes keep coming back . So buddy I feel for you.

It might be worth checking out this bulletin. I'm certainly not replacing the gasket every time I take it apart, but it also mentions the correct orientation and the specific torque for the clamp. I used a Fat Wrench torque screwdriver, commonly used for firearms maintenance, to get the clamp to the correct torque.

http://sprinter-source.com/forum/showthread.php?t=48673

thebitmaster
07-22-2016, 01:26 AM
My new boost pressure (MAP) sensor, boost temperature sensor, and crankcase vent heater arrived today.

1) The MAP sensor, which I ordered from Europarts SD, had a severely damaged gasket. Almost half of the gasket was worn away and malformed. This is the third time I've received a damaged part from them. Last time it was the intercooler, which arrived with a large dent in the manifold where it very well might have caused a leak, and that was due to inadequate packing. It was inside a cardboard box with only one small cardboard insert and no additional packing material to protect its fragile manifold. When I complained to them about the shipping in e-mail, I never got a reply. This time I raised additional Cane, and am complaining here publicly in a venue where they get a lot of business. The first problem was a ding in an aluminum pipe that clearly didn't effect its functionality, and I never informed them of that problem until the second problem occurred. I had high hopes of finally seeing and end to this months-long saga as I tried literally the last part that could be causing the problem, and now I have to either wait for them to ship me another MAP or buy a more expense and non-Bosch sensor locally, which I really don't trust.

https://europarts-sd.com/ is the site that I've had the problems with.

2) I installed the new boost temperature sensor, and it did nothing to fix the problem. I was hoping that even if the sensor itself wasn't bad, that maybe the gasket old gasket was causing a leak. I never actually did a soapy water test on either the boost pressure sensor or the boost temperature sensor because I didn't realize they were in the pressure path.

3) I got the new crankcase ventilator heater, but plugging it into the connector didn't solve the error code I was getting:

2415-004 R39/1 (Vent line heater element) Signal line is interrupted

This is both good and bad. It's good because it might mean I get to leave the old heater on there. It turns out that I haven't the slightest idea how to get the old one off. It's held onto the rubber hose with a metal clamp that appears to be all one piece with no method of removal. I see signs it might have been crimped on:

https://c3.staticflickr.com/9/8848/28355728082_063ac5e204.jpg

How the heck does that clamp come off without damaging the hose? A dremel? Liquid nitrogen?!?

Also the new one came with a gasket.

https://c1.staticflickr.com/9/8894/27843749584_8853b50602.jpg

Where is that gasket supposed to go? Certainly not into the turbo inlet hose...there's no way it would fit. Currently, there is *no* gasket on the part of the sensor that fits into turbo inlet hose. I tried putting the gasket in there by itself, but it's clearly too big and can only be made to fit by almost bending it in half. At first I got excited when I saw the new gasket, thinking that maybe there's *supposed* to be a gasket there and maybe it's absence was causing an inlet leak and contributing to my low turbo boost code.

So where is the gasket *supposed* to go? I can't find anything beyond the code testing for the crankcase ventilator heater in the FSM beyond the stuff for diagnosing the code, which is electrical and doesn't address the physical process of removing the heater from the end of the pipe. Also, there is a little arrow on the heater. Right now that arrow points toward the turbo inlet and not toward the crankcase side. Judging from the construction, I think this is a rare instance where it actually doesn't matter which side is which with regards to the hose and the turbo inlet. It looks like it's just a copper surface that gets heated up and is perfectly symmetrical on both ends. I'm wondering, though, if mine is hooked up "backwards." Which way should the arrow point?

But the reason it's bad is that means that it's probably not a problem with the heater, but with the heater wiring. As I mentioned earlier, the hot wire was at the correct voltage and the signal wire to the ECM passed a resistance test. The volt across them with the engine idling was 10.some volts. I also mentioned earlier than fiddling with the heater blew a fuse that the turbo actuator and EGR also live on. I'm wondering if maybe there's a short or broken wire somewhere that might be affecting the boost actuator.

