When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
I am the fairly new owner of a 1986 944 Turbo with 157,000 miles on it. It's my second '86 951 and it is quite a bit more "used" than my first one - which was bone stock and only had about 86,000 miles on it.
This one has a Vitesse MAF conversion with a vrmomonitor which I know nothing about as well as a 3" stainless full exhaust system.
Anyway, when I first start it it will run perfectly smooth, then if I drive it for a while and I'm at a stoplight or just letting it idle, it will almost feel like it is going to kill. Then it catches itself and it will idle fine for a little while and then go back into the same convulsion and recover routine. If that weren't enough I think I need to replace my motor mounts, so every time it happens it scares the **** out of me - thinking I'm going to have to attempt to re-start the car.
Curious if anyone knows where to start to debug this one?
I'll try to post up some video of it soon. Any insight would be greatly appreciated.
Can you hook up a PC and do some logging with the MoMonitor while this problem is occurring? That would help us immensely. Otherwise, that's a pretty tough one to diagnose over the Internet.
Here's a few things I see:
1) The idle is not set correctly, it's way too high. It might have been set high on purpose to compensate for a problem that the previous owner was having.
2) If the problem ONLY occurs at idle, and ONLY after the engine is warmed up, then that helps isolate the problem. Some of the usual cuplrits are:
a) Idle Air Control valve - It could be that the DME is trying to close the Idle Air Control valve because it sees that the idle is too high, and then when the RPM's completely cut out, the DME is reopening it to compensate for the excessive RPM drop. Just a theory, this might not be it though. I don't think the MoMonitor logs IAC control, but if it does, that could be really helpful.
b) It has the appearance of the computer (DME) temporarily losing connection with the speed and reference sensors. When a 951 loses either the speed and reference sensor signals, the engine cuts out and the tachometer loses the RPM signal completely. This is not an easy one to troubleshoot. There are a few possible problems - the connection between the sensors and the DME (a problem that most all 951 owners experience if they have the car long enough), the operation of the sensors themselves, or a bad installation whereby the gap was not set correctly (i.e. gap is too large - only way to confirm that is to uninstall and reinstall the sensors, and you'll need an old one with a washer glued to the bottom to use as a reference). Do you know if the previous owner ever replaced the speed and reference sensors, and if so, was that recent?
c) TPS switch - usually this is just an on/off type of problem, and usually repeatable, but it's possible that the TPS switch adjustment could be sitting just on the borderline between open/closed, and vibrations from the engine might just be enough to set it off. Easy enough to check it though. The guide at Clark's Garage is a good one - http://www.clarks-garage.com/pdf-manual/ign-03.pdf. The main check for this problem is to check the adjustment, to make sure it clicks to the closed position, and clicks back to the open position as soon as the throttle is opened.
One of the common problem areas for the connection to the speed and reference sensors is the plastic connectors to the DME (engine computer) wiring harness. There are two connectors, one each for the speed sensor wiring and reference sensor wiring. The connectors get very brittle with age and fail. They are located on the top of the engine bay area near the firewall, and they should be attached to a metal bracket. When the car is warmed up, try wiggling these and see if you can make the idle hiccup occur. If you can make it occur, then you know that you have a connectivity problem with either the speed or reference sensor.
Here is a pic of the bracket holding the speed and reference sensor connectors, and the O2 sensor plug:
I'm wondering if the idle might have been set higher to compensate for this hiccup or if it maybe was to prevent the rattling due to possibly needing new motor mounts?
The problem does only seem to affect the car when it is warm and only at idle as far as I can tell.
I'll see what I can do to find and download the software to try to get a laptop hooked up to the MoMonitor. I know literally nothing about how any of that works.
And I'll check those connectors next time I have it warmed up. I'll also see if I can find any trace of replacement on the service records in the file.
The tach bounce is odd. I don't recall ever seeing anything like it. JMJ951 post is spot on.
I wonder if there is a bad solder joint in the idle control circuit in the DME. Try setting idle by following the procedure in Clark's-garage. When you short the pins on the test port/socket the idle control is taken out of the loop, note if the bounce remains before adjusting idle.
The Monitor will be of great help, you can see many variables as processed by the DME. Drop me an email and I'll send you the instructions and software for your PC.
Well, it definately looks like someone has been in here working with the sensors in the past. I took one of the sensors off the car (the one with 3 pins in the plug and the bracket trapped on it), freed the bracket from being locked onto the wire, and soldered the pins (they were just crimped on before) and then put it back together. I'll see if it works any better tomorrow when I can give it a test drive. When I stripped off some of the insulation in the process of soldering it, the #3 pin along with some shielding wire just fell off. I don't know if that was because the wire was almost ready to fall apart on its own or if I cut the wire when I removed the outside layer of insulation.
I'm almost afraid to look under the plug that has the electrical tape wrapped around it. That connector only has 2 pins - guessing that is the speed sensor?
Fast951 I will shoot you an email - I have a few other questions for you too about the Vitesse items on this car.
I would get a new sensor. To save some $ get the sensor for BMW 535i MY87-88. It is a bit longer than the 951 sensor, but much cheaper. Stick to the Bosch brand.
Replied to your email with the software/instructions.
Keep us posted. It'll be interesting to know the root cause of the tach bounce.
Very interesting indeed! I’m curious to see the result too.
Definitely replace those sensors; the BMW ones are $60 US each not bad and cable length is a non-issue.
Anyone think his vacuum is a little low? On my car, with 17 in vacuum at warm idle, the boost guage (also vacuum gauge under “1”) shows just less than .4. His cold is super low vacuum and warm seems a tad on low side?
Last edited by Dan Martinic; 05-25-2018 at 01:22 PM.
Reason: Added pics
Ok I got the MoMonitor app installed - thanks fast951!
Unfortunately when I went out to hook it up, after starting the car it started and ran rough and ended up killing. I think I am in for some new sensors.
So just out of curiousity, which one is the ref sensor and which one is the speed sensor? And is one supposed to only have two connectors in the plug and the other one 3 or are the two sensors identical to each other?
Thanks Dan, I was doing a bit of research which also confirmed that the sensors are identical. As far as your question about the MoMonitor, I'll let Fast951 answer your question as he is the expert in that arena.
I am posting some more pix of pulling the second sensor out and some video of my MoMonitor in action.
I did get the car to run for a little while but now it exhibits a similar but worse behavior than before and it is now doing this when cold. I think my problem has to do with the sensors probably needing replacement as Fast951 suggested. I will play around with it a little more and wiggle the wires while it's running like jmj951 first suggested too.
Here's what the MoMonitor looks like - it connects to a cable that is somehow connected with the DME (I think - I didn't install it):
PS if that video is your car’s reading, and if this MoMonitor is very accurate, then my suspicion of low vacuum seems correct: it should be closer to 20, not 12 or so that you are showing on the MoMonitor (if I understand the monitor right).
I’d say you might have a couple of teeny vacuum leaks
Yes, this one was quite thick I noticed. Thanks - glad that is supposed to be there - wasn't sure.
I also went out to the car and started it and tried wiggling the wires around by the sensors and the plugs and that didn't seem to make any difference. I ordered a couple of sensors - really hoping that takes care of this issue. Will have to wait a week or so before they arrive tho...
05-23-2018, 04:42 PM
Idle hiccup when engine is warm
