Mk3 2.0 8v Gti ADY intermittent engine failure problem- Help!

Hi All, my 1995 Mk3 2.0 8v Gti has developed an intermittent problem recently.

On the first occasion after driving for a few miles and accelerating off from a junction, the car had a sudden jerk when the engine appears to stop but restarts itself virtually immediately. This whole event lasted a second or so then the car continued running perfectly to my destination a couple of miles further on. After a break of about half an hour the car started as usual but the engine failed completely about 1 mile down the road and would not start again. On checking under the bonnet it was apparent that I had lost virtually all of the coolant via a pin hole leak in the metal coolant pipe feeding the heater. Ah I thought no wonder! Towed car home where it started again first touch. Obviously turned it off immediately, replace damaged coolant pipe, refilled with coolant and a few days later the car seemed back to normal.

All well for a couple of weeks, used daily with no problems at all then yesterday the sudden jerk problem happened again following a pull in top gear up a hill. Car then drove fine to destination a couple of miles on, parked up but on return to car, about an hour later, it would not start. Tried several times, engine spins fine on starter, all dash lights on ok, but no sign of life. Then after about five minutes on fourth try it started immediately I turned the key and ran perfectly for the rest of the day, including several stops and starts. All well again today with no signs of the problem.

Has anyone experienced this type of problem? I wondered if I have damaged one of the sensors when the overheating happened, or if the ECU relay 30 or Fuel Pump relay 167 giving problems. If I bought a code reader, would this help identify this type of intermittent problem, if so which reader would you recommend.

Hi,
Yup my 1997 mk3 has just started doing the same thing !

After years of uninterrupted service the little mutha started cutting out indescriminately. Not the end of the world but a real pain 'cos all of a sudden, whilst happily driving along, you no longer have servo assisted steering & brakes.............d'oh !

Gave up and took it to garage. They diagnosed "sludge" in the throttle body and cleaned it out. So far so good but it's only been a few days.............i'll let you know.........!!

Quote:

---

Originally Posted by fablon

Hi, Yup my 1997 mk3 has just started doing the same thing ! After years of uninterrupted service the little mutha started cutting out indescriminately. Not the end of the world but a real pain 'cos all of a sudden, whilst happily driving along, you no longer have servo assisted steering & brakes.............d'oh ! Gave up and took it to garage. They diagnosed "sludge" in the throttle body and cleaned it out. So far so good but it's only been a few days.............i'll let you know.........!!

---

I had this happen (intermittent shuddering, failing to start and then "self-healing"), over the last fifteen/15 months to my wife's 98 Variant 3, it helped that I have a Vag-com. When it happened, I'd scan and store results and when it became more frequent and finally, the car would not start, so then going over the logs from the last year, I noticed that the engine ECU always seemed to disappear and wouldn't respond"NO RESPONSE FROM CONTROLLER".

Forum searches brought up the usual, throttle position sensor/throttle needing cleaning(I did this a year ago), HT leads, Dizzy and ignition coil, fuel filter,fuel pump, MAF sensor, Crank shaft position sensor, Coolant Temperature Sensor etc . . . Found one where unlike the others, a guy specifically mentioned something different, the ECU/Engine Electronics relay. Because of my scan results, I zeroed in on the ECU electrical line, fuse was good and ECU was most expensive so I started out replacing the engine ECU relay (number 30).

Hope this helps,


Some say they get a "dry joint" which in plain english is a loose connection when solder gives way, other theory is the constant click on and click off and arcing/sparking, causes soot build up on the contacts. Anyway roughly 10 UkPounds or 15 US$ later and you should be sorted. Hope this helps.

thanks Kafuluzed, mine's started cutting out again having come back from the garage after the aforementioned " throttle body cleaning" procedure. I'll take it back and ask them to check over the items you have identified,
cheers,

fablon

Hi fablon,if you have had the "throttle body" cleaned then you need to get it re-aligned using a proper VAGCOM scan.The cheap scanners available are a waste of time and money,they "won't do the job".My N reg 8v with the ADY engine used to cut out and turned out to be faulty "crank sensor".Got that fixed but kept cutting out or stalling when engine got warm so changed relay 30 and 167 and everything is now fine.

