Rear EGR crossover pipe keeps breaking.

NMccann92

New member
I have a 2007 Dodge Sprinter that has been breaking the EGR crossover tube that branches from the exhaust manifold Y-pipe in the rear of the engine that goes to the EGR cooler. It has gone through 3 tubes now and always either seperates from the bottom flange closest to the egr cooler or cracks at the bottom of the tube in the same area. Anyone ever had an issue like this? I have the top bracket bolted theres another clamp but I cant see where it would go to. I have started working on this truck about a year ago and am not sure if the people working on it before removed the bracket and didnt put it back on, or I possibly may have when I replaced the turbo. Thank you.
 

edfrompa

2008 ROADTREK on F/L 2500
Might take a look at your motormounts.

Just my $0.02, others may see things differently.
Good luck, Ed
 

irvingj

2015 RT SS Agile (3.0L)
Ditto on the motor mounts. Sounds like excessive movement in there somewhere.
 

NMccann92

New member
Took a look at the motor mounts and didn't see anything too out of the ordinary. It does vibrate a little at idle, not too much though. It is at Mercedes now to see if they can see anything with computer.
 

rollerbearing

Well-known member
I have no experience looking at Sprinter motor mounts, but on other vehicles it helped to see loose mounts by loading the engine. All brakes on, wheels chocked, observer to the side. Goose the motor while it is in gear. I remember seeing some of the V8s in MB SLs lifting several inches on one side. Wonder that the air cleaner didn't push up through the hood.
 

lindenengineering

Well-known member
The motor mounts --well engine mountings because its an internal combustion engine will not affect the durability of the tube you are mentioning. Simply because the EGR cooler assy and the exhaust turbo Y pipe are all bolted firmly the engine structure.

So what causes it to break?
Ham fisted installation first off!
DO NOT bend the tube to fit it--If it won't align up then there is something misaligned--maybe the cooler not fully bolted up to the head. Investigate.
Remember the concertina elbow end goes towards the turbo.
Don't forget there are two anti vibro retention clamps that hold the pipe to the turbo Y pipe .
Tip:- Offer the pipe up with the flange bolts and gaskets as a unit. Then start all the bolts in their respective threaded holes then carefully secure it so that you don't bend it during the snug down.
Take into account that the EGR cooler end casting is made of very soft grade alum (cost about $900 to replacee) and it is very easy to tear out the threads. Again use new bronze bolts for the turbo end of the pipe , (6x1mm torqs) from the dealer .
If you do tear out the thread(s) use a timesert to repair it!
Overall don't force it nor bend it!
Now you are not on your own!!!!

There is a moron who works for a MB dealer West of us who must be using a big ring wrench to align the tube judging by the mangled condition we see 'em in since they will last about 500 miles before breaking up again! (They appear in my shop for a giggle or two)
Flat rate is a "wonderful thing" !!
Hope this helps & all the best
Dennis
 

NMccann92

New member
I would imagine it is the fact that only one of the brackets is bolted down as I dont know where the other one goes. As far as installation, I've never had to force it tjey've always just slipped into place. All bolts are installed by hand tight until both ends of the tube are seated down and tightened firmly but overtight. I have not used new bolts the last couple times so maybe thats it also. Thankyou Dennis for the insight, it is at the dealer now I told then to replace tube but to also figure out what the cause is. I will report back what the diagnosis is from Mercedes.
 

NMccann92

New member
Turns out it was missing hardware for the EGR tube which I can only assume to be the second bracket needing to be bolted down. Although I have no idea where it would go, but we will see if it holds up. On the way back from the dealership truck went into limp mode after boost was fluctuating and then complete loss of turbo function. Went back to shop to inspect under hood and removed turbo outlet tube to see if the o-ring was rolled because they replaced a couple of the CAC system o-rings and seals, only to find the o-ring was missing! Thankfully we have parts on hand at the shop, installed the o-ring and retest. Everything okay at this time. Be wary of the dealership, just because it's Mercedes doesn't mean they are perfect.
 
With no metal working experience, I attempted a repair of a blown-out EGR long pipe using Muggy Weld's SSF-6 1150 o F silver solder. I screwed up just about everything that was possible to do wrong, but I think the fix has worked. Still have to pressure test it with something besides my mouth. :bounce:

The details are here, but in general I wrapped the pipe with stainless steel wire to cover the hole, then filled it all in with the SSF-6 to glue it all together. I also blew another hole in the pipe in the process because I'm a beginner, but the same fix worked for that too. :)

I general, this technique would probably work without the wire coil if it's the more common failure of the flange starting to leak right at the seam. In that case the solder could be used to simply flood the gap.

Once I got done I went back and flooded the *other* end (which was intact) with the same stuff to re-enforce that join. I saw some evidence that the flange might have been brazed on at the factory in the first place, which might explain why that's a common failure mode.

I also re-built one of the supports using a stainless hose clamp, as the original had broken off. I have to wonder what kind of abuse caused that to happen in the first place. Those supports are thick!

https://scottmauer.wordpress.com/2018/01/23/adventures-in-hot-metal-part-1/
 
The original problem was fixed with this technique, but it didn't work on the pencil-sized hole I blew in the pipe. I did some pressure testing at less than 30 PSI and it was obvious.

