Help diagnosing my problem
Discussion
Hi guys, this is my first problem post on this forum. I have had a good search
Symptoms are:
Delay of a good second or so between pushing the pedal and the engine responding.
Running very rich (to the point that it blasts soot all over the garage door and other cars from 2m away!)
Sometimes stalls by itself when started cold (so need to hold down the throttle for a few minutes)
The main problem is that the car goes through phases for running fine and doing the following:
Dip the clutch and the engine stalls - revs just drop to zero.
When i put my foot down midway through the rev range the car just kangaroos violently. Its not so bad in a low gear at high revs or in a high gear at letting the torque move the car but accelerating normally i.e not racing away but not slow either it violently stops and starts and stops producing power.
What i don't get is that this happens for a period of time (minutes - hours) but then will be fine again for a silimar amount of time. Stopping the car when pulling over and shutting it down can either cause it to start happening when it wasnt before or stop happening when it had been up until that point.
I appreciate that as i got my car cheap (just over 2k) that there could be more that one part at fault but things like the throttle delay arent my main concern, just stopping it kangerooing!
I have had a look round the forum and have found a few things that could be the problem. Throttle pot, coil, various sensors. The guy who sold me the car suggested the throttle pot and i had a mate look round it and he said that the leaking exhaust manifold could be counfusing the lambda sensor (which i am going to bodge repair tomorrow to see)
Its the intermittent nature of my problem which i though might stand out as a particular part failure.
Off to a garage soon as i have little mechanical skill but it would be nice to have an idea as it seems the previous owner had been trying to sort the problem already and failed eg a brand new dizzy.
Car is an 1990 S3 2.9 btw
Massive thank you to for any help, your knowlege could save me hours trying to find the problem.
Symptoms are:
Delay of a good second or so between pushing the pedal and the engine responding.
Running very rich (to the point that it blasts soot all over the garage door and other cars from 2m away!)
Sometimes stalls by itself when started cold (so need to hold down the throttle for a few minutes)
The main problem is that the car goes through phases for running fine and doing the following:
Dip the clutch and the engine stalls - revs just drop to zero.
When i put my foot down midway through the rev range the car just kangaroos violently. Its not so bad in a low gear at high revs or in a high gear at letting the torque move the car but accelerating normally i.e not racing away but not slow either it violently stops and starts and stops producing power.
What i don't get is that this happens for a period of time (minutes - hours) but then will be fine again for a silimar amount of time. Stopping the car when pulling over and shutting it down can either cause it to start happening when it wasnt before or stop happening when it had been up until that point.
I appreciate that as i got my car cheap (just over 2k) that there could be more that one part at fault but things like the throttle delay arent my main concern, just stopping it kangerooing!
I have had a look round the forum and have found a few things that could be the problem. Throttle pot, coil, various sensors. The guy who sold me the car suggested the throttle pot and i had a mate look round it and he said that the leaking exhaust manifold could be counfusing the lambda sensor (which i am going to bodge repair tomorrow to see)
Its the intermittent nature of my problem which i though might stand out as a particular part failure.
Off to a garage soon as i have little mechanical skill but it would be nice to have an idea as it seems the previous owner had been trying to sort the problem already and failed eg a brand new dizzy.
Car is an 1990 S3 2.9 btw
Massive thank you to for any help, your knowlege could save me hours trying to find the problem.
My first thought was throttle pot - might be worth swapping for a known good one just to see.
The cold start idle problem on its own, could be the idle speed control valve, but given the combination of symptoms, it's more likely to be something else.
You don't say, but given that it has a lambda sensor I assume it's an S3C? My S3C had a leaky exhaust manifold for ages (just a slight blow) but there was no evidence of this affecting the lambda sensor or the running of the car - it just sounded rubbish. I suppose that depends how bad the manifold leak is.
It does sound more like a fuelling problem than an electrical problem. This might be related to the exhaust leak (if the leak is serious, and between the lambda sensor and the cylinder head) or it might be something like fuel pressure?
The other thing that comes to mind is a leak in the inlet manifold or the various rubber tubes, especially to the MAP sensor (a wee square black box on the bulkhead beside the clutch reservoir). The ECU uses the lambda output, the throttle pot position and the MAP (Manifold Absolute Pressure) to determine where it should be on the fuel map. There are 2 or 3 wee rubber pipes joined at a T-piece down behind the dizzy - make sure none of them are perished or nipped or folded shut.
