[GHewgill]

It is not a spur gear, the pump is driven by a coupling sleeve over the splines.

There might be a rubber O ring on the axle of the injection pump that needs to be replaced and a new gasket is also advisable to prevent oil leakage.

The shaft of the pump needs to be free of oil and grease.

Take the pump out on 24 degrees BTDC and look if the marks on the pump line up when you take it out. That gives you an indication of whether timing of the pump was off the mark.

Also cover the open pipe connections of the pump with same plastic cling film (or special plastic caps) to prevent dirt entering the pump.

Thanks for the tips. I pulled the pump out and to my surprise the splines on the coupling bolted to the injection pump are internal, not external as your diagram and my book have it. I could not make out any similar markings or gap in the splines as your diagram shows. I simply put the pump back in but a little lower than I removed it. I then put things back and checked the onset of injection timing. I managed to get it right around 24 deg BTDC.
The bad news is that after putting everything back together it still smokes - a lot; marginally less than it did before and it does have more pep, but it still smokes. AGGRAVATION!!!!!!!!!!
Reflecting on the situation, I bet the fuel timing has been wrong for a long time. The same guy who elected not to refill the oil in the air filter probably was responsible for the bad timing. I keep "fixing things" that are wrong but are not the source of the problem.

 

[GHewgill]

Maybe I need to go back to thinking about the injectors. I don't have a neat hydraulic pump with a 20 bar gauge on it, so to test the injectors I simply attached them one at a time to the # 1 nozzle and connected a remote starting switch to the starter and observed the spray pattern, force of spray etc; all highly subjective I know, but I would recognize an odd spray pattern or a clogged injector or a leaky one. THey all delivered a small but forceful stream straight down. I caught the fluid in a dish and they all delivered enough fuel to aerosolize a little bit of diesel fuel as the stream struck the dish. I also tested three injectors I removed from a 240D at a junk yard in the same way. They behaved similarly. Am I kidding myself here? Should I buy new injectors or get a hydraulic tester? Any other thoughts of what could be at fault here that is slowly getting worse all the time? After each service I've performed I get marginal improvement but the trajectory of performance is downward.
Thanks,
Greg

 

[Govert70227]

Thanks for the tips. I pulled the pump out and to my surprise the splines on the coupling bolted to the injection pump are internal, not external as your diagram and my book have it. I could not make out any similar markings or gap in the splines as your diagram shows. I simply put the pump back in but a little lower than I removed it. I then put things back and checked the onset of injection timing. I managed to get it right around 24 deg BTDC.

There is a coupling sleeve over the splines, that connects the pump with the intermediate shaft. The sleeve can be pulled off. The sleeve has the splines on the inside of course.

Maybe I need to go back to thinking about the injectors. I don't have a neat hydraulic pump with a 20 bar gauge on it, so to test the injectors I simply attached them one at a time to the # 1 nozzle and connected a remote starting switch to the starter and observed the spray pattern, force of spray etc; all highly subjective I know, but I would recognize an odd spray pattern or a clogged injector or a leaky one. THey all delivered a small but forceful stream straight down. I caught the fluid in a dish and they all delivered enough fuel to aerosolize a little bit of diesel fuel as the stream struck the dish. I also tested three injectors I removed from a 240D at a junk yard in the same way. They behaved similarly. Am I kidding myself here? Should I buy new injectors or get a hydraulic tester? Any other thoughts of what could be at fault here that is slowly getting worse all the time? After each service I've performed I get marginal improvement but the trajectory of performance is downward.
Thanks,
Greg

I hope you mean a 200 bar gauge, because at 20 bar the injectors should not be opening. Normal opening pressure is about 115 bar, lower limit is 100 bar. The spray pattern should be a small closed cone, not a single stream, although some injectors have a pre-injection stream. The only way to properly test the injectors is with a hand pump. There are two ways: either make or buy a hand pump and test the injectors yourself. With the aid of the hand pump you can also replace the nozzles in the injectors, although you need shims to adjust the opening pressure of the injectors. The other way is bring the injectors to a diesel workshop, have them test and replace (if necessary) the nozzles. It is about an hour's work for them.