Hi everyone,

sorry to have taken a while to reply. I had an interesting (don't know how else to put it) Easter weekend.
Had a scheduled on-line exam that weekend (anytime at my convenience for 12 consecutive hours between Thursday and Tuesday). I planned to hop from studying to playing with my 1 year old baby daughter, sorting out my wife and sitting the exam Saturday between 9pm and Sunday morning.After this off course, these two ladies would have me to themselves.

Day 1, Good Friday Morning:
On study break, I was done giving the "scarlet" lady:zx11: (who is I suppose lady number 3 ), the wife's MK3 Variant/Estate, AGG, 8V, 2.0L, a thorough wash (engine,undercarriage,interior and body) when after everything it wouldn't start.

Cranking was fine(starter turned engine), but no catching(This automatically eliminated the starting system).

I blew air (from my trusty blower) in to all the usual sensor connections and proceeded to use WD-40 as I have done over the last 17 years on my own MK2 8V GTI but she wasn't having it.

Thinking my 5 week old (and mark you), used ECU relay number 30 replacement gremlins were back or maybe now it was the fuel pump relay number 167, out came the Vag-Com.

Unlike before however, the ECU was responding to the VAG-COM scanner and none of the vehicle controllers showed any errors. So I knew it was either the fuel or ignition/sparking system.

The output tests for the fuel system in Vag-com made the fuel relay click(I could hear it in in the relay box) and
the fuel pump purge(I could hear this in the boot/trunk).
You can test this system yourself by removing a spark plug from the block. In my case, plug number 1(the one nearest to the pulleys and belts) confirmed this as it was wet and smelt of raw petrol.

Day 2, Saturday:

Couldn't concentrate on studying and mind wondered;

So the trouble must be the ignition system I figured.

Openned up the steering assembly going for the ignition rotor (the one the key turns when you turn it in the steering wheel lock cylinder).
Continuity tests proved that it was fine.

I took off one HT lead and with a screwdriver touching it's center to the cylinder head, cranked,but would not get a spark.

I swapped the entire cap and 5 cables from my trusty old MK2 but with the same results.

I then figured it must be the coil pack, tested the input signal cables from the ECU.
The two outer contacts gave the battery's 12v with ignition "ON" so I figured that the signal to send spark (which comes down the third contact in the middle could also be missing. As this required complex testing equipment/knowledge of electronics, I decided another elimination test would be easier and took the coil off to a breaker's yard (40km and a 1hour minibus taxi ride away) to test it on his car(it's identical to the wife's, although not as pristine ). His car fired up instantly:confused: with my suspect coil, so that idea went out the window. He suggested it was the crankshaft/engine speed sensor. I protested seeing as he sold me a new one approximately 5 weeks earlier, unless it was a crap part. He shrugged his shoulders and said I should take a wiring specialist. I then said to him I was more than capable of diagnosing the wiring issues myself but wanted to bounce ideas off a few colleagues.


Day 3, Easter Sunday:

Didn't start exam, wasn't faring well on the practice tests, attention diverted to car.

Found a thread on a Golf 4 which wouldn't start after washing the engine and his problem turned out to be with earthing connectors. I took the battery "tray" off, checked these ones (which were clean) but sprayed WD-40 anyway, still the same. The car cranks but won't start.

Checked and WD-40d all the visible Earth connections I could find, but no dice crank, no spark.

• Went after MAF sensor connector and intake air temperature sensor connector (after getting my first and only VAG-COM error related to this incident all weekend) about it all weekend with wd-40, CRANK, NO SPARK

• Checked throttle body (was still clean a year ago from last time), CRANK, NO SPARK

• Throttle position sensor didn't get my water "blast", so I ignored it.

• Played with wires from MAF sensor and fresh air box into firewall but, CRANK, NO SPARK

• "Maybe timing belt has snapped?" as on these cars it needs to be changed every 40k km. Cover showed previous owner had done it at 39,300 or something and I had bought it at 69,000km, and currently it was at 81k on the second belt. Removed the timing belt cover and cranked.Belt was in one piece so the next thought that it had probably slipped and lost timing. Decided to revisit this later.

Anyway, after more research I found out that,
a) The ECU sends a signal to the coil pack controller integrated circuit down the middle line (there are 3 lines connecting to the IC on the mk3 coilpack) and this then sends voltage up to the coil itself to send out to the dizzy and plugs. No spark signal from the ECU, no spark out of the coil.

