PDA

View Full Version : OM647 Compression test quick question


doug022984
02-29-2016, 09:21 PM
In an effort to rid myself of some glow plug MIL codes, I'm getting ready to change out some glow plugs. While I have them out, I was planning to do a manual compression test since I recently acquired my sprinter with 223K miles and no record or maintenance or anything for engine health other than it runs. I don't have a DRBIII or other means of doing it electronically, so that's why I'm going to do it manually while changing the glow plugs.

I have the service manual and the process seems very simple (paraphrasing: get engine to temperature, pull plugs, install tool, crank engine with starter, and record results for each cylinder). My question is in my experience with doing compression tests on gas engines, disabling the fuel system or injectors is always an important part of the compression test. The sprinter manual makes no mention of disabling the injectors or fuel system while performing the compression test. Is there a reason this doesn't have to be done on diesels or sprinters specifically? Should I do it anyway to be safe? If so, I'm guessing that disconnecting the camshaft position sensor would be the easiest means of disabling the injectors?

autostaretx
02-29-2016, 11:04 PM
Remember that a Sprinter has a nominal compression ratio of 18 to one.
That's almost 300 psi, and a good engine can be a lot higher.

Choose your gauge, hoses and screw-in head attachment to handle that pressure.
-------

The fuel injected per revolution at 993 rpm in my OM647 is 0.000732 cubic inches per squirt in neutral.
Yes, 7 ten-thousandths of a cu inch.
12 cubic millimeters (1 mm-3 = one one-thousandth of a cc, so 12 thousandths of a cc (or milliliter))

I would think that disabling the camshaft sensor would kill the squirts, but i haven't tried to find out.
(since the "timing" of the engine is *when* the squirts happen, it darn well should kill them)

--dick

lindenengineering
03-01-2016, 01:04 AM
Better to disconnect the crank sensor.
Read C/R for a turbo engine is 400/425-psi
Minimum C/R is 325.
Allowed 10% variation between cylinders and 25% across the whole 5 pots.

Any C/R at/or below 300 and you will get bad starting especially on transition between glow plug operation and a partially warm engine. Gobs of smoke emit once it catches!
Hope this is of use.
Dennis

MillionMileSprinter
03-01-2016, 02:08 AM
Dennis suggests disconnecting the crank sensor, but I think he may mean the camshaft sensor. Both will keep the engine from starting, but the crankshaft sensor is located under the van, while the camshaft sensor is located right there in the head, to the left (passenger side in the US) of injector #5. I've always had the injector cover off when I did glow plugs, so it's very easy to get to .

doug022984
03-01-2016, 02:51 PM
Thanks everyone for the prompt responses and guidance. Sounds like I'll pull the cam sensor due to its easier access. Hopefully the glow plugs come out without being too stubborn.

lindenengineering
03-01-2016, 03:21 PM
Thanks everyone for the prompt responses and guidance. Sounds like I'll pull the cam sensor due to its easier access. Hopefully the glow plugs come out without being too stubborn.

Oh you wish!:bounce:

In similarity not not quite the same if you get my drift!
The Ford 5.4 Triton engine has some funky style spark plugs that break off in the head!
Renown for it and the subject of a class action law suit !----Sound familiar.
The technique to getting those nasty buggers out is some engine heat, kroil, a deep special guide socket AND a 3/8th/1/2" airgun no less!
The other day I had a similar bash experiment with a 1/4 drive air gun and some glows in the 647. Spun out with no breakage!
Maybe the windy gun and scaring them out its the real answer!--If you can stomach it!
Best of luck
Dennis
ps
For those interested in Ford--A Triton spark plug:-
http://www.summitracing.com/parts/mof-sp515?seid=srese1&gclid=CI7W8erkn8sCFVc1aQodRWAAfw

doug022984
03-07-2016, 04:23 PM
This project quickly became more complex and postponed. #1 plug came out so easy, it was like it wasn't even torqued. #2 plug did about 1 full turn then started to tighten up like it was galling, so I stopped. #3 plug was the best: Upon pulling the connector off, a whole rusted piece of electrode or whatever the internal part of the plug is came out with it. This was clearly the plug causing my engine light to be on, but what was worse: no hex present to even put a wrench on. Evidently the previous owner tried to remove this plug and broke it off. I just stopped there and put #1 back in, tightened #2 back up. Will revisit this when the engine is out in another month.