I *really* wish the MAP was good so I could see if that fixes the turbo problem. The last thing I want to do is going on a wiring harness goose chase to find that short. I'm gonna re-do the vent heater part of this and start a new thread for it.

thebitmaster
07-22-2016, 04:06 PM
Removed the solid clamp on the crankcase vent heater with a dremel. Not happy with the results, but it's off.

http://sprinter-source.com/forum/showthread.php?t=49069

There was some minor damage to the hose, but I don't think it went all the way through. I'll order a new one next time I go for parts, but am not looking forward to the same operation on the PCV complex. I'm installing the vent with the round part of the connector toward the front of the vehicle, as I see here:

http://www.benzworld.org/forums/attachments/w164-m-class/750193d1390600265-question-about-crankcase-vent-oil-going-1.jpg

Once that's done, I'll go for a test drive and see if the new heater's relative resistance stability helps with the boost problem. Considering how wildly the old heater's resistance was changing *when it was disconnected and unpowered* I can only imagine how much havoc it might have been causing on the turbo actuator, which is on the same fuse *and* who's wire harness literally touches the heater one.

If the problem still exists, GDE sent me a new tune to try when I submitted my review today. Hmmm. Note that I tried reversing the tune I already have earlier, and instead of taking 1/4 mile for the low boost code to occur, it happened almost immediately.

If the vent heater doesn't fix it, then I'll try the new tune, and the factory firmware again while I'm waiting for the replacement boost pressure MAP to arrive.

thebitmaster
07-25-2016, 10:45 PM
So I finally have a usable new MAP/Boost pressure sensor in my hand, and I'm stuck on that second screw! Lindenengineering said that it could be removed with a mini ratcheting wrench, but I'm not certain what size? It looks like 7mm. I cannot *BELIEVE* I am this close and being stymied by one screw!

https://c3.staticflickr.com/9/8597/27932824234_35c66b8db8.jpg

:yell:

lindenengineering
07-26-2016, 01:22 AM
You need a Snap ON wobbly 1/4 drive torq bit socket and several short extension swaps to crack it loose. Plus the 3" ratchet.
Dennis

lindenengineering
07-26-2016, 01:46 AM
Here's a suggestion which might be beneficial!
Why not remove the throttle body to get immediate access to the sensor !
This is held by torq bolts and a long (90 mm long 6x1mm) screw!

There is a square section ring seal that seals the body/adapter to the cross over tube. A source of lost turbo boost in many a circumstance and I always bust it open for examination simply because I don't want to overlook it and see the bloody thing back again!
Dennis

thebitmaster
07-26-2016, 03:16 AM
You need a Snap ON wobbly 1/4 drive torq bit socket and several short extension swaps to crack it loose. Plus the 3" ratchet.
Dennis

Dennis,

I really appreciate your quick reply!

I think I'm getting a little confused by your tool terminology. I've got a Duralast E8 torx bit, 1 inch in length, that I can't even get onto the back bolt when it's not hooked up to *anything* else. There isn't enough clearance even for that, much less something to hook it to. I used that bit on a universal 1/4" universal joint with an long extension to get the front one off (after removing the EGR tube to get myself some room) at which point I dropped it and lost it forever.

So I have four pieces, the 1/4" ratchet, the 4" 1/4" extension, the universal joint, and the E8 socket. Can you break down your pieces the same way? I did a Google search for "Snap ON wobbly 1/4 drive torq bit socket" and wasn't able to put the pieces together. Maybe a picture?

I've been chasing this boost leak forever, so the idea it might be happening at the throttle body seal is both wonderful and horrible. I thought I'd run out of things to fix except for the turbo itself and the actuator motor. The MAP was last in line. I still get a hissy noise when the turbo runs, even though I've pressure tested the boost path from the turbo to the end of the last intercooler hose to death.

I had to back out the big screw in the throttle body to get to the first screw on the MAP. I have the Dodge/Chrystler/Jeep Service manual, but I can't find Throttle Body in there. It's probably under some alternate name like so many other things. Which section would it be in?

