GolfGTIforum.co.uk
Model specific boards => Golf mk3 => Topic started by: dazmatic on 16 February 2013, 18:12
-
Basically, I've been having some seriously weird problems with my 16 valver.
I had previously asked over at VWVortex but I'm getting nowhere there so what I will do is brief over everything.
Car started hesitating ~2k RPM, even when cold and just pulling out from the road where I live.
Tested TPS and not a single dead spot on it, works a treat.
Then went on to change the CPS and still no change.
Swapped out the coil for another, still no change
Throttle body also cleaned
The problem seems to have changed however from a hesitation at ~2k RPM, to no hesitation whilst driving at all, but once temp reaches high enough for the fan to turn on, the engine cuts out intermittently, as in, it'll cut out but just before it dies completely recovers itself and will eventually cut out completely.
If I then leave it for a few seconds, it'll start and run fine until of course the temp gets high again.
So, since then I have also changed ECU relay and fuel pump relay for known working pulls from another car.
Then changed the fuel filter and no change and today changed the fuel pump and no change.
However, I noticed something particularly strange.
Whilst having my VAGCOM on the car and using the measuring blocks and the car running, as usual it got up to temperature and started intermittently cutting out like usual, but when it did I lost communication with the VAGCOM, as if the ECU was switching on and off sporadically.
Now, the ECU switching on and off would explain the cutting out but what supplies the ECU with it's power?
NOW FOR THE WEIRD BIT
Obviously, first thought was the ECU relay failing (again?) so I decided to pull the ECU relay and replace the contacts with a pair of spade terminals soldered via cable to a switch to simulate the ECU relay contacts opening and closing.
When running the car on the drive, I flicked the switch off and the car continued to run! Eventually it got up to 'temperature' and started cutting out and recovering like usual so and I could sit there and flick this switch all I like simulating the ECU relay contacts and it made no difference.
It's like the ECU is parasitically getting power from elsewhere.
Now, I am aware that in a previous life with a previous owner that it did have an alarm/immobiliser fitted...
Oh my god, could it be the immobiliser failing?!
-
My car was doin this, changed the coil, still doing it, new ht leads sorted it.
-
Oh yeah, leads were replaced maybe 10k miles ago for new, as was the coil, plugs etc.
-
Ah right. Can't help you then sorry, might be worth pm'ing DanChave as he has had a lot of running problems with his and he might be able to help
-
There must be a reason as to why my car can run without the ECU relay in it?
How does the immobilizer work?
I've got a Laserline immobilizer but I'm not sure whether that would cut the ECU out entirely or just prevent the car from starting?
-
defo pulling number 30 relay?
-
Definitely relay 30! I have 2 of them, original one from the car and the 2nd hand pull and have both in my hand!
-
Ok,
Now it's getting EVEN weirder... just went out to the car.
The car will now start without even putting the immobiliser keyfob into the slot as in, the immobiliser doesn't even work now.
EDIT:
Just been outside to the car again as when I was looking at the relay box the other day, it was a complete birds nest behind it!
I had noticed the other day that there was a black cable with an inline fuse just handing there, so, thinking that it was for the old scorpion alarm system which is no more, snipped the loose cable off.
There was also a few other snipped cables behind there, which, look snipped I.E. no exposed ends.
There was one however, a black cable, that did have an exposed end and it didn't occur to me this morning thinking about this immobiliser malarkey that maybe perhaps it had something to do with it.
Thinking that car started even without the immobiliser and no flashy red light on the fob front, lead me to think that the immobiliser wasn't actually working.
So, I went out there with a multimeter and tested between the black cable I snipped and this black cable that had the exposed end. Turns out there was 12V between them, so I figured theres a fuse, whats the harm in connecting them.
So I rewire this fuse end to the stripped cable that comes out from the back of the fuse box and the other end of the fuse to the cable I snipped. I gingerly, placed the fuse in the holder and screwed the 2 together successfully with no pops or bangs.
I look up, turn the key to start and nothing! So the immobiliser is working again, but, the car will still start without relay 30 BUT only after placing my keyfob in the holder.
I'm thinking that perhaps the black cable somehow worked loose from the inline fuse turning off the immobiliser and that perhaps, parasitically, the ecu was drawing power from the immobiliser hence why it will still run without relay 30 in.
So, I will put relay 30 back in and take it for a spin and see if I break down!
-
So now it starts does it cut out still when up to temp?
-
Ok...
The immobiliser now immobilises, but, the original hesitation problem is back AND now when up to temperature it cut out completely, left me stranded on the side of the road again.
So, im going to scrap it next week! :(
-
Is the immobiliser part of the alarm system? as most are or can be wired in to the cooling fan to allow the fan to run on for a while after the alarm has been armed without setting it off.
-
Nah, it would look like there was a scorpion alarm system fitted by the dealership and then later removed as the alarm horn is still under the bonnet.
This immobiliser is a laserline and fitted sometime later.
Trouble is, I've just done a thorough test of both ECU and fuel pump relays and it is neither of them. I've also pulled the aforementioned fuse out again and went for a spin.
