1986 300SDL Shake, Rattle, and......miss? - PeachParts Mercedes-Benz Forum

Diseasel300 · #1 06-01-2016, 12:09 AM

I've been working on my basket-case of an '86 300SDL and trying to track down and solve the deeply disappointing running quality at idle and low RPM's.

As a back story (cut very short, I'll detail in the project car forum later on), the car was traded into a used car lot in '07, car lot went bankrupt and was sold to another owner, new owner was going to refurb the car for his own use, fell ill, and the car sat in a back field until January of this year (9 years sitting for those keeping track). Since this is Texas, it survived largely intact and I basically stole it for a song with the intention of using it as a training bed to learn to work on a diesel. (I specifically wanted a fixer-upper).

Cut to the chase, everything in the car works, and it seems to be a restorable car with a minimum of money outlaid, but lots of elbow grease. I've been working on the engine on the weekends to get it going again, it ran on maybe 3.5 cylinders when I got it and is now running somewhat on all 6, although far from acceptable (or driveable) condition.

Important things first:

I've done a Diesel purge. I've replaced the copper washers in the delivery valves and the related O-rings. I rebuilt the original injectors, but they were too far gone to keep the halves from leaking. Installed 6 new Bosch injectors. Head is original #14 head, however it is crack-free, has new valve guides, new valve guide seals, and valve seats ground/machined.

Prior to removing the head I did a compression test (stone cold, not warm/hot) and got:
Cyl 1: 385 PSI
Cyl 2: 380 PSI
Cyl 3: 385 PSI
Cyl 4: 370 PSI
Cyl 5: 380 PSI
Cyl 6: 365 PSI

Cylinder 4 and 6 both had poor seal of the exhaust valve due to worn valve guides, explaining their lower pressures.

The engine is burning some oil (I assume stuck rings from sitting nearly 10 years with old/crappy/sludgy oil in it), but has virtually no blowby coming out of the breather, what little blowby it does have is not even enough to pressurize the valve cover if the breather hole is blocked. While the head was off the cylinders were checked for wear/scratches/pitting and found to be in good shape. This largely tells me the bottom end is pretty solid.

I just can't seem to get the thing to idle or run at low RPM's worth a crap! It feels like it's missing out randomly on 1 or more cylinders. It'll run smooth, then suddenly miss out and surge a bit (almost like a governor hunt), romp around, engine rocking around, smooth out, repeat, etc. If you give it a bootful, it is very hesitant to rev, almost stalling, until the RPM's get above 2000, then it begins to smooth out, then becomes perfectly smooth around 3000 RPM or higher, if the RPM's are backed off and held around 2000 RPM, it'll be perfectly smooth for a few seconds, then that random missing starts showing back up. As the RPM drops back down below 1800 RPM, it becomes more noticeable again.

I've tried cracking injector lines, and all 6 make the same difference. It becomes VERY obvious when the line is cracked and you have the dead miss.

It all screams to me fuel-delivery issues. The other big red flag (to me at least) is that if the car sits overnight, it'll fire right up the next morning after maybe 2-3 compression hits. However if it sits for a few days, it'll fire, die, then act as if it has to build prime again (15-20 secs of crank). I have it running from a temporary bottle right now as I need to clean the fuel tank, and I've confirmed that I have no injector lines leaking (nuts dry, injectors dry, delivery valves dry). All of the rubber fuel hoses are new as well, and the fuel heater is bypassed.

Could this be a lift-pump issue? It certainly seems to be pumping fuel through the clear filter, but the leaking down and random soft missing just screams air in the lines, leaking injection valves, mucked up injection pump, something! I would expect a mechanical fault with the engine to produce something obvious, excessive blowby, low compression, a consistent soft miss or dead skip, etc, but this seems to be completely random, I can't nail it down to a single cylinder.

Trials? Suggestions? Tips? I'm fresh out of ideas!

vstech · #2 06-01-2016, 06:42 AM

How do the fuel filters look?