Sorry to ask so many n00b questions. The most complicated thing I've worked on before this is a Geo Metro, which is basically a lawn mower with a throttle body fuel injector. :)

thebitmaster
07-26-2016, 05:53 PM
Also, I talked to Steve at https://europarts-sd.com and got things straightened out. He never received my e-mail about the damaged intercooler, and offered to replace it in the future if it gave me problems. He also had immediately shipped me a replacement boost MAP as soon as he found out about my problem, which I now have. *I* managed to miss *that* e-mail. So although there were several issues in a row, they've been resolved to my satisfaction. In the future, if I don't get a reply to a complaint e-mail, I'll follow up with a call.

thebitmaster
07-27-2016, 04:45 AM
I finally replaced the boost MAP sensor. It turns out that a 1/4" 12-point box wrench is exactly the right size to use on the E8 torx screws. It was a *massive* pain, especially to get the back one back on, mainly because the screw is only accessible for less than 90 degrees because the MAF has triangular flanges on the side. That 90 degrees is largely blocked by the EGR housing, leaving only a tiny amount of room for 1 point-at-at-time rotation. I chose to break those flanges off on the old one. My wrench I had to come in from the lower left, facing the vehicle, under the spherical reservoir, while reaching in from the top with the other hand to hold the end of the wrench in place with my fingertip.

What *would* be perfect, however is a 1/4" *offset* 12-point box wrench with a very thin edge. That would allow the wrench handle to be above the flanges just enough to get it on tight. They had a perfect one at the auto parts store as part of a set. Instead I ordered an individual one off Ebay for $6 including S/H.

The Bad News is that it didn't fix the problem. The readings from the new sensor were pretty much the same as the old ones, and I can still reproduce the low boost code quickly. It seems like it take a little longer, but I only had time for a few test drives.

Tomorrow I'm going to try some WD-40 on the moving parts of the turbo motor, despite having lubed the linkages with heavy grease before.

Lindenengineering suggested disassembling the throttle body to check the seal, which I'd like to try. The turbo still exhibits "hissiness" despite all my pressure testing. I've now replaced both the boost temp sensor and the boost pressure sensor, which were my last two eligible sources for both bad sensors and gasket leaks. Unfortunately either I can't find the throttle body in the FSM I have, or the page is one of those who's graphics are fubared and offers no usable information. I'm looking into online resources, including this forum, for help.

I started another thread regarding what appear to be high back pressures on the NCV3 list, but I'll copy that post here as well next.

thebitmaster
07-27-2016, 04:48 AM
My question here is, do these exhaust back pressure readings look normal? If back pressure is too high, it would reduce the turbo's performance, right? Because the exhaust wouldn't be flowing through the turbo unobstructed?

https://c2.staticflickr.com/8/7306/2...c017961e93.jpg

Engine Speed: 1707 RPM
Intake Pressure 998 hPA (14.47 PSI)
Boost Pressure 1290 hPA (18.70 PSI)
Exhaust Back Pressure 2033 hPA (29.48 PSI)

https://c5.staticflickr.com/8/7520/2...1cc8f37a13.jpg

Engine Speed: 1590 RPM
Intake Pressure 989 hPA (14.34 PSI)
Boost Pressure 1272 hPA (18.44 PSI)
Exhaust Back Pressure 1916 hPA ( 27.78 PSI)

https://c7.staticflickr.com/8/7452/2...a665839ab7.jpg

Engine Speed: 1869 RPM
Intake Pressure 989 hPA (14.34 PSI)
Boost Pressure 1019 hPA ( 14.77 PSI)
Exhaust Back Pressure 1124 hPA ( 16.30 PSI)

thebitmaster
07-27-2016, 07:54 PM
I'm also still intermittently getting the afore-mentioned crankcase ventilator heater code despite having replaced the part with a new one. That part shares a fuse with the turbo actuator, and their wire bundles overlap, which makes me nervous. When I tested the power and EMC wire for that parts, it looked ok. I did a lot of wiggling to see if I could induce a short or something, but naturally it didn't happen when I was doing it.

thebitmaster
07-27-2016, 10:12 PM
I ordered a new resonator even though the old one has been pressure tested to death.

