Weston

I'm having a really irritating problem with my '83 944 stumbling and cutting out... It starts up and seems ok at first, but I get going down the road and it will stumble or cut out, and the tach needle will rapidly drop (there's no way that the engine is really slowing down that fast). Sometimes the engine will come back alive and the tach needle will fly back up to where it should be. It's hard to say for sure, but I have a suspicion that this is related to throttle position, as it seems like it usually happens when I am slightly increasing throttle; letting off of the throttle seems like it may help it come back to life, but I'm not 100% convinced. It will sometimes die at idle too. I would normally suspect TPS, AFM and/or fuel pressure, but I don't know of a way that those would cause the tach to drop so fast.

It has been cold here in Colorado and this problem seems to be temperature related... Every time that this has happened, the car had been sitting in the cold (below freezing) for several hours. But if I try it again after it has been sitting in slightly warmer weather (40-50 F) for a few hours, it drives perfectly except for a little stumble at 2000 rpm when increasing throttle (it's fine when holding steady throttle through 2000 rpm; it just doesn't seem to like more throttle at that spot). I re-tracked the AFM the other day in hopes of fixing the 2000 rpm stumble and the 5000 RPM cut that I was experiencing... the 5000 RPM issue seems to be a lot better now (maybe not 100% cured though), but it didn't change anything else.

Any ideas?

Thanks

skene

It sounds like an issue with your reference sensors. One of them feeds the signal to the tach and if it's got a shoody connection your tach will drop and your engine will cut out...which sounds like what's happening.

It's very possible that the cold weather is causing a bad connection somewhere in those wires/connectors. I'm not sure if your 2k throttle issue is related, but it might be?

Weston

Thanks for the info. I took a look at the reference sensors and this is what I found...





The other sensor was pretty much the same deal when I pulled it apart... the sensor's connector was broken and not really held in place by anything, and the wires themselves on both sensors were not inspiring confidence either. I tried to fix them with zip ties and duct tape, but it still acted up around 2000 and 5000 rpm as I drove around. It died at one point and I could get it to start back up after waiting a minute, but every time, it would die as soon as I stepped on the gas... moving the wires around fixed that and got it home, where I promptly ordered two new sensors ($100/each on eBay seems to be the best deal for new ones). I figure the engine vibration was affecting the connections, and was also why the throttle position and/or RPM seemed to be part of the issue.

Jakerx

Yep that definitely looks to be the source of your problem. Get it replaced!

Weston

The new reference sensors are in and they definately make a difference, but there is still a problem. The engine still hesitates badly at 2000-ish RPM when giving it more than a little throttle, and the tachometer still drops faster than the engine, but it's not a complete signal loss like before (it seems to just fall about 500 rpm when the engine cuts out, then shoots back up when the engine decides to work again). I also still have the same kind of cut out around 5000 rpm, but it's not always there... it seems to be more common in the cold, so I figure it's running lean. I checked out the TPS and it is working fine. Any ideas?

I ordered a new fuel filter and new gaskets for pretty much everything that could cause a vacuum leak, and I'll go over the vacuum lines again when the intake manifold is off. I'm thinking it's a fuel pressure issue at this point, but I'm still confused as to why the tach is still dropping a little bit. I'm tempted to run new wire from the reference sensors to the DMU, but that seems like a pain, and I'm not convinced that it would fix my problem.

The car was running fine before it was parked and sat outside in Colorado for about a year. I know it's probably just one little thing that's causing 90% of the problem, but I'm still not sure what that is...

Thanks for any help

Weston

Update...

There were 3 big problems:
1) My reference sensor connectors were broken and not making a solid connection all the time. Putting in new sensors improved things and the tach didn't completely die anymore when the engine cut out, but there was still a problem.

2) There was an oddly wired kill switch & relay on one of my DME relay's wires. I found this when I ripped my interior out to install a roll cage (it's a 944-spec race car). It did not appear to be a solid connection, and slamming the door did cause the engine to die once. Getting rid of all of that solved all of the cutting out, but it still ran somewhat rough (seemed ok at the time, but that was before I knew how it was supposed to run) and it failed emissions horribly... the CO level was at least 10 times higher than normal, and HC was about 3 times higher than normal.

3) My O2 sensor was dead, causing it to run pig rich (around 12:1 A/F at light throttle; should be 14.7:1). This is the reason that it failed emissions, and it also caused the engine to run less smoothly, with distinct vibrations at certain spots (namely 1800 rpm), which triggered problems 1 and 2.

M758

Glad you got it all fixed up. Sadly most 944 spec cars will have issues like this for a sometime as you "clearout" all the old issues. The good news is these should not be a problem any more!