You haven't addressed the leak of the pneumatic governor. If the car is idling normally (700-800 rpm) than the throttle was screwed back to produce enough vacuum so that the engine doesn't idle too high. That will increase smoke (and fuel consumption).

Smoke out of diesel engines is (if it isn't coolant because of a leaky head gasket) is mostly unburnt diesel. Diesel will not burn because of lack of compression, not enough air, too much diesel injected or injected at the wrong time.

I would probably do a compression test and a compression leak down test to see whether the pistons, cylinders and valves are working properly. It could be that the engine is worn out. The engine has to be warm for these tests.

Diagnosing a problem isn't always easy, certainly not on an old car with neglected maintenance, but take it one step at the time, easiest/cheapest things first, more complicated/expensive things second.

 

[GHewgill]

Yes, I meant 200 bar not 20. An order of magnitude issue, probably because I really think in pounds per square inch, not bar or Pascal.

That makes sense about the splines. Silly me, I read your post when you mentioned the collar, but I was still expecting to see a male and female part rather than two external splines with a collar joining them. Oh well, despite my misapprehesion this adjustment seemed to go OK.

I will fiddle with the idle at lunch hour. Some weeks ago when this problem began to appear I had raised the idle since I found it smoked less at a higher idle, but had rather optamistically turned it back this weekend thinking that I wanted to reset everything to "normal."

Yes, there is the leaky pneumatic governor situation, but I found that when I put the viscous oil on the poppet shaft it did help it hold vacuum, yet made no perceptible difference in how the engine ran. This inclined me to believe that this is really not a problem, even if it could be better.

The flow from the injectors is conical but very tight, to the point that it appears to be a straight stream, isn't that right? This is how it appears to me on the video I've seen anyway (and how it looked when I connected the injectors). If it was indeed coming out as a solid stream I can't imagine that it would be straight down; wouldn't it be directed to one side of the other since the nozzle opening creates this cone when the tip rises out of the whole in the nozzle when the pre-set pressure has been achieved? If it were a solid stream that would mean it was only coming out on one side of the nozzle or the other.

Yes, I do realize that I'm rather handicapped because I don't have proper test equipment, so I'm relying on observation and inference (and I'm too cheap to take it in to the shop .

Thanks again for the advice; I will persevere.
Greg

 

[Mike D]

You are shooting in the dark. Do the compression test before you go any further. Anything below 300 p.s.i. is suspect.

 

[GHewgill]

You are shooting in the dark. Do the compression test before you go any further. Anything below 300 p.s.i. is suspect.

Thanks Mike, yes I was thinking last night that I need to get the tool (I have a compression tester for gasoline engines but doesn't fit diesels) and do that to rule it out. Low compression was not my first thoguht because it starts so easily, but there's nothing like some hard data.
Greg

 

[GHewgill]

Thanks Mike, yes I was thinking last night that I need to get the tool (I have a compression tester for gasoline engines but doesn't fit diesels) and do that to rule it out. Low compression was not my first thoguht because it starts so easily, but there's nothing like some hard data.
Greg

Hey, Im quoting myself! OK, bad joke. I bought the compression tester and did the test. All four cylinders are at 360 psig (26 bar).

Further suggestions?
Thanks,
Greg

 

[Mike D]

Ok. Now you know; compression is good, valves are not burnt, head gasket is not blown and it is not an internal engine mechanical problem.

Next step. What did you determine the pump timing to be? You must check the pump by either drip or rise method at the #1 compression stroke. I'm guessing you are but make sure you are on the compression cycle. Pull the valve cover and check. (turn the crank until the #1 intake valve opens then continue turning to 24-30 degrees BTDC).

 

[Govert70227]

I will fiddle with the idle at lunch hour. Some weeks ago when this problem began to appear I had raised the idle since I found it smoked less at a higher idle, but had rather optamistically turned it back this weekend thinking that I wanted to reset everything to "normal."

Yes, there is the leaky pneumatic governor situation, but I found that when I put the viscous oil on the poppet shaft it did help it hold vacuum, yet made no perceptible difference in how the engine ran. This inclined me to believe that this is really not a problem, even if it could be better.

You've established with a vacuum gauge that the governor doesn't hold vacuum, that is a problem. That will cause smoke and diesel not being burnt because of lack of air.