Any signs of WVO use? Try putting a quart of wesson oil in your temp fuel... the IP could be shot.

funola · #3 06-01-2016, 08:07 AM

On your fuel filter housing banjo with 3 barbs, put a clear line on the barb to replace the line that returns to tank (on OM617's a cigar hose). There should be solid fuel in there at all times, running or not. If the car sits overnight or longer, and you see a bubble in the clear line at the barb (that's the highest point in the fuel system), you have a leak somewhere in the fuel system.

barry12345 · #4 06-01-2016, 10:47 AM

Sounds to me like several possibilities perhaps.

Yes you could also have a lift pump that delivers little volume at reduced rpm as well. Of course as another poster mentioned you also have to make sure you are not processing air by installing a clear line. You have changed the fuel filter of course? Also I would want to make sure the egr valve is staying closed when it should be all the time.

To answer what I would do with this sick puppy? First close off the return line from the injector pump. No change results install the clear line as suggested looking for air.

Next get the cheap liquid dampened 0-30 pound pressure gauge from harbor freight. It is cheap. So you can observe your fuel pressure in the base of the injection pump.

Your indication of fuel drain back by the prolonged cranking after the car has sat some time will probably be caught as well. To me it sounds as the revs come down either air or low fuel pressure or a combination of the two may be present.

Anyways no consideration can be given to the injection pump or anything else until these things have been done. Chances are the tests I suggest will locate the issues. With your original compression readings I would not have pulled the head. These fuel systems are pretty simple if simple tests are used to solve issues with them. Only guessing and not testing makes them seem complex.

ah-kay · #5 06-01-2016, 11:08 AM

I don't think 'you stole the SDL for a song' with all the money and effort you have sunk into the car. Why took the head off? The compression numbers were all in spec. It is a lot of work and 6 new injectors are $$$$.

I would try the lift pump. I had problem with 2 cars ( SDL and 87 300D ) with missing and idling problem. Changing out the lift pump and all is well. It is easy to replace but $$$ also for new ones. I got them from JY for pennies.

Diseasel300 · #6 06-01-2016, 11:50 AM

Thanks to everyone who has replied so far, certainly some ideas to try.

Fuel filters are both new, car does not show any signs of being run on veg. EGR valve is disabled and blocked with a plate (Nobody wants that kind of crap in their intake manifold).

To clarify why the head came off: Valve guides. Seals were worn and guides were sloppy. New guides and seals installed, valve seats and faces cut and lapped. No cracks in head, no warping measured.

Most of what I've invested in this car so far is time, which I have plenty of. I spent the last 18 years slowly restoring/repairing a 500SL Euro, it'll likely be more of the same with this one. The car isn't a DD, nor is it something I consider an "investment". More of a hobby. Some people may not see it that way but to each their own.

I'll give the veg oil a try. I'm assuming if the IP is scored or leaking this should make a difference to an extent since it's thicker oil? I will also try the clear line and blocking off the return hose from the injection pump filter and report back with any changes in behavior when I get some time and a pressure gauge. Out of curiosity, what is considered normal fuel pressure with plain pump diesel?

barry12345 · #7 06-01-2016, 01:06 PM

Well the expected fuel pressure in the normal range should be perhaps 9-15 pounds in the base of the injector pump. If it gets as far as having to actually check the pressure is still an unknown at this time. You test to isolate the actual cause over guessing.

It is usually both cheaper and less frustrating. Nothing wrong either with treating this as a hobby. Always lots to learn. Sounds like your basic engine is in decent mechanical condition as well.

No reason to suspect the injection pump itself yet. The majority of the time they are still just fine anyways. From your descriptions I think I noticed one issue with your lift pump.

If it is enough to be concerned about will show in the tests. Just update your posts as you move along. There are lots of members that will help.

To me at least there are lots of puzzles to solve and my personal favorite is learning more and more about these fuel systems. You were right incidentally as you do have problems in your fuel system.