Continuity tests showed all wires had good connections to the ECU. The wires could have broken copper cores inside that kept being "on" or "off" depending of whether I had stretched things out or returned them back to their original "snaking" routing or the problem was elsewhere (unless the ECU was acting flaky).

b) The ECU determines when to send the spark signal based on what position the crank shaft is at. It figures this out using the signal from the crank shaft/engine speed sensor. No sensor signal (to the ECU this probably means the engine is not rotating so why send a signal to coil to spark?

This signal being in the 5v range, my experience from PCs (where a short of a 3.3v line with a 5v or 12v line "kills" components) led me to suspect that the water from the engine wash had somehow shorted the sensor 5v circuits to the 12v ones of the rest of the engine either "killing" or "numbing" the sensor or causing the ECU to perceive "no crankshaft movement"

c) You can test the crankshaft position sensor if it is out of the engine block. Turn ignition to on but don't start the car. Wave a piece of metal(I used a screwdriver) across the area that get's inserted into the crank case.
There should be all sorts of whirrs, whizzing and clicking in the engine area (as the ECU figures the crankshaft is rotating and sends all manner of instructions to fuel injectors,the coil IC,switches and even more sensors and motors.

I tried to crank the car and just as I was giving up, it started. I let it idle while I bragged to my brother on the phone how I had "figured it out". Only to come back 10 minutes later and find it had stalled. Cranking gave me the same results (CRANKING, NO SPARK)

Figured I would sit the exam that night/Monday morning rest at night, head off to work on Tuesday and maybe implement on Wednesday(when I would take time off)

Day 4, Easter Monday:

Didn't do exam, figured I would do it in the evening and woke up to study for it.

Late in the afternoon I was back on the car(couldn't get it out of my mind)
My old crank angle sensor produced the desired response, (half an hour, 6 nuts/bolts, two jacks -1 front engine mount/starter bracket and an allen bolt later ) on removing the 5 week old "new" one from the engine block, it wasn't as sensitive.
I literally had to make the screwdriver rub/tap it gently before the engine components would respond in the test c) above.

Because the old one had it's insulation plasticized(the outer insulation and that of the inner 3 wires was cracking, flaking and falling off every time I touched it), I had replaced it 5 weeks earlier(when relay 30 turned out to be the culprit) as a matter of reliability.Didn't want my 1 year old daughter and wife calling me to rescue them:1zhelp:, at the most in opportune time.

I was thus shocked that with the old one taped up and put back in, the car started almost immediately, cutting out randomly after idling between 5 and 10 minutes.I figured that on both the ends where the 3 wires came back together,some shorting was taking place and decided to live with that until I can get something new of equivalent quality. The original is marked Siemens, France on the sensor portion, while the replacement came in a box that said Schneider Autotek, but with no markings on it and with two holes on the "sealed" part of the sensor, was probably Chinese/Asian in origin. Could be these holes let water into the sensor portion.

Tuesday, post Easter weekend (Wife and daughter 0:aargh4:, Scarlet lady 4 days)

The car was ok for the first 10 - 15km test. While driving:drive: the 40km to the aforementioned breaker to get a better sensor or my money back, it cut out thrice in the first 10km. Each time I got is to start again simply massaged the connector end to keep it straight and the three wires away from each other.He gave me the run around so I have decided to drive like this and won't be giving him my cash, will order on-line from now on.
I have done 200+ km since and problem is "under control" as no cutting out has occurred since that day(APR 2nd I believe it was).

Bottom line, it's a matter of elimination, patience and experience. If your car is made after 95, you need a VW scanner, go for a genuine HEX+CAN, ROSS-TECH, VAG-COM for around US$300, (not a knock off cheap copy) or find a mate with one that will, scan your car for free, in exchange for a pint, or a reasonable fee. Your better off going to the dealer for a scan (if you can't do the two above) then getting an experienced mechanic outside to "interpret" the scan results, than using a knock off/cheap scanner.

A mate or mechanic with the same model can save you buying a part if you can swap suspect parts one by one for those on theirs. Once you confirm the source of your problem, you will need to purchase one, not many parts.

Just remember dealerships, are businesses not charities and will gladly spend a lot of their time (labour):bigok: and/or replace (even when yours are functional)a lot of parts in exchange for a lot of your money.