What happens is the following. Under normal conditions you have to open the air valve half way and than the engine runs at let's say 3000 rpm. If the governor cannot hold vacuum, you only need to open the air valve for a little bit for the engine to run at 3000 rpm. The problem is that the air valve doesn't let enough air through to the cylinders, but the amount of diesel injected is still the same.

Good to hear that the compression is in order. Did you check with the air valve completely open?

If you have checked through the injector openings, don't forget to use new heat shields under the injectors.

 

[GHewgill]

You've established with a vacuum gauge that the governor doesn't hold vacuum, that is a problem. That will cause smoke and diesel not being burnt because of lack of air.

What happens is the following. Under normal conditions you have to open the air valve half way and than the engine runs at let's say 3000 rpm. If the governor cannot hold vacuum, you only need to open the air valve for a little bit for the engine to run at 3000 rpm. The problem is that the air valve doesn't let enough air through to the cylinders, but the amount of diesel injected is still the same.

Good to hear that the compression is in order. Did you check with the air valve completely open?

If you have checked through the injector openings, don't forget to use new heat shields under the injectors.

Thanks for the post Govert. OK, I'll see if I can get the cover part with the poppet valve and all the seals. I acknowledge that it leaks and therefore isn't what it should be, but I'm a little skeptical about that being the source of the problem for a couple of reasons: 1) when I put the viscous oil on the poppet shaft it definitely holds vacuum better, but the engine runs the same - no change in idle and no reduction in smoke, 2) My experience with "adjusting the vacuum" in an attempt to determine whether it is over-fueling or not. Let me explain. I brought a small lab vacuum pump from my work (I'm a chemist) and connected it to the pneumatic governor port on the injection pump. My thinking was to "dial-in" the right vacuum for a reasonable idle and then open the air valve on the throttle body and see if supplying more air to the engine would reduce smoke. Well, my experiment wasn't very successful. Even though this vacuum pump has an air bleed valve, I found it very difficult, actually impossible, to get the engine to run at idle speed. It would either race or stall. Well, race is an exageration, but way above idle anyway. If I increased the vacuum marginally from a normal running speed (I'm guessing around 3,000 RPM, but I have no tachometer so it's only a guess), it would slowly slow in speed and then stall. I'm still not sure why this should be, but perhaps the system requires the natural feed-back that occurs between the engine vacuum and the delivery of fuel. Maybe de-coupling these two isn't possible. So, even though my "experiment" wasn't successful, I think I learned a couple of things. One is that the engine speed is determined by the injection pump - it will pull more air through the throttle body even if it is restricted. When it did this there seemed to be less smoke, not more, even though this is certainly an over-fueling situation. If the leaking seals on the pneumatic governor were consequential, why isn't the idle unacceptably high (as in my vacuum pump situation)?

Perhaps I'm mudying the waters here, and your explanation of the fuel/air ratio makes perfect sense, but then I go back to my point 1) above. Barring some revelation I will replace the part, but it seems like I might be fixing yet another thing that really won't solve the problem.
Thanks again,
Greg

 

[R.P. McMurphy]

...i just want to say that both of you guys are an asset to this place, the exchange of knowledge between you is most respectable and invaluable. :bowdown:

 

[Govert70227]

The diesel engine has a tendency to run out of control even with a limited amount of diesel injected, certainly if there is no load on it. If that happens, air flow increases, vacuum increases and less diesel is injected. It is a feedback loop. Feeding a constant vacuum to the governor doesn't work.

That is also why there was less smoke at 3000 rpm. Only a limited amount of diesel was injected, but the speed of the engine wasn't controlled by the governor. At 3000 rpm more air is drawn in, but the amount of diesel was still low.

The oil seal on the poppet cam shaft is temporary, and sometimes it will take a while to make the seal. The puppet cam shaft isn't the only place where it can leak, the diaphragm can leak too.

 

[GHewgill]

The diesel engine has a tendency to run out of control even with a limited amount of diesel injected, certainly if there is no load on it. If that happens, air flow increases, vacuum increases and less diesel is injected. It is a feedback loop. Feeding a constant vacuum to the governor doesn't work.