I also ordered new gaskets for the front EGR pipe and for the throttle body. When they arrive I'll replace all parts, do a test drive, and repeat a full-system pressure test if the code remains.

I sprayed some WD-40 on the turbo actuator linkages, which I had previously greased with heavy grease. I'm going to do the portion of the FSM's electrical testing on it that I can do without a Star tool, and maybe take it off and see if anything looks crazy. I tested the freedom on the linkage before and it felt nice and loose.

Does anybody know if there is supposed to be any leak-by if pressure is applied at the turbo but the rest of the system is left intact? There was in previous tests, but I assumed it was normal. When I blocked the end of the intercooler output hose, I got a 100% leak-free setup where my compressor never cycled. With the hose still hooked up, the compressor would cycle fairly often. I never did any soapy water spritz testing above the intercooler output hookup because I didn't know what to squirt and didn't realize leaks could still happen up there.

thebitmaster
07-28-2016, 05:29 AM
Does anybody have the pinouts for the NCV3 boost servo motor? I spent over an hour searching through the FSM, and although it has the test procedure for the motor, I can't find the wiring diagram for the connector *anywhere* in either the repair side or the wiring side.

thebitmaster
07-28-2016, 05:05 PM
At Dennis's suggestion I'm going to loosening the exhaust after the turbo to see if that solves the back pressure problem.

Grins2Go_Brett
07-29-2016, 12:23 AM
Wow. Definitely a lot of information here. I'll be watching to see if the throttle body seal cures your problem. BTW, I want my cookie for reading the entire thread! :smirk:

thebitmaster
07-29-2016, 01:20 AM
So I disconnected the exhaust pipe from the turbo, and I very rarely got the boost code after miles of hilly driving. This is the biggest improvement I've ever seen, and pretty much means either actual back pressure from a clogged exhaust or a bad pressure sensor are causing the problem. I also had a few glow plug issues right at the end that might have been because of the exhaust gas escaping into the engine compartment. I'm also very low on oil.

Now that we know either the exhaust or the pressure sensor are part of the problem, what's they best way to narrow it down?

thebitmaster
07-29-2016, 01:22 AM
Wow. Definitely a lot of information here. I'll be watching to see if the throttle body seal cures your problem. BTW, I want my cookie for reading the entire thread! :smirk:

Dude, you don't just deserve a cookie, you deserve a medal!

It looks like exhaust back pressure or a bad sensor for the same might be causing the current problem, see above.

72chevy4x4
07-29-2016, 02:47 AM
but you still got the code w/ the exhaust disconnected? If you got the code 'at all', I'm inclined to believe there is a pressure sensor that has readings on teh borderline, and the disconnected exhaust decreased the backpressure enough to get it within specs most of the time.

by disconnected, do you mean there is no exhaust connected or it's cracked open somewhere?

thebitmaster
07-29-2016, 04:01 AM
but you still got the code w/ the exhaust disconnected? If you got the code 'at all', I'm inclined to believe there is a pressure sensor that has readings on teh borderline, and the disconnected exhaust decreased the backpressure enough to get it within specs most of the time.

by disconnected, do you mean there is no exhaust connected or it's cracked open somewhere?

Good points. I just removed the bolts connecting the turbo to the exhaust pipe. I verified there was a small gap, but didn't disconnect any other brackets to make the gap really big.

The ambient and intake pressure sensors track within 1 hPa, so I think they're good. I changed out the boost sensor, and the new one is the same as the old one, so I don't think that was it either. It's very possible the exhaust pressure sensor is messed up somehow, but it's pretty expensive, so I'm going to take it off an look at it before I go any further.

The back pressure was still higher than I'd intuitively expect most of the time, so I'm guessing only a little pressure was getting out at the join. I could certainly smell it. There were times when the boost pressure was pretty high, maybe 1500 hPa, and the exhaust back pressure was only 1700 hPa, which is the closest I've ever seen them at mid-level boost. More often it was closer to with the screws still in and pipes together, which would be like 1300 hPa boost and 1800+ hPa exhaust back pressure.