You could use the vegetable oil test at this point but personally I would avoid it. The thicker viscosity could mask weak valves in the lift pump as well as weak injection pump elements in your case. Plus automatically because the flame front is of longer duration will make the engine sound different perhaps as well. In other words it would not be conclusive at this point but may prove useful later. Have to eliminate the relief valve possibility and lift pump plus any air issues first.

Usually their fuel systems decline result from either just wear and age but some from less maintenance over the years than needed. For example the periodic fuel pressure checking is required in the service manual but most are never done.

Diseasel300 · #8 06-02-2016, 02:54 PM

A partial update:

I'm not sure I understand how to measure lift pump pressure going to the IP. Both ends of the hose are a stiff transparent plastic with Banjo fittings on both ends. Other than cutting this line and replacing it later with a new fitting, is there something I'm missing?

Here's what I did do:

Measured output volume through return hose: FSM shows leaving ignition off and using a special relay bypass tool to crank the engine without fuel. I did this by tying the STOP lever down and cranking from the key switch. FSM shows a minimum of 150cc delivered volume through the return line in 30 secs cranking on the starter. My pump delivered ~200cc in 30 secs.

While I had my setup in place, I inserted a 50PSI gauge I have in the return line and clamped it in place, then cranked on the starter again. Maxed out ~14PSI, then slowly (over about 10 minutes) returned to 0. No fuel leaks are obvious.

I'd assume this means the lift pump is probably OK (pressure and volume-wise) and the long crank after sitting for a period of time is due to a check/relief valve somewhere leaking back?

I've not done a compression test since refitting the head with new gasket, however the compression hits on the starter "sound even", e.g. no skip or speeding up. I know sounds don't mean a whole lot, just making the observation.

I've also read a few threads (on this forum and elsewhere) stating that the new Bosch injectors made in India (these are the ones I have) can cause problems. Would it be worth doing a pop/leakdown/pattern test and a fresh compression test while I have them out?

I'll be busy for the next couple of days days without any time to really mess with the car, but I certainly appreciate any analysis and further ideas worth trying here!

jt20 · #9 06-02-2016, 03:15 PM

Have you verified injection timing? how?

could the camshaft be out 180 degrees since the head install?

any smoke when its running?

Also, it takes a bit of skill to replace the orings and washers on the delivery valves properly - if they are off, it will cause issues.

ps - I checked this thread bc of your username - very funny, very fitting.

Diseasel300 · #10 06-02-2016, 03:50 PM

I strongly doubt the camshaft is 180˚ out. If it was, I'd expect to have other symptoms. I made witness marks on the chain and cam sprocket and took the extra step of zip-tying the chain to the sprocket and keeping it under tension while the head was out to maintain timing.

Before the head came out, the cam was ~1˚ ATDC, after it went back in, it was still aligned to the same point. Chain, sprockets, and chain guides all appear to be in good condition, no cracks, sharp edges, gouges or wear apparent.

I *HAVE NOT* checked injection timing, in fact I was planning on asking about that in the previous post, but the phone rang and I forgot to ask about it.

There is smoke present, never goes away. Very difficult to tell if it's oil, diesel, or both. With as long as the car sat, I'd expect some oil smoke, however it also stings the eyes a bit, which from what I've read is usually fuel-related (timing, volume, injector issues, compression, etc). I can troubleshoot a gas engine with my eyes closed, but Diesel is certainly a learning experience (but I enjoy it).

The back story on the name of the car is from when it was fired up for the first time on what seemed like 3.5 cylinders. Sounded like someone with COPD hacking up a lung! The guy I had helping me said "It sounds like it's got a disease!" and the name "The Diseasel" was born and has stuck around since...

barry12345 · #11 06-02-2016, 07:22 PM

Three or more points perhaps of interest. There are two check valves in the lift pump and when you mentioned drain back I suspected them. Both of them function as anti drain back valves as well as functioning of the lift pump. These two valves probably remain the greatest reason for the cranking required to get the engine firing after a longer period of sitting. Near the end of this post is where more useful information currently might be found.