I went with SSF-6 initially because diesel exhaust can get up to 1200 o F, and normal silver solder melts much lower. There may well be other high-temp silver allows I'm not aware of. I went back and took off the wire today, cleaned everything up, and made a stainless collar out of hose clamp to cover the hole instead. I was hoping it would then be just like soldering a copper pipe join. It sort-of worked, but the gap was too big, there wasn't enough flux inside the join, etc., so it just fell off again. I went back and did it again with flux coated bronze, and *now* it's pressure tight. I didn't realize that bronze was probably a better choice to begin with. Not to mention literally 30 times cheaper!

The good news is that I probably don't have to trust my own horrible repair. The newer pipe in the van just has a hairline leak right at the flange, which *is* very amenable to the SSF-6. My plan is to take it out, seal it up now that I'm less incompetent, and put it back, and only keep the older pipe as a backup. Wish me luck!

I have details for all this, but they won't post until tomorrow. Will post the link here when it goes live.
 
So in the process of trying the "wrap the hole in stainless steel wire" technique, and the "make a stainless steel collar out of a hose clamp" technique, I not only didn't fix the original problem, but managed to completely destroy the pipe.

Drumroll, please :rad::



Luckily the newer pipe still on the van, while barely hanging on by a thread:



(note that the crack goes *all* the way around the pipe...)

...was much more amenable to fixing with the Muggyweld SSF-6 and a propane torch instead of the oxyacetylene that had gotten me in so much trouble before.



I clamped everything into place to make sure the gaps would be as hairline as possible, then started above the crack and chased the SSF-6 down to the flange join. The final result looks really solid, no light, no extra flux caking things up, and it passes a 25 PSI pressure test.

I put the pipe back on the van, and for the moment it's holding. However, I think it will break again, and here's why.

Originally I though that I had stripped one of the bottom screws going into the EGR cooler the first time I did the repair. We'll remember that this is the second time I've fixed this pipe since owning the van. I didn't understand that when using thread grease one had to completely re-calculate recommended torques, and one of the screws stripped upon re-installation into the soft aluminum of the EGR cooler. My solution, since the threading is much deeper than the screw was long, was to find a longer screw and put in 5 exhaust-grade lock washers to take up the slack. Naturally I was supernaturally careful the second time with the longer screw not to overtighten. However, the second pipe failed in the same way, and I think I know why. I think the EGR cooler screw was stripped by whoever installed the last pipe before I got the vehicle. Because that side wasn't bolted down as well, it would probably flex a little when the pipe got hot. The other side was tighter, so it stayed in place. Over time, this differential stress eventually caused the pipe to spring a hole in the far side, as well as start to pull away from the flange.

Once I did the second repair, I had a verified-stripped hole. I compensated with the longer screw that went past the bad threads with lock washers on it. Even if this was as tight as the other side (which it probably wasn't because of my paranoia about stripping it more), it's my belief that the lock washers probably provide enough extra thermal expansion to cause exactly the same problem to happen again. Also, when I installed the second pipe, again being super careful about the stripped hole, I think it may have stripped *again*.

Moral of the story, I think that hole was stripped in the first place, and the lack of solid contact with that side of the flange caused both pipes to fail. What I *should* have done this time was drill out not one but *both* EGR cooler threads and install a heliocoil or similar thread replacement, then gone back to two identical factory torx bolts, using a torque wrench to guarantee identical tightening. Unfortunately there comes a point in every repair where getting it done become a higher priority than getting it done right, and I was already way past that point by the time I got the second pipe fixed. I also categorically refuse to remove the EGR cooler to re-do the threads, meaning it will all have to be done over my head.

If and when the new pipe starts to leak, I will re-tap both holes, or execute my Other Crazy Scheme to more permanently eliminate the problem. Since I have the GDE tune, my EGR valve never opens, so all of this plumbing exists for no other reason that to provide leaks, making all this work even more frustrating and time-wasting.
 

Bobnoxious

Deplorable and adorable.
Don't be too hard on yourself, I give you kudos for trying. Doesn't look like flux was use when brazed? As Dennis mentioned prior, there must be preload on the component. How much prying is necessary to align the mounting holes?
 
Don't be too hard on yourself, I give you kudos for trying. Doesn't look like flux was use when brazed? As Dennis mentioned prior, there must be preload on the component. How much prying is necessary to align the mounting holes?
Thanks, Bob.

Both the bronze brazing rods and the SSF-6 are flux-coated rods. One heats up the work piece first, dissolves the flux over the area to join, then follows up with the metal. I didn't understand that the bronze flux was so glassy. I actually managed to get a positive pressure test on both the wire wrap and hose clamp repairs, but in both cases the gaps were mostly filled with flux, not metal. Not only am I very inexperienced at all this, but it's actually very difficult to get this right without TIG welding. The pipe is so thin that even when it's dull red, it still doesn't hold enough heat to melt the bronze rod on contact. You poke it gently with the rod, and that part of the stainless turns black. The rod remains rock solid. This means that one has to apply the flux, keep the workpiece at the sweet spot between melting the bronze and melting or splitting the stainless, *and* melt the bronze rod with the torch all at the same time. That's certainly way beyond my pay grade three days after firing up an oxyacetylene torch for the first time.