It still kept hesitating again and then eventually cut out and died when it reached 86oC on the oil temp gauge.
I've decided I've wasted far too long trying to fix this now and am going to either sell it spares/repairs or scrap it.
Can't be dealing with it anymore.
-
Well, I've completely changed my mind.
I've put my polo up for sale instead and im keeping the golf for a bit longer to try and fix it.
Is it possible that the new crank position sensor is faulty?
Is there a way of testing it? I've got an oscilloscope but not sure whether any fault would show up on it if I tested it.
-
Crank sensor worth a check. They become wet and corrode over time resulting in cutting out etc. also, check ecu earth under battery tray. Make sure wire isn't black inside as water gets under tray and burns wire inside out. If this is the case, you can re wire an earth back to the engine block.
-
As above - I had issues with corroded wires under my battery tray too.
Can you possibly tell if you are getting fuel supply when the fault happens?
Check that your alternator isn't overcharging, it activates a over voltage protection thing inside Relay 30 which cuts power to the ECU to protect it. My engine would hesitate during it's few seconds of overcharging but it never cut out.
With your oscilloscope you should get a square wave pattern like an on and off then probably a longer square wave which represents TDC, i don't know what pins to test without looking tbh but if it were to fail at a certain temp then the pattern would be anything but a square wave. However when it's hesitating it could be like the sensor is failing so you might see something there.
If you are getting fuel supply right up to the injector rail when the car cuts out then you can rule out the pump/filter/relay 167.
But basically when the engine stops there's no spark so from a diagnosis point of view it's complicated because the crank sensor signal along with the hall sensor in the dizzy is used by the ECU to determine when to tell the coil to spark.
It's not unheard of for new crank sensors to fail, especially the cheaper make ones. That and the coil are known to fail when hot.
-
Hi there,
Haven't replied for a while, had given up on the car.
The tax is due at the end of the month so I either fix it or it goes.
Trouble is, I've now replaced the fuel pump for a known good one and I've tried to recreate the problem by pulling the relay. When doing this, the car cuts out BUT slowly, as the pressure drops.
The actual problem is more of an instant loss/regain, like ignition cutting out rather than slow death. It can't be the coil as it's stone cold and has also been replaced for known good.
Also, I don't think it's the hall sensor in the dizzy as I've disconnected it with the car running and the car carried on running! (didn't throw up any error codes either which was strange)
I've watched the signal on my scope from the crank sensor and your right in that there is a square wave pattern and a long wave for TDC. I watched the signal whilst the car was cutting out on the drive and the signal stayed a constant square wave.
I think it's safe to say that it isn't the CPS which means this second new one I've bought is redundant.
I'm completely lost now, not a clue!
-
Perhaps the ECU itself is to blame? :undecided:
-
I'm pretty certain its a temperature thing.
Soon as the oil temperature reaches 86/88 degrees Centigrade, that's when it starts cutting out. It will run lovely up until that temperature. I'm thinking that maybe perhaps it's something to do with oil temperature or something?
-
Engine temp sensor? I had the same problem changed everything then after bein told to change the crank sensor it didn't do it for a while then came back. Then changed the engine temp sensor and as soon as i changed that it never did it again ane that was 3 years ago.. worth a punt for less than a tenna part
-
Which temperature sensor was that and where is it on the block?
I know that there is two water temperature sensors, one for the dash and another for the engine to control the fan.
But what about oil temperature?
Is it the same case that there is 2 sensors for oil as well, one for the dash and another for the ECU?
I put my VAGCOM on it last night, no codes at all.
-
burn it :laugh:
-
Believe me, this car hasn't been used for 4 going on 5 months now, burning it has become an option!
-
Well, I have just replaced the crank position sensor with another new one.
This new one is also knackered! I fitted it and plugged it into the loom and the car doesn't start or fire at all now. No spark. Tested the new sensor with my multimeter and across the middle pin to one of the outer pins it measures 1 ohm, to the other pin it is 11 ohm so it would be safe to say that the sensor is knackered.
I also got the sensor I pulled off and tested that and there wasn't a reading between any of the pins as in higher than the 2 meg/ohm that my meter reads up to.
Funny thing is, I was curious about what type of sensors these are so I cut the thing open to get access to the electrics inside.
Turns out, there is a magnet and a circuit board inside the sensor that consists of 3 capacitors, a resistor and a 4 PIN IC. Funny thing is, I measured the connections from the cable directly where they meet the circuit board, across 2 of them I read 700ohm and between the opposite I read nothing.
Exactly what a new sensor with nothing wrong should be like!
Looks like the epoxy resin that the sensor is housed in is may be causing trouble as with all the epoxy removed, I'm getting a proper reading so something is going wrong with the housing.
I've just bought ANOTHER new crank sensor which will be my THIRD sensor and put a complaint into the supplier that supplied the faulty sensor I fitted today.
-
Just a quick update:
It's ALIVE!! Fitted the new sensor and it's working an absolute treat. Gave it a good thrash, got it to about 94 degrees centigrade oil temperature and it just kept on chugging!
I've never had so much trouble caused over a single sensor!
Still doesn't explain how the car still runs without an ECU relay though.