But those valves can still be leaky and supply enough operational lift pump pressure. The relief valve before the relief overflow line from the injection pump was not really tested .

You can close off that return line for a running test. Do not leave a non dampened meter in the position you describe though as you might destroy it with the pulses that are present there when the engine runs. Anyways you would not be able to read it . The gauge needle would just swing one way to the other very rapidly.

Because your placement of the gauge deadheaded the system. . It is completely safe to run the lift pumps with a deadheaded situation by the way. I would have suspected or wanted to see a higher pressure than the 14 pounds. Yet much may depend on the system not really in running operation when you took the reading.

The common measuring point for the injection pump base pressure is either take a banjo fitting and drill and tap it for a fitting in the top. Then you can read the active pressure with a running engine from it. You also can buy a banjo bolt already done as well. Only a liquid filled gauge has any use there though.

What you did do though. You did indicate in a pretty static way that the lift pump was providing enough pressure for the engine to run reasonably. Well below what it should have been in that deadheaded situation yet still enough.

What was lacking was the relief valve could be totally open and you have not established it was not. That is what the running deadhead would establish.

So at this point I would put reconditioning of your lift pump or replacement if a kit is not available by yourself as a possible future need. All the 123 series cars have home rebuildable lift pumps with a kit available cheap. I am not sure yours has.

Your lift pump being good enough for the time being or not is still in the realm of reading the base pressure in the injection pump base with a running non dead headed system. Then closing the return off to see what the top pressure the lift pump is delivering. Personally I would have liked to see perhaps 25-30 pounds in a running deadheaded situation like yours. Your newer fuel filters will not obstruct the pressure much in the running closed off test. Old unknown fuel filters can seriously impact the available pressure though.

Volume pumped is a requirement to develop pressure. So I seldom use it as an indicator. Substituting dynamic pressure readings instead. You will become a lot more familiar with your fuel system before this is over.

Guessing at them makes sometimes undesirable needless costs occur. Plus in all too many cases leaves other issues present that deteriorate your performance, fuel millage and starting up qualities. Plus If you go through a road breakdown because you left an unknown condition behind you will wish you had not in many cases.

Fuel system failures are the most common reasons for these cars breaking down on the road I expect. Very cheap to address on the 123s to make them right again if done at home.

A prime reliability factor on these old cars. Your 603 engine car has a different lift pump than the 617s have. I have always wondered if the internal valves where the same as the earlier lift pumps.

Anyone know? If they were I would have thought it common knowledge by now. I have never seen a 603 lift pump.

With an updated #22 head I see the 3 liter 603 as the best diesel car engine ever produced by Mercedes before any real electronics made the scene. A real shame they never offered this package from the factory. Almost bulletproof and more efficient plus powerful. Quieter as well in comparison to the older engines.

Injection timing will not make a really signifigant difference unless really way off when general troubleshooting. Rotating the engine to the required damper indication point and taking a small mirror to have a peek in the timing port is enough. If the indicator is roughly centered consider it close enough. Of course if it is not visable rotate the engine 360 degrees to the same mark and look again. You have to at least see it.

If the injection pump were 360 out I would expect at least a lot of exhaust smoke from the tailpipe. In comparison to what was there before you pulled the head in your case. .When you see the indicator in the injection pump port. If you can get a peek at the lobes of the first cylinders cam through the oil fill hole. Have a look to verify that they are generally pointing upwards as well. I understand how some of Mercedes diesels run with the injection pump off 360 degrees but not the 603.