I'm still not exactly sure how the flux on the bronze rod is supposed to work. It dries to a glassy finish, and clearly causes problems if it's still present after the brazing is done. Is it supposed to be fully burned off first? I know it's supposed to clean and prep the join surface first, but how it does than when it turns into glass is beyond me.

The SSF-6 was a little simpler because it flows more easily than the bronze at 1150 o F. It's *way* more expensive ($25-30 per rod), but by the time I got to the second pipe, I had it figured out enough to get a good seal. I had previously thought my propane plumbing torch wasn't hot enough. Turns out it is *and* it's much harder to accidentally destroy the underlying stainless with it.

With both pipes, there was never any torsion on them during installation. They fit perfectly in place as everything was gently tightened down. That's one reason I think the bad threading is the source of this recurring problem.

An additional wrinkle is that both of my temp sensors in that area read over 500 o F when the engine is cold and off. The EGR cooler one, and the upstream from the cat one I think? I haven't figured out how to verify the voltage ranges to see if they're actually fubared or if my iCarSoft II is maybe not right for those. All the other temp sensors read correctly. I disconnected the EGR cooler temp sensor, and it then reads 1100 o F and throws a code. I wonder if maybe those sensors are wrong and that's somehow making something else burn too hot?

However, after a 20 minute drive in 50 o F temps, an IR thermometer good to 1200 o F pointed at the bottom of the pipe where it joins the EGR cooler was only 150 o F with the engine still running. I was surprised it was that low. The whole reason I used the very expensive SSF-6 instead of normal silver solder is that the Interwebz says diesel exhaust can get up to 1200 o F. Normal silver solder is only good to 400-500 o F. Gonna try more readings later. It's important to keep in mind, though, that while there's still pressure in that pipe, there's no flow. It's blocked at the EGR end by the valve always being closed. I would guess that means it won't get as hot, though there might be more *differential* heat between the hot end at the exhaust side and the cool end at the EGR valve.

Also, a simple question. Is there supposed to be some kind of splash guard to protect the EGR pipe from water? It seems like hitting a hot exhaust pipe with cold water is asking for trouble. I certainly don't have such a shield on my van.
 

NMccann92

New member
Usually EGTs (exhaust gas temp.) is read at the exhaust manifold, because thats where it is going to be the hottest. Now the EGR crossover is getting those exhaust gasses obviously. However it may cool down a little by the time it gets to the bottom of the tube. You will probably get a hotter reading at the top. Also when you read the temperature the truck probably had some time to idle and cool down. 1200 degree EGTs would be WOT conditions with excessive fuel being dumped in, or atleast in my expierience. Im not exactly too familiar with Sprinters average numbers.
 
Usually EGTs (exhaust gas temp.) is read at the exhaust manifold, because thats where it is going to be the hottest. Now the EGR crossover is getting those exhaust gasses obviously. However it may cool down a little by the time it gets to the bottom of the tube. You will probably get a hotter reading at the top. Also when you read the temperature the truck probably had some time to idle and cool down. 1200 degree EGTs would be WOT conditions with excessive fuel being dumped in, or atleast in my expierience. Im not exactly too familiar with Sprinters average numbers.
Thanks for the info. I'm mainly concerned about the temp at the lower end because that's where my good-to-1150 o F SSF-6 repair is. Since the GDE tune keeps my EGR valve closed, no exhaust ever flows through the pipe unless it's leaking. The only heat transfer is from conduction along the pipe. Checking the exhaust end would definitely be useful and interesting. A better worst-case scenario would be after hard driving on a hot day. Knowing those numbers would help other people considering repairing their own pipe.
 

Bobnoxious

Deplorable and adorable.
SMILLER is the go to Guru with respect to GDE tune and emission components. Very knowledgeable. I believe the only time you should see any temperatures near 1200° is during post-injections/DPF regenerations.
 

mcles

New member
I have a 2007 Dodge Sprinter that has been breaking the EGR crossover tube that branches from the exhaust manifold Y-pipe in the rear of the engine that goes to the EGR cooler. It has gone through 3 tubes now and always either seperates from the bottom flange closest to the egr cooler or cracks at the bottom of the tube in the same area. Anyone ever had an issue like this? I have the top bracket bolted theres another clamp but I cant see where it would go to. I have started working on this truck about a year ago and am not sure if the people working on it before removed the bracket and didnt put it back on, or I possibly may have when I replaced the turbo. Thank you.
I'm having the same issue. I suspect the issue may be different from what has been passed around and I hope for some feedback.
I first notice my van going into limp-home mode, then it clears up, back and forth a couple times then the tube cuts loose as described.
I suspect blockage of the exhaust system.
I haven't been running the van long enough to get it properly warmed up so I suspect the DPF is getting plugged, creating excessive back pressure and that is causing the failure.
Could this be true or am I full of BS.
This tube is a real PITA to replace so I would really like to get this issue sorted out.
 

Top Bottom