Jury is still out about whether or not it's the pressure sensor or the cat. The DPF pressure drop, reported by my scan tool, is negligible. I'm gonna look at the sensor tomorrow when its cooled down. However, I drove for over an hour with only those three codes, two very close together, and I would have gotten about 30 or more during that same time, so we're definitely onto something. I wish it were easier to try swapping a pipe for the cat, but it looks like everything is welded. I've got two welders on property, though, so maybe it's not to bad...

thebitmaster
07-29-2016, 01:42 PM
I'm continuing to pursue this on this thread:

http://sprinter-source.com/forum/showthread.php?t=49189

This one has gotten really long, and we're on to a different potential solution. The conversation will continue on *that* thread until I reach a resolution, at which point I'll come back and post a summary here. So please tune in there if you want the latest!

Grins2Go_Brett
07-29-2016, 06:39 PM
Dude, you don't just deserve a cookie, you deserve a medal!

It looks like exhaust back pressure or a bad sensor for the same might be causing the current problem, see above.

No medal needed. :thumbup:

Glad to see (in the other thread) that you seem to have found your leak. Keeping my fingers crossed for you. :rad:

SprinterHelp
09-06-2016, 05:05 AM
2007 2500 3.0
Going thru same horror, lhm intermittant, more hotter days. Started after heavy heavy rain entered air intake and was very intermittant.
Became problem after turbo intake loosened up and drove several hundred miles in rain taking in raw unfiltered air.
Inspected everything in circuit hoses, seals egr replaced tube, hoses, metal connectors, cleaned DPF, EGR cooler and accessible sensors.
Have code on lcd display [\]O not sure if thats O2 sens. P242f code needs to be erased.
New EGR, runs very very good but still drops lhm at high boost when it wants to, more when driving same speeds, much less when driving without cruise. Turbo begins to whine lightly low load set off lhm, much less if stepping to the floor so to speak any many times no lhm when flooring it up to 29.5 MAP which seems about max.

I did some work on low milage sprinter today changing actuator and see that the green ring seal on 2007 avion motorhome sprinter was much wider in girth and sealed much better, appeared to be glued in place with brown glue....no more after clean. Looks like mine has shrunk. (TWSS)

Also noticed pvr oil fumes entrance at turbo intake was so tight that it cannot be removed. Mine is lightly sealed in comparison and can be removed with little effort.

Need to check turbo vanes seem much more worn at 340k as opposed to 35k. Slight browning and scratches in turbo near blades, just the slightest play cant imagine vanes striking housing yet.

Will repair seal issues tomorrow.
Hope sensors do it.

SprinterHelp
09-11-2016, 01:17 PM
Comparing the vehicle running similarities I after checking intercooler which shown no leaks went to check the authors leak location. Dishsoap and water i found the lower egr seal bubbling. 299 code last one showing. Glad i read this very long thread three times. Anyways guess three weeks ago i did that gasket repair incorrect.

So will see in an hour.

SprinterHelp
09-12-2016, 06:31 AM
Stripped egr bolt lower side of tube.
Tapped, leak no longer..... still getting 0299...
God please be in the seal behind the turbo.

The vehicle now surges lightly at idle.

Amen
Have not gotten around to y cable, but one thing is this all happened due to heavy rain, each wet period lhm became more prevalent. Maybe thats my surge, scanguage ii shows 13.9 14.00 volt.

Rob S
09-12-2016, 02:30 PM
If this is an NCV3 the (2007 Model Year so yes it should be an NCV3) the air intake is under the LH side
of the grill in the top rear of the hood.
The air then needs to go all the way across to the false grill on the RH side of the hood.
Then it has to go through the air filter box and the air filter before it gets to the large
O' ring that seals the outlet from the air filter,

Unusual, that...

On my 2007 Dodge (2008 Navion) NCV3 the "across the hood" air passage only supplies the cabin climate control by way of Cabin air filter located passenger side against the firewall.

Combustion air enters just under the lip of the hood on the drivers side.

I think they are all originally like that.