That is also why there was less smoke at 3000 rpm. Only a limited amount of diesel was injected, but the speed of the engine wasn't controlled by the governor. At 3000 rpm more air is drawn in, but the amount of diesel was still low.

The oil seal on the poppet cam shaft is temporary, and sometimes it will take a while to make the seal. The puppet cam shaft isn't the only place where it can leak, the diaphragm can leak too.

Hi Govert,
OK, that makes a lot of sense to me. So, what would you do, replace the cover with all its parts and then see if things are OK and if not have the injector pump diaphram replaced? I presume that is not something I could likely do myself, but would have to have done professionally.

One thing I did notice with both a hand vacuum pump and my little lab vacuum pump is that I couldn't get a good vacuum on it even with dousing the poppet cam shaft seal with oil. The fuel shut-off lever test that you described worked fine with the oil, but actually pulling a vacuum didn't go above 10 inches of mercury (250 torr or 330 mbar), If my recollection is correct. I didn't keep the vacuum on it that long not knowing if I could do harm to the unit if it never experiences high vacuum. Would you expect the vacuum to keep rising? Do you think this indicates a leak elsewhere (like the diaphragm?) or have I simply overcome the film strength of the oil on the shaft?
Thanks,
Greg

 

[Govert70227]

Or does it? The piston speed increases when the engine speed increases, but does that mean it can suck in more air through the closed air valve?

I recently worked on a OM616 engine with a governor with no leaks at all. I did the vacuum test: removed air hose, pushed lever in stop position, connected vacuum gauge, released lever. The gauge showed slightly below 100 mbar of vacuum and that is what you should expect. The FSM says 70-90 mbar, see:
07.1 Diesel Injection System - OM616
and select 07-125 over there.
I can make pictures or a movie this saturday of that governor, for comparison.

If you connect a vacuum pump to the governor, and if there are no leaks, the vacuum can go as high as the pump can provide (usually around 800 mbar of vacuum). It will not damage the governor, you are doing the same thing as putting the engine into the stop position.

According to my information you cannot get the engine to run well with a leaky governor housing. A slow leak is acceptable, but we are talking half a minute/minute here.

You can replace the diaphragm yourself, it is very well described in the FSM. You do need a micrometer and you have to watch out for small parts.

(That is my hand)

Fixing the diaphragma to the control rod is the most difficult, but do-able.

 

[GHewgill]

Hi Govert,
I assume your "or does it" refers to whether the engine can draw in more air? From what I can tell these engines's butterfly valves do not limit air flow nearly so much as in a typical gasoline engine. For one thing they still draw air through an ostensibly closed butterfly even though there seems to be no bypass - you can feel the air flow around the edges of the disc. Also the pneumatic line from the injector pump is attached to a channel in the throttle body which is open to both sides of the butterfly valve which has a notch cut into to it to accomodate this channel. I think a lot of air must be able to pass through this channel. This also suggests to me that the vacuum remains relatively low in this line. Besides all this the shear volume of air from the tail pipe of an idling diesel is amazing - even these engines with throttle bodies.

Can you tell me what the part number is for diaphragm? and do I need to pull the pump out (please say no )

Thanks,
Greg

 

[Govert70227]

I was doubting my own words whether more air per stroke is sucked in with increased engine speed.

The channel in the throttle body is the venturi, that produces the vacuum. If you open the air valve the venturi is bypassed, so the vacuum decreases.

The pump can stay on the engine to replace the diaphragm, but there is not much space to work in a W115, but it is possible.

I held my hand for the air intake of a idling W115-200D, the amount of air sucked in was considerably less than with my W123-200D, which does not have a air valve.

I have to look up that part number, MB sells them, but other manufacturers too, sometimes at lower prices.

 

[GHewgill]

Thanks Govert, I'd appreciate that very much. So far I'm not impressed with MB dealerships parts departments, but I may be premature in my judgment. Also I've tried searching the internet for the part number of the cover and so far no luck. Do you have internet suppliers that you recommend?
Greg

 

[Govert70227]

The MB part number for the diaphragm (or membrane) is A0000750607

The MB part number for the poppet cam cover is A0000750004
That number isn't know in the Bosch catalogue, probably because the poppet cam cover doesn't have a single number in the Bosch catalogue. I do include two PDFs from the BOSCH website: Bosch Automotive Tradition which lists all the parts of the pneumatic governor and an exploded view. It is in German, but the Bosch parts numbers are there. You might want to go the way of replacing the axle seals yourself, that is cheaper than buying the entire cover ready made.