I will mention just how people get this messed up when they think it will not happen. They set their damper position to zero. They do not make sure the cam timing marks are aligned because they have forgotten that there are two top dead centers. Then get the head done and when reinstalling line up the cam marks. This puts them 360 out in injector pump timing. If you tied the cam sprocket to the chain of course this cannot occur. It is removing the sprocket from the chain that is the enabler. Actually a quite common error and guys buy the cars cheap where the owners have given up trying to figure what is wrong on occasion after they did their heads. Not that long ago one of our members drove a 606 engine Mercedes a long way home with this issue. You should currently have no more exhaust pipe smoke that before you pulled that head anyways. Because if you have signifigantly more and especially if it stings your eyes you have to verify the timing.

As for the difference between diesel and gas cars when it comes to valve guides and seals. My old gas Mercedes would suck substantial oil down the intake guides when the seals were poor. Mercedes over oiled the cams in comparison to most engines.

Primarily because the intake valve guides where under vacuum on gas cars. Your diesel engine has no intake vacuum of any importance. Instead it gets also pressurized by the turbo that will drive any oil trying to get down the intake guides back.. The exhaust guides are also never submitted to less than the pressure presented by the back pressure of the exhaust system on both gas and diesel cars so little oil in general goes down them. Although a valve guide broken off flush with the head will allow too much oil to be processed in these diesels. The top of the head is really flooded by oil on these engines is the probable cause.

I can see no way you messed up the timing from your description. But being the slow thinker I am. Is there a chance the timing of the pump was 360 out when you purchased the car? In that case you would have just duplicated the wrong timing as it was already existing. Also when you checked the timing originally it could have been what you observed but false. You would be far from the first.

I should have mentioned this as soon as you said the exhaust smoke stings your eyes. You have to check it out. Could easily be the answer even if a very long shot. Actually unless the new injectors are totally garbage it is a strong contender. I believe your main problem is perhaps incomplete combustion. You have compression. So fuel almost has to be injecting with some defect.

Since you want to recondition this car over a longer period and it is not now needed as a daily driver. I see no reason that some more of my extensive trouble shooting methods should not be posted for your consideration as this project goes along. They are generally cheap to employ as your engine and its systems in good condition provide a fairly strong engine for a car diesel. Years ago I could not believe just how fast the 603 engined cars were with engines in good condition. There is a way called the milli volt method for example that is very applicable if done right. You can get these engines to almost talk to you about their general condition with it if done properly.

Sorry about the long post.

Diseasel300 · #12 06-02-2016, 08:45 PM

Barry, you may be onto something here. The smoke level is comparable to how it was before the head was done, a bit less than it was, but still pretty disappointing seeing as how the bottom end seems so tight and clean (cylinders looked beautiful, not glazed, not a scratch or score in sight). The smoke odor is pretty nasty, a very acrid smell. On a gas engine oil smoke usually just stinks (and makes you reek of it), but this is somewhat different, I assumed since it's a diesel and not gas engine, but now I wonder...

How can I verify base timing? You mention a port, where is this located? I'd like to make sure the base timing is at least in the ballpark. I know the IP is injecting fuel, but it would certainly be nice to make sure it gets injected in the compression stroke instead of the exhaust stroke.

While I took a lot of time to make sure everything was kept in time when I had the engine apart, who knows what happened to it in the past? I know it has had injector work done in the past since it had 3 (very crusty) rebuilt/used injectors, 1 unknown brand (no markings at all) injector, 1 Audi/VW injector, and 1 115bar injector from a N/A diesel in it. I wonder if it's possible someone pulled the pump and put it back in out of time (or 360 out).

I did a bit of reading after seeing your response and it seems that a 6 cylinder diesel will run (poorly) with the timing 360 out. Symptoms are a supremely crappy idle (if it will idle at all), smoke, hideous throttle response, and very poor power. All of which are present here.

I'm not diverting from testing fuel pressure, I ordered a 30PSI liquid-damped gauge and some spare Banjo fittings this evening so I can hopefully test fuel delivery pressure sometime next week.