However, the long air pipe and the air filter top dead centre would still prevent any water getting to turbo, unless the filter element is missing

I think we might be more talking about moisture affecting and worsening the corrosion of the connections of the Y-cable and or ground cables. If these have not been checked, replaced, cleaned etc, that would be high on my list.

bcislander
09-12-2016, 05:10 PM
Unusual, that...

On my 2007 Dodge (2008 Navion) NCV3 the "across the hood" air passage only supplies the cabin climate control by way of Cabin air filter located passenger side against the firewall.

Combustion air enters just under the lip of the hood on the drivers side.

I think they are all originally like that.

However, the long air pipe and the air filter top dead centre would still prevent any water getting to turbo, unless the filter element is missing

I think we might be more talking about moisture affecting and worsening the corrosion of the connections of the Y-cable and or ground cables. If these have not been checked, replaced, cleaned etc, that would be high on my list.

+1, the intake at the LH rear of the hood (bonnet) only supplies air to the cabin.

Combustion air enters via a downward facing tube (#200 in the diagram) that plugs into an intake 'box' located at the LH front side of the engine compartment, near the top of the radiator. IMO, short of a major flood, there is no way that water could enter the turbo via the air intake system.

+1 on the Y-Cable/Ground strap. Given the voltage spikes & 13.9/14.0 V 'running' voltages, the Y-Cable and grounding straps should have been the first things replaced (Y-Cable) and/or cleaned (Ground strap).

SprinterHelp
09-14-2016, 01:55 AM
+1, the intake at the LH rear of the hood (bonnet) only supplies air to the cabin.

Combustion air enters via a downward facing tube (#200 in the diagram) that plugs into an intake 'box' located at the LH front side of the engine compartment, near the top of the radiator. IMO, short of a major flood, there is no way that water could enter the turbo via the air intake system.

+1 on the Y-Cable/Ground strap. Given the voltage spikes & 13.9/14.0 V 'running' voltages, the Y-Cable and grounding straps should have been the first things replaced (Y-Cable) and/or cleaned (Ground strap).

Beg to differ, running thru iowa about month and a half or two had severe heavy rains. Ran into a few 10 inches of puddling doing 50 60 ish, one time between two semis, no fun. Several cars stalled out. Water had entered up thru the grill snorkel and soaked my filter with road oil and grime which nearly collapsed the filter to the point of stalling.... It was late and nap time, next day went straight to get new filter. Ever since encountered limp home ocassionally when significant rain. Were not talking rain I am talking the type of rain just before a twister. Never seen so heavy.

Well going to check that Y cable, my voltage still fluctuating 13.7..8 14 with drops to 13.1 ocassionally.
Swirls are working properly.
Modus scanner shown no codes except charge leak whic I cant find.
:bash:

Cheyenne
09-14-2016, 08:00 AM
...next day went straight to get new filter. Ever since encountered limp home occasionally when significant rain.


Is the new filter seated correctly and not allowing water past?

Keith.

thebitmaster
12-04-2017, 10:03 PM
My epic previous experience with this problem is repeating itself:

https://sprinter-source.com/forum/showthread.php?t=49189&page=4

TLDR I replace the EGR crossover tube because it had a hole about a year ago. I was very careful to make sure the two clips were in place and the entire assembly was finger tight before I symmetrically tightened down both sides of both bolts. There was little or no tension in the system, all the screws lined up fine. I *did* manage to strip one of the EGR cooler holes despite using a torque wrench, so I replaced that bolt with a longer 10mm bolt and some extra washers to eat the extra length. After the repair my Low Boost code finally went away and there was no longer a bad smell.

Today I found that exhaust gas is coming out of a crack at the join of the tube and the header just under the EGR cooler, very close to where I had the problem before. This is super frustrating, as the part is expensive, a huge pain to replace, and I did all this not too long ago. I suppose there's a small possibility I did something wrong, but I also have to wonder if excessive back pressure could make this more likely. I've never got a code, nor a decent answer to what nominal back pressures might be.

I'm gonna talk to Europarts SD about any warranty on the pipe. I also have the GDE tune, which means that my EGR valve is always closed. Obviously there's still pressure at the failure junction, but if the hardware itself isn't necessary any more, I'm going to make or buy some blocking plates and call it a day.