Can anyone advise what special tools I may need for extracting broken glow plugs? I plan to do it after the engine is out and cylinder head removed. I have access to a machine shop with drills and mills.

lindenengineering
03-07-2016, 05:39 PM
With the head off, seek a machine shop with spark erosion machine to erode them out!
That's what I/We use on these 5 banger engines.

On the Vee 6 we use a different technique with aircraft precison drills, short jobber lengths won't do!------She said!~:laughing:
Dennis

doug022984
08-20-2017, 06:22 PM
Piggy backing off my own thread since the subject line still applies. Anyone have recommendations on good adapters to screw in to the glow plug ports for doing a compression test? The harbor freight set sits so low that you can't get the guage quick connect on. I bought another "sprinter specific" adapter that was a nice hose extension setup, but it had the wrong threads. I can't seem to find the manual's recommended special tool #8927 for sale anywhere, so I'm hoping others might know of alternatives. My guage has the 1/8" quick connect like the harbor freight kit does, but it's a better quality one that shouldn't break after 1 use.

RileyC
08-24-2017, 01:21 PM
I have a Mityvac analog diesel compression tester that I ended up switching the end of the quick connect from the Harbour Freight diesel compression test kit onto. That way you can use the fitting from that kit that works with our vans. I had to use some long curved needle nose pliers to attach and release it because of the tight clearance in the glow plug cavities. It seemed to work well.

kite
02-17-2018, 06:02 PM
I'm going to do a compression test today. I built up a ridiculous stackup of air fitting adapters to connect a mityvac glowplug adapter to a harbor freight gauge. An added benefit is it lifts the quick connect above the fuel lines, so will hopefully be easier to connect and disconnect the gauge. Hopefully no leaks...

Quick question - to disable the fuel system, is pulling the fuel pump fuse OK? I did it, the engine cranked but didn't start. This seems like the intended outcome, but I want to make sure I'm not missing something.

lindenengineering
02-17-2018, 07:57 PM
Yes you can pull the fuse!
Disconnect the cam or crank sensor.
What is at gauge set to????
Remember you will need at least 400 lbs for a diesel, else the needle will break off the tube.
Dennis

kite
02-18-2018, 03:51 AM
Thanks Dennis,

Yea, the gauge is a diesel compression tester that goes to 1000psi.

Since I wasn't able to get to this today, I'll ask a couple more questions.
1) to clarify from your last post. Should I remove the fuel pump fuse OR cam/crank sensor? Or do I need to remove a sensor AND fuse?
2) the harbor freight tester I got has a 10mm glow plug fitting, but I can't use this b/c I can't get the quick connect all the way down. The mityvac adapter I got is 10x1.0, and fits just fine. But it is different in that it has an O-ring near the end of the adapter. do I want this O-ring? I'm kind of worried it could slip off inside a cylinder.

The actual GPs just seal against the head, correct?

pics of each attached. the o-ring is in the red circle on the mityvac. you'll see the adapter stackup as well....

Aqua Puttana
02-18-2018, 11:19 AM
...

The actual GPs just seal against the head, correct?

...
Correct. The threads compress the seal face against the head.

I would compare the tool adapter length to a glow plug to be certain the reach is correct.

Some pics of the seal face are here.
https://sprinter-source.com/forum/showthread.php?t=63238

How will the O-ring seal unseat?

vic

kite
02-18-2018, 05:49 PM
Ah, thx for the pic. I thought that was the seal but good to know for sure. Clearly my adapter is way shorter than a GP. BUT it does have the o-ring. I guess I could just try it and see if it seals.

My concern with it slipping off it is isn't in a groove, it just sits on a smaller diameter nipple at the end of the adapter. If the o-ring is tight enough to seal in the bore, it might roll/slip off during removal...

Nautamaran
02-18-2018, 07:28 PM
The adapter needs to match the GP length so it can reach the conical seal at the end of the major bore, otherwise your compressed volume includes the empty GP bore hole and your readings will be artificially low. If your O-ring is sized to the major diameter of the plug stem I wouldn’t worry about losing it down into the cylinder, but if it is designed to seal the clearance around the glowing tip beyond the seal surface I agree - it could get stuck in there, and cranking seems as likely to suck it in as blow it clear?

-dave

autostaretx
02-18-2018, 08:19 PM
You can simply unplug the connector at the cam or crank sensor.

If the ECU cannot see either of those sensors, it will not attempt to squirt fuel into the cylinders (since it cannot determine which stroke is happening where).