If we determine that the fuel pump needs overhaul/replaced, I'll do a teardown and document it for comparison to a 617 LP. Since it's now confirmed that there's a check-valve in the LP, and considering how gummy everything else has been, I'd imagine the LP will be coming out at least for a cleaning in the near future.

barry12345 · #13 06-02-2016, 10:17 PM

I am no expert but believe the newer diesels than the 616 and 617s run much better when the pump is installed wrong. I again have never heard any reports on the 603 in this condition. Still it would not surprise me they would also run better than the earlier engines as well.

Look up the port location where you remove a bolt plug to observe the injection pump mark. Its location will be found by other members either posting it or in our archives.

From what the injectors looked like it may have been someone trying to solve the problem before your acquisition. The basic engine condition sounds good so it is worth pursuing.

Having a fuel pressure gauge to me is never a waste of money. These fuel systems are very old and almost always can use some attention. Bringing them back does increase reliability and performance is well established.

They usually are cheap to rework where needed. You do have to own a gauge to do it properly though. Depending on your output pressure of the lift pump will determine what needs attention. Since your lift pump is perhaps expensive. I decided some time ago. That if the running pressure is still really decent a cheap check valve can be put in the line to eliminate drain back. That's if it is a proven issue on the more expensive lift pumps. Some members have limited funds so stretching their dollars is required.

I was not born with a silver spoon in my mouth in a working class family. Still seem to have this considering costs in my mind still when the need is no longer there. An old habit that is hard to break. Even though we reside in an area of eastern Canada that time has passed by to some extent.

It is only ten to fifteen dollars to rebuild the 616 and 617 lift pumps at home and pretty easy to do. Your type may be priced pretty steep and perhaps a used one acquired and tested is an option otherwise. I am almost sure that no kit is available for yours but not positive. I Liked your ideal of at least checking for gummed up dirty valves first if the need is indicated. You are not short on fuel as you cannot burn what is being injected now. The smoke out the back should be both stinging the eyes a little and perhaps quite dark. Not a lot of it is a requirement.

You may not have drain back but instead if the car sits a longer time the residual unburnt fuel may flow down through the rings requiring a substantial amount of new fuel injected to light off. Harder to start is another claim of those that have wrong time injection. Actually you have all the reported symptoms more of less if you think about it.

You obviously know a lot about mechanical things. Sometimes I do not like the way I present things. I never want it to seem that am talking down to anyone. In my experience everyone knows more about something than I do.

I am not a working mechanic either. It is just a hobby repairing things basically. I enjoy attempting to help others as many have helped me over the years. This is also a pretty civil forum and we have some members that really know a lot about these old diesels. I seriously doubt I would own a handful of these older diesels if the forum did not exist. It is always nice to see newer members on site. You bring some interesting problems and knowledge to the mix.

Diseasel300 · #14 06-03-2016, 05:00 PM

*That "Eureka" Moment*

Had a few minutes free this afternoon, popped the valve cover and spun the engine over to ~20ATDC for Cyl #1 (I know the manual calls for 15˚, I wasn't going to spin the engine over twice more for 5˚).

Popped the 17mm timing cover off the side of the Injector Pump and Voila......no tang.

Followed the instructions/photos in Post #28 in this Thread:
http://www.peachparts.com/shopforum/diesel-discussion/328989-oil-pump-chain-broke-86-300sdl-how-replace-2.html

Very clearly NO TANG in my viewing port. With being ~5˚ off from 15˚ATDC, I'd expect to still see the tang, parts of it, or the shaft visible in the hole if the timing were anywhere close to correct.

Now I need to know: Is there a How-To on how to get back to base timing? I know the IP has to come off and I need the holding tool for the IP tang, but that's about the end of my knowledge at the moment.

jt20 · #15 06-03-2016, 07:55 PM

Begin by spinning the crank until you find the tang. then check the camshaft location. compare with the crank readout.

2 ways do deal with this without special tools:

Drip method: pull the IP, re-install at proper location, drip time by removing the delivery valve

or

Use the tang in the IP at 15 deg on crank and estimate where to install the pump (you can purchase a pretty cheap locking tool for it)

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode