Vacuum Problems - Troubleshooting Help Needed - PeachParts Mercedes-Benz Forum

mach4 · #1 10-11-2013, 07:06 PM

The problem I am having is an intermittent one, making troubleshooting particularly vexing.

The engine is an '85 617.952 with the EGR and ARV disabled. The chassis is an '85 107.

The symptoms are

1 Difficulty with shutdown - Occasionally the car is difficult to shutdown, either doing so slowly or a couple times, not at all (used the manual shutoff). Sometimes revving the engine a bit helps, other times not. The shutdown is a manual setup with a vacuum solenoid allowing vacuum to the shutoff valve when a switch on the dash is pushed.

2 Clunking into 1st gear when slowing down to a stop

3 Hard 2-3 shift on acceleration

These symptoms work in tandem. If I'm slowing down with the intent of shutting the car down, if it clunks, it's going to be difficult to shutdown. If there is no clunk, it shuts down normally (or close to)

Just got back from a 5800 mile road trip where things worked properly 90 plus percent of the time. Then on the very last leg it refused to shutdown completely. The symptoms would randomly appear and disappear seemingly at will.

I've checked all the rubber hose connectors for any obvious leaks and they appear to be fine. Brake pedal is always normal; no hard pedal. Door locks always work properly and the reservoir holds for 5 or so cycles as it should. Trunk and fuel locks all work.

I used a mechanics stethoscope on the vacuum pump and there are no unusual sounds.

I've checked the routing against the vacuum line diagram and all the orifices and dampers appear to be in their proper place and the lines connected as they should.

I haven't checked recently, but last time I did, the transmission vacuum control valve functioned properly.

I've tested vacuum at the following points with the readings shown. The 12 inches is steady state and it drops to 10 upon acceleration - presumably due to the transmission VCV doing its job.

And things can be extremely random. For example, we stopped for dinner last night and the car wouldn't shut off. Used the manual shutoff on the engine. Then after dinner, started the car and, as a test, immediately hit the shutdown switch and it shut down completely normally.

It seems like I've checked everything. I have a spare vacuum pump I'm planning to rebuild as soon as I can source the proper bearings, though that doesn't seem to be the problem.

I'm out of options and open to suggestions and ideas.

TnBob · #2 10-11-2013, 08:01 PM

Mityvac or similar tool is a huge help in chasing vacuum issues. Sure great for finding hidden vacuum leaks in a circuit.

kerry · #3 10-11-2013, 08:30 PM

Should have higher vacuum than 12 on the shut off circuit but it sounds to me as if the check valve in the vacuum pump may be malfunctioning sometimes.

mach4 · #4 10-11-2013, 09:00 PM

Quote:

Originally Posted by TnBob

Mityvac or similar tool is a huge help in chasing vacuum issues. Sure great for finding hidden vacuum leaks in a circuit.

Yup, got one as well as a boost/vacuum gauge that I can hook up and drive for testing.

mach4 · #5 10-11-2013, 09:02 PM

Quote:

Originally Posted by kerry

Should have higher vacuum than 12 on the shut off circuit but it sounds to me as if the check valve in the vacuum pump may be malfunctioning sometimes.

If that were the case, wouldn't I be getting stiff brake pedal sometimes? That's never happened.

vstech · #6 10-11-2013, 09:53 PM

the brake line is isolated from vacuum leaks in the small stuff... only way you'd lose brake vacuum is if you broke the main vacuum line, or the pump is defective...

I do not like the way you have the vacuum connected and flowing through the vcv correctly.
post up pics for others to critique, and lets fix the problem.

mach4 · #7 10-11-2013, 10:53 PM

Quote:

Originally Posted by vstech

the brake line is isolated from vacuum leaks in the small stuff... only way you'd lose brake vacuum is if you broke the main vacuum line, or the pump is defective...

I do not like the way you have the vacuum connected and flowing through the vcv correctly.

Post up pics for others to critique, and lets fix the problem.

I'll get some pics tomorrow.

I'm quite sure however that I've got things set up properly. If you look at this detail from the full vacuum diagram for the '85 617 Cali version, the side connector to the VCV for the transmission clearly just vents to atmosphere - I should have it plumbed to the cabin, but it is just a vent line.

Below is the full vacuum diagram.

The vacuum source for the VCV is the "front" port on the check valve (140) and runs through orifice 63 or 63a depending on whether the vacuum transducer (125) is active or not and through the green vacuum damper (72) to the VCV.

One thing that I should point out on my configuration is that the '85 671.952 is mated to a 722.4x transmission. When my tranny went out, I replaced it with a 722.3x from an 83 but left the blue flying saucer Vacuum Transducer (123) in place. The transmission has worked perfectly with this configuration even thought that's not what is technically called for.

I think what you're getting confused is the '83 that has a Y-connector with one end going to vent and the other to the 3-2 valve. It's still just a vent. And the Vacuum Control Valve connects directly to the transmission, rather than going through the Vacuum Transducer (blue flying saucer device).

'83 Vacuum Setup

Hope this helps clarify my unique situation.