The cold start idle problem on its own, could be the idle speed control valve, but given the combination of symptoms, it's more likely to be something else.
You don't say, but given that it has a lambda sensor I assume it's an S3C? My S3C had a leaky exhaust manifold for ages (just a slight blow) but there was no evidence of this affecting the lambda sensor or the running of the car - it just sounded rubbish. I suppose that depends how bad the manifold leak is.
It does sound more like a fuelling problem than an electrical problem. This might be related to the exhaust leak (if the leak is serious, and between the lambda sensor and the cylinder head) or it might be something like fuel pressure?
The other thing that comes to mind is a leak in the inlet manifold or the various rubber tubes, especially to the MAP sensor (a wee square black box on the bulkhead beside the clutch reservoir). The ECU uses the lambda output, the throttle pot position and the MAP (Manifold Absolute Pressure) to determine where it should be on the fuel map. There are 2 or 3 wee rubber pipes joined at a T-piece down behind the dizzy - make sure none of them are perished or nipped or folded shut.
Edited by tvrgit on Wednesday 3rd January 19:58
I am no mechanic but some of your symptoms are similar to what i was experiencing when my Air Flow Meters were playing up. Some days I would get in the car drive for an hour no problems, other days it would play up going 5 mins down the road to get a paper.Sometimes it would start kangarooing and then stop ,or I might switch off and it would correct itself , even called the RAC one day as it was undrivable , the guy jumped in the car and drove off , had no problems ,(still took it round the block a second time, I think he was just having fun)
I have read on here people, clean them up , fit second hand ones or fit new ones (very expensive)or even change the system .
Hope this helps .
I have read on here people, clean them up , fit second hand ones or fit new ones (very expensive)or even change the system .
Hope this helps .
The S3C had catalytic converters but is otherwise identical to the S3 (it also had hydraulic tappets but you can't tell the difference without pulling the engine to bits). The 3 obvious differences are
1. The S3C had lambda sensors and the S3 doesn't. The lambda sensor looks like this:
The sensor is in the bottom centre of the pic. It should have two - one in each manifold.
2. The S3C has (or had) catalytic converter. See in that pic, the far away exhaust downpipe? See the darker (older) section about a foot long that connects the manifold to the J-pipe under the sump? That is a replacement pipe for the catalytic converter - again there was originally a cat in each side. If your car has a separate pipe section like that, or the cats are still in place, then it's an S3C. If the manifold joins straight to the exhaust system that goes back under the car, then it's not.
3. The S3C has no air flow meters. There are two large corrugated tubes that go from the throttle butterflys (the two round air intakes top centre of that pic) forwards to the foamair filter, which is mounted on the nearside of the chassis. If those pipes go straight to the air filter, it's an S3C. If there are two square metal and plastic boxes (the air flow meters) with wires coming out, then it's not.
Have a look and let us know what you find.
1. The S3C had lambda sensors and the S3 doesn't. The lambda sensor looks like this:
The sensor is in the bottom centre of the pic. It should have two - one in each manifold.
2. The S3C has (or had) catalytic converter. See in that pic, the far away exhaust downpipe? See the darker (older) section about a foot long that connects the manifold to the J-pipe under the sump? That is a replacement pipe for the catalytic converter - again there was originally a cat in each side. If your car has a separate pipe section like that, or the cats are still in place, then it's an S3C. If the manifold joins straight to the exhaust system that goes back under the car, then it's not.
3. The S3C has no air flow meters. There are two large corrugated tubes that go from the throttle butterflys (the two round air intakes top centre of that pic) forwards to the foamair filter, which is mounted on the nearside of the chassis. If those pipes go straight to the air filter, it's an S3C. If there are two square metal and plastic boxes (the air flow meters) with wires coming out, then it's not.
Have a look and let us know what you find.
thanks again
Ah yes pretty sure its a S3C then - has 2 lambda sensors and the plastic pipes go straight from the air filter to the inlets on top of the engine.
On a side note in interested because i read that all S3c's have an lsd as standard.
mycroft's problems sound similar to mine but as i dont have an air flow meter, do the lambda sensors do the same job? would replacing those be an idea?
I have also had a look round the tubes to the map sensor which seem to be ok.
Ah yes pretty sure its a S3C then - has 2 lambda sensors and the plastic pipes go straight from the air filter to the inlets on top of the engine.
On a side note in interested because i read that all S3c's have an lsd as standard.
mycroft's problems sound similar to mine but as i dont have an air flow meter, do the lambda sensors do the same job? would replacing those be an idea?
I have also had a look round the tubes to the map sensor which seem to be ok.
The lambda sensors don't quite do the same job.
Basically (and this a quick explanation, not a bleeding manual, before anybody else jumps in to say "ah but there's also this bit"
on the non-cat 2.9 cars, the ECU works out the fuelling based on the amount of air being drawn in (measured by the air flow meters), and the throttle position (measured by the throttle pot). Obviously engine temperature etc also come into play, but assuming everything else is equal...
On the cat-equipped cars, the throttle pot does the same job. The MAP sensor measures manifold depression as a proxy for "engine load" while the lambda sensors measure unburnt fuel in the exhaust. There is no direct measurement of the amount of air going into the engine. The required position on the ECU fuel map is basically determined by a need to minimise unburnt hydrocarbons while catering for engine load and throttle position.
So although they don't do the same job they could well lead to the same symptoms.
Lambda sensors are not cheap though (about £60 each IIRC - I thought I was going to have to buy one because I couldn't get one of mine out when I replaced the manifolds - the price encouraged renewed efforts with more penetrating oil and a bigger spanner!) so I wouldn't suggest replacing them unless you know for sure that is the problem.
I am still more inclined to suspect an air leak on the inlet manifold or rubber tubes etc.
I have just thought of something else though - the engine wiring loom lies along the top of the right-hand cylinder bank, between the rocker cover and the fuel injectors. There is a connector in there which is very susceptible to water ingress, and can lead to various inputs from sensors (including the throttle pot, ISCV and engine temperature sensor) becoming confused. I think the coil feed and distributor signal to the coil is also there. Make sure the internal connections are clean and dry, especially if you have either jet-washed the engine or been driving in the current pishy weather.
Basically (and this a quick explanation, not a bleeding manual, before anybody else jumps in to say "ah but there's also this bit"
on the non-cat 2.9 cars, the ECU works out the fuelling based on the amount of air being drawn in (measured by the air flow meters), and the throttle position (measured by the throttle pot). Obviously engine temperature etc also come into play, but assuming everything else is equal... On the cat-equipped cars, the throttle pot does the same job. The MAP sensor measures manifold depression as a proxy for "engine load" while the lambda sensors measure unburnt fuel in the exhaust. There is no direct measurement of the amount of air going into the engine. The required position on the ECU fuel map is basically determined by a need to minimise unburnt hydrocarbons while catering for engine load and throttle position.
So although they don't do the same job they could well lead to the same symptoms.
Lambda sensors are not cheap though (about £60 each IIRC - I thought I was going to have to buy one because I couldn't get one of mine out when I replaced the manifolds - the price encouraged renewed efforts with more penetrating oil and a bigger spanner!) so I wouldn't suggest replacing them unless you know for sure that is the problem.
I am still more inclined to suspect an air leak on the inlet manifold or rubber tubes etc.
I have just thought of something else though - the engine wiring loom lies along the top of the right-hand cylinder bank, between the rocker cover and the fuel injectors. There is a connector in there which is very susceptible to water ingress, and can lead to various inputs from sensors (including the throttle pot, ISCV and engine temperature sensor) becoming confused. I think the coil feed and distributor signal to the coil is also there. Make sure the internal connections are clean and dry, especially if you have either jet-washed the engine or been driving in the current pishy weather.
Thank you for the advice.
It also has the habbit of sometimes for a few seconds getting practically no air into the engine at all, its stutters along slowly with thick black smoke comming out the back until it carries on as normal.
Interestingly since i patched up the exhaust leak on the manifold the kangarooing still happens just as violently - but occurs much less often.
It also has the habbit of sometimes for a few seconds getting practically no air into the engine at all, its stutters along slowly with thick black smoke comming out the back until it carries on as normal.
Interestingly since i patched up the exhaust leak on the manifold the kangarooing still happens just as violently - but occurs much less often.
Gassing Station | S Series | Top of Page | What's New | My Stuff