--dick

kite
02-19-2018, 04:46 AM
The adapter needs to match the GP length so it can reach the conical seal at the end of the major bore, otherwise your compressed volume includes the empty GP bore hole and your readings will be artificially low.

Won't the volume of the gauge also affect the reading? and won't gauges vary?
A quick estimate of the amount of extra volume makes me think the GP bore volume will be in the noise of this measurement. Also, it will be consistent, so at least I'll find any outliers.
Estimate GP major diameter is 6mm
length missing = 40mm
volume = 4.5cm^3
vs 2700/5 = 540cm^3
so it's less than 1% of the volume of the cylinder.

If your O-ring is sized to the major diameter of the plug stem I wouldn’t worry about losing it down into the cylinder, but if it is designed to seal the clearance around the glowing tip beyond the seal surface I agree - it could get stuck in there, and cranking seems as likely to suck it in as blow it clear?

Good point. The O-ring is indeed sized to match the major diameter of the plug stem. So even if it comes off it'll be stuck above the seal surface. I guess then I'd have to do some fishing to get it out, b/c like you said, cranking to clear it sounds risky.

Aqua Puttana
02-19-2018, 10:46 AM
...
so it's less than 1% of the volume of the cylinder.

...
Isn't the compressed volume what you should be using for calculation?

Won't the straight, no taper non-sealing threads leak if the seal face doesn't contact?

I do agree that it will be an ok test for comparisons between all cylinders. The absolute data will be suspect.

vic

doug022984
02-19-2018, 04:26 PM
I ended up using a similar adapter that did NOT seal on the conical face down in the bore like the glow plug normally does. It sealed on the O-ring around the fitting like yours. I too believed this would give me slightly lower readings, but did the test anyway. I ended up with 375 to 400 psi between my 5 cylinders, which is acceptable per the manual. So even if the setup increases volume slightly, it wasn't enough to give the cylinders an unacceptable compression reading. I'd move forward with what you have.

One tip from my troubles even though I realize its compression testing 101. The O-ring is too large to find its way all the way down in to the cylinder bore, BUT make sure you depressurize your gauge before moving to the next cylinder. I was trying to move too quickly while the engine was still warm and pulled the quick connect fitting before depressurizing the gage and it blew the internal O-ring right out of the quick connect fitting. I found it sitting on the head right next to the GP Bore.

kite
02-19-2018, 04:51 PM
Isn't the compressed volume what you should be using for calculation?
vic
You're right, my calc isn't great. I guess the extra volume should be taken into account in the initial and final volume. I think it's just time to try it though.


Won't the straight, no taper non-sealing threads leak if the seal face doesn't contact?
vic
If the O-ring has a good seal, shouldn't this eliminate any pressure differential across the GP threads?

I ended up using a similar adapter that did NOT seal on the conical face down in the bore like the glow plug normally does. It sealed on the O-ring around the fitting like yours. I too believed this would give me slightly lower readings, but did the test anyway. I ended up with 375 to 400 psi between my 5 cylinders, which is acceptable per the manual. So even if the setup increases volume slightly, it wasn't enough to give the cylinders an unacceptable compression reading. I'd move forward with what you have.

Nice! The one I got looks like this (https://www.amazon.com/gp/product/B007QV59T8/ref=oh_aui_detailpage_o05_s00?ie=UTF8&psc=1).

Did you use the same one, or something like this (http://www.freedomracing.com/compression-test-adapter-9543.html?gclid=Cj0KCQiAiKrUBRD6ARIsADS2OLm1lOOcZG JPRi-cIK1-RGbvmP18ameBtP9lOv3ohKr1JZYWktEAsokaAup3EALw_wcB)? This one looks like a better design...

Thanks for the compression testing 101 - haven't done this before! Will report back...

doug022984
02-19-2018, 05:26 PM
This is the exact one I used. https://www.amazon.com/s/ref=nb_sb_noss/135-7523096-9694712?url=search-alias%3Dautomotive&field-keywords=S%26G+tool+35250

kite
02-19-2018, 05:33 PM
Interesting, it seals above the threads. Glad it worked, hope mine does too (or I'll be buying one like yours).

Nautamaran
02-19-2018, 07:24 PM
Won't the volume of the gauge also affect the reading? and won't gauges vary?
True. My apologies... I was clearly bogged in academic mode yesterday. I would expect guages to be made with small I.D. components with this in mind, or to be liquid filled?

But you’re right: I should have done the math on the short adapter before commenting...

2.7 litres = 2700 cm^3 engine displacement
18:1 compression ratio
2700 / 5cyl / 18 = 30 cm^3 chamber volume

6mm = 0.6 cm diameter glow plug bore (0.3cm radius)
40mm = 4.0 cm missing length
Pi*r*r*L = 1.13 cm^3 increase in chamber volume

1.13 / 30 * 100 = 3.7% chamber volume increase, so any gauge reading should be multiplied by 1.037

At 350 psi the gain is 13 psi... not quite noise, but not earth shattering either.
Put another way, you’ll loose about 2.5% compressed pressure per inch of unfilled GP bore length.
:cheers:

-dave

kite
02-20-2018, 02:08 AM
Pi*r*r*L = 1.13 cm^3 increase in chamber volume
-dave

DOH. I did pi*d^2 on my first pass...

your math looks pretty good.

Alright so all this talk and the whole deal takes only about 15 minutes. All cylinders were 250-260psi after 8 cranks (pressure had pretty much stopped climbing).
I think (and am hoping) that I've got a leak in the system somewhere. My guess is it's the o-ring on the adapter, b/c it really isn't tight as I push it down the glow plug bore. For a good seal you'd think it'd take some effort to push it down until the threads engage.

Or am I just being optimistic? Is it possible it's this low and still starts and runs fine?
I like to think that it's good they were all within 10psi, but it could just be that I've got a "blow off valve" at ~250psi...

Nautamaran
02-20-2018, 04:43 AM
You could lather it up with some grease and have another go?

From Post #3 above:
Better to disconnect the crank sensor.
Read C/R for a turbo engine is 400/425-psi
Minimum C/R is 325.
Allowed 10% variation between cylinders and 25% across the whole 5 pots.

Any C/R at/or below 300 and you will get bad starting especially on transition between glow plug operation and a partially warm engine. Gobs of smoke emit once it catches!
Hope this is of use.
Dennis

So if you've got tough starting (and I assume you must or we wouldn't be here?) then your 250-260 readings may be correct? These scale up 3.7% (10 psi) to 260-270 with glow plugs installed, which puts you squarely in Dennis' smoky start range.

You may wish to squirt a teaspoon of diesel fuel into each cylinder to wet the rings and dissolve a bit of carbon. It's cheating a bit, but could help diagnose the compression losses?

-dave

kite
02-20-2018, 05:31 PM
So I got curious how my compression was after I busted my air filter seal in baja over the holidays. Upon later inspection, the turbo showed significant erosion from dirt in the intake, intake hoses were coated, etc. I figured why not learn how to do this measurement. Of course, I hoped to learn that I hadn't done significant permanent damage. pic (https://sprinter-source.com/forum/showpost.php?p=619321&postcount=47)

Fortunately I'm NOT experiencing tough starting, in 50F weather, nor smokey starts. And with the knockdown for the GP bore and stupid tall adapter stackup I made, perhaps the readings are correct.

Thanks for the back and forth on this one. I still might try to lube the o-ring, or buy a better adapter. It'll have to wait a week though, on my way out of town this morning.

rodery
05-22-2018, 02:18 AM
So I bough the adapter to do the compression test on my van from amazon and this part does not fit. I have a 2010 sprinter with 270000 miles. I believe the amount of oil I am loosing is due to a valve stem seal leaking as appose to a bad cycliser ring but wanted to do a compression test. What adapter is available for me?

lindenengineering
05-22-2018, 03:22 AM
Ford F250 with 7,3 diesel kit will do! (1995 to 2000)
375 to 400 psi optimum!
Anything below 320 a concern.
Dennis

WabiWanderer
09-16-2018, 08:58 PM
Looking to do this test through the glow plugs on my 2004 Sprinter 3500 SHC with OM647 engine this week. Thank you for the information in this thread!

Can someone please confirm these two items would suffice? S & G Adapter for 7.3 Ford Diesel with S & G Gauge

https://www.amazon.com/Tool-Aid-35250-Diesel-Adapter/dp/B000HHFDDU?keywords=S%26G+tool+35250&qid=1537125396&s=Automotive&sr=1-1-catcorr&ref=sr_1_1

https://www.amazon.com/Tool-Aid-34700-Diesel-Compression/dp/B000HI5AAU/ref=pd_bxgy_263_2?_encoding=UTF8&pd_rd_i=B000HI5AAU&pd_rd_r=23fca847-b9e5-11e8-b3e1-116141892cee&pd_rd_w=Qa2EW&pd_rd_wg=HNlIi&pf_rd_i=desktop-dp-sims&pf_rd_m=ATVPDKIKX0DER&pf_rd_p=3f9889ac-6c45-46e8-b515-3af650557207&pf_rd_r=0ZW6HMFDB7MX0YBN93X2&pf_rd_s=desktop-dp-sims&pf_rd_t=40701&psc=1&refRID=0ZW6HMFDB7MX0YBN93X2

lindenengineering
09-16-2018, 10:23 PM
Yes!
Dennis

owner
09-17-2018, 08:05 AM
You can simply unplug the connector at the cam or crank sensor.

If the ECU cannot see either of those sensors, it will not attempt to squirt fuel into the cylinders (since it cannot determine which stroke is happening where).

--dick
This isn't true for ROW sprinters at least. The ECU can gain 360 sync just from the crank signal pulse train. It can do this very quickly depending on when the crank tooth gap (or other unique pattern) is detected.

Once it has 360 sync, the abscense of the cam signal (which happens (or doesnt happen) at a predefined windiw of 360 crank angle once every 720 deg) tells the ECU which 360 phase it is in. So the ECU doesnt wait for an extra 360 of crank just to see the cam, which it already knows has to happen that next cycle.

So basically it short circuits that process and will squirt fuel into those cylinders that are in the 360 phase that always have no cam signal.

With cam disconnected this has a 50% chance of firing the correct cylinders and causing ignition events. And the other 50% of the time it will inject into the exhaust stroke.

Then on the next 360, when the ECU knows for sure it must see the cam in that same window of crank angle, but again it doesn't, then and only then will it stop trying to fire cylinders.

So with cam disconnected you have a 50% chance of firing up to 2 full ignition events every time you crank her over. Im not sure if this applies to NAFTA, but I know for sure it applies to ROW.

An interesting feature is if you disconnect the cam sensor with the engine running, it will keep running in a fairly decent silent limp mode. Low turbo boost and trans short-shifting. And if you re-connect during that same run it will cancel limp mode and resume proper running (it will store a code though). I toyed with the idea of rigging this up to a switch, which could double as a security feature preventing starting, and an "economy" mode.

Aqua Puttana
09-17-2018, 02:57 PM
This isn't true for ROW sprinters at least. ...

:idunno:
It may relate to the different emissions systems found on the NAS aka NAFTA Sprinters eg. - CARB 50 State.

...
An interesting feature is if you disconnect the cam sensor with the engine running, it will keep running in a fairly decent silent limp mode. Low turbo boost and trans short-shifting. And if you re-connect during that same run it will cancel limp mode and resume proper running (it will store a code though). I toyed with the idea of rigging this up to a switch, which could double as a security feature preventing starting, and an "economy" mode.
The NAS aka NAFTA Sprinters also have a power reduction/rpm limit for loss of Camshaft Position Sensor CMP. It is subtle enough that an operator may not even notice the limiting.

The CMP signal loss may trigger a power reduction and not a complete loss of turbo LHM.

...

NOTE:
"If there are faults in the boost pressure control system, the full load injected quantity is also reduced [by the ECM]."

So when there is a boost control related problem (hose leak, sensor problem, turbo, etc.) the RPM's aren't limited, BUT the full load is. That explains "can't go up hills", "no power, but on the flat I can get to 70 mph". A different form of LHM, but it is still an LHM power reduction set by the ECM.
...
vic
As always clicking on the Blue Arrow Icon within any quote box will take you to the original post/thread.

vic

lindenengineering
09-17-2018, 07:17 PM
I will follow up with this post.
Yes you can do a scan compression etc and it does give you what I will call an indication plus a rough running etc! In most cases that is all that is needed.

But me being me, being old fashioned, I always follow it up with an actual pressure test with a gauge in the hole.

This policy can be best described by a quotation posed to me by the local vicar in the UK village where my recently deceased mother's ashes were to be interred.

How do they do cremations and services for the deceased in America?
Essentially I answered cremation first, then the receipt of the cremated remains, then memorial service--To which he exclaimed "Well done America that is how its supposed to be done, good show"!

Then followed it up with "We Brits are such a distrusting bunch of bastards that we must see our loved one go into the fire and have certified remains before we will accept them"!

So with that colorful analogy I do the same & get physical verification physically!
Dennis