The system has worked for some 25k miles, so I'm a little reluctant to think it's the setup that's wrong... but, that's why I'm asking.

vstech · #8 10-11-2013, 11:29 PM

I see said the blind man, thanks for posting up that pic.

since viewing this, I'm going to have to say you have a leak in one or more of those hoses we were looking at on your setup.

start by isolating the two circuits on the checkvalve.
plug one hose, then the other, and see what's isolating it. too bad we didn't have time to do it when you were in my yard!

whunter · #9 10-12-2013, 01:07 AM

Quote:

Originally Posted by kerry

Should have higher vacuum than 12 on the shut off circuit but it sounds to me as if the check valve in the vacuum pump may be malfunctioning sometimes.

Very possible.

* A failing shutoff element.

* A failing transmission vacuum valve and/or modulator.

* A failing vacuum pump would explain the seriously low reading.

* Failed climate control elements can also be part of the issue.

FYI: The OM616, 617 engine shutoff element requires at least a solid 13 inches of vacuum for correct operation.

.

DeliveryValve · #10 10-12-2013, 01:17 AM

^^^ agreed, what he said..

At least for the climate control, you can plug up the line and run it without it to see if it is the culprit.

Shut off element and trans modulator can be checked with a mity vac to see how long it holds vacuum and their function.

.

pmckechnie · #11 10-12-2013, 08:47 AM

Having worked on intermittent problems many times I would like to suggest you connect a vacuum gauge to the vacuum system so you can see when the problem is there and when it is not. It may be failing more often than you think and a gauge would help you know when it fails. For instance, on your 5000+mile trip, it could have failed many times but you didn't know it because it didn't fail when you were turning off the engine, locking the doors, etc, at that exact time. If I am not mistaken the 380SL had a economy gauge in the instrument panel which is just a vacuum gauge. If it is still there you could use it as an indicator. You don't need to know exact readings, just an indication that something is not right. Then you could disconnect different systems, one at a time and be able to see if the problem has gone away. If not, go to the next system. Get the point? Then after the problem is found and fixed you can keep the gauge connected and you will have an indication of the vacuum system health.

Paul

Brian Carlton · #12 10-12-2013, 09:18 AM

Quote:

Originally Posted by mach4

b.

140 is the check valve. It simply prevents vacuum from bleeding back into the pump at shutdown.

The vacuum levels should be identical at all ports.

Replace the valve. All will be well.

mach4 · #13 10-12-2013, 09:20 AM

Quote:

Originally Posted by pmckechnie

Having worked on intermittent problems many times I would like to suggest you connect a vacuum gauge to the vacuum system so you can see when the problem is there and when it is not. It may be failing more often than you think and a gauge would help you know when it fails. For instance, on your 5000+mile trip, it could have failed many times but you didn't know it because it didn't fail when you were turning off the engine, locking the doors, etc, at that exact time. If I am not mistaken the 380SL had a economy gauge in the instrument panel which is just a vacuum gauge. If it is still there you could use it as an indicator. You don't need to know exact readings, just an indication that something is not right. Then you could disconnect different systems, one at a time and be able to see if the problem has gone away. If not, go to the next system. Get the point? Then after the problem is found and fixed you can keep the gauge connected and you will have an indication of the vacuum system health.

Paul

Good suggestions. You are right that the SL has an "economy" (vacuum gauge) however I've repurposed that as a turbo boost gauge.

I've got a boost/vacuum gauge that I can use to tap into various points in the system for monitoring over time from the cabin.

mach4 · #14 10-12-2013, 09:23 AM

Quote:

Originally Posted by Brian Carlton

140 is the check valve. It simply prevents vacuum from bleeding back into the pump at shutdown.

The vacuum levels should be identical at all ports.

Replace the valve. All will be well.

I've got a spare. I'll give that a try.

mach4 · #15 10-12-2013, 02:23 PM

Thanks for all the suggestions and ideas.

I pulled the check valve on the vacuum pump and it was good. It looked clean but I gave it a good cleaning anyway.

Tested the VCV on the IP and it was fine.

Tested the transmission modulator and it was fine.

I tested the vacuum shutoff solenoid for leaks and it was fine.

I then rechecked the vacuum at the rear port on the main vacuum line tap and it showed 8 in-hg.

I then plugged the one line that provides vacuum to all the chassis components and re-checked the vacuum as before and it went from 8 to 22 in-hg.

So while I don't have the problem solved, I know it's not in the engine bay.

The thing that threw me off was that since the vacuum locks continued to work after the engine was off, I assumed that I had no vacuum leaks in the chassis. Bad assumption. But the fact that the locks continued to work for the 4-5 cycles does seem to eliminate those from the equation.

Now the symptoms begin to make sense. The one time that the engine failed to shutdown altogether, was the one time that we were using heat on the trip. In fact it is the first time I had the heater on since probably May. So I'll start looking in areas of the climate control components for the leak first. Whunter and Delivery valve called it - thanks.

So first point of troubleshooting the chassis vacuum was to hook up a vacuum tap at the point where the chassis gets its vacuum from the VP. I drove it around with the gauge registering 6 in-hg in the off position. Moving it to other positions did nothing except defrost, where the reading went to 10 in-hg. Not sure what that means but it's a start.