The MB part number for the gasket between poppet cam cover and the rest of the housing is A0010742380

Although my friendly neighbourhood Mercedes-Benz dealer is very good (I was at the parts department once, said that a wanted a certain part number. Upon hearing the number the man helping me said: "that is that rubber hose for the engine ventilation") I always look up the numbers in the EPC (Electronic Parts Catalog) first. For Americans access to the EPC is free, look here for more info: EPC-net Online

If you want to keep the car and work on it yourself, get a Factory Service Manual, that has a ton of information.

I know only one internetsupplier Auto Parts at AutohausAZ - OEM Auto Parts - Discount Replacement Parts, Resources and Car Care Tips but I don't think it has these parts.

 

[GHewgill]

The MB part number for the diaphragm (or membrane) is A0000750607

The MB part number for the poppet cam cover is A0000750004
That number isn't know in the Bosch catalogue, probably because the poppet cam cover doesn't have a single number in the Bosch catalogue. I do include two PDFs from the BOSCH website: Bosch Automotive Tradition which lists all the parts of the pneumatic governor and an exploded view. It is in German, but the Bosch parts numbers are there. You might want to go the way of replacing the axle seals yourself, that is cheaper than buying the entire cover ready made.

The MB part number for the gasket between poppet cam cover and the rest of the housing is A0010742380

Although my friendly neighbourhood Mercedes-Benz dealer is very good (I was at the parts department once, said that a wanted a certain part number. Upon hearing the number the man helping me said: "that is that rubber hose for the engine ventilation") I always look up the numbers in the EPC (Electronic Parts Catalog) first. For Americans access to the EPC is free, look here for more info: EPC-net Online

If you want to keep the car and work on it yourself, get a Factory Service Manual, that has a ton of information.

I know only one internetsupplier Auto Parts at AutohausAZ - OEM Auto Parts - Discount Replacement Parts, Resources and Car Care Tips but I don't think it has these parts.

Hi Govert,
I'm having a devil of a time finding the parts you specified. The MB dealership says that they're all back order items from Germany without an arrival time, meaning they may or maynot be available.

The Bosch link you gave me earlier has a dealer finder but the dealers are independent foreign car repair shops. I called a couple and they have only limited access to Bosch parts. From the diagram you posted it looks like the shaft seals are simply O-rings (also how they're identified on the parts list you gave me earlier). Might I be able to use any good quality nitrile O-ring of the proper size? The diaphragm is going to be harder of course; no such thing as a generic or standard diaphragm. I have a couple of e-mails into diesel parts suppliers on the US east coast who do carry Bosch equipment, but I won't know until they reply.

One question I had about the poppet cam operation: you've identified "b" as the idle position and "a" as the full load position, but it appears that "a" is the retracted position and the same direction (right on the diagram) that vacuum pulls toward. I think I must be looking at this thing wrong somehow.
Greg

 

[Govert70227]

According to the MB website: Mercedes-Benz Classic Ersatzteile the cover A0000750004 is not available, but the diaphragm A0000750607 is. Don't forget we are still not sure whether the diaphragm is leaky, but if it is, it can be ordered.

You can try at engine overhaul companies, but as the USA hasn't got a lot of diesel cars, it will be more difficult to find the parts. In Europe there are a lot of these old MB diesels about, sometimes converted to marine use, so the parts are still kept in stock.

You can try replacing the O rings, but that might not be enough, sometimes the play between axle and bearings is too big. Remember the order of the different parts when you take the cover and axle apart. And you need a new gasket, although a liquid gasket and the old gasket might work too.

The poppet cam is there to stabilise the rpm at idle and at low partial load. The vacuum fluctuated too much at low engine speeds, so that the rpm will lope, also described as galloping or sawing. The poppet cam comes forward at low speeds, the diaphragm will rest against the poppet cam.