Still having GTT issues

[BZG Wagon]

Ever since I bought my GTT I have been having an issue with a knock (or pink) in the engine under full throttle from roughly 2500rpm to 4000rpm. It gets worse if I run it on anything less than 98 octane.

I was in Christchurch two weeks ago on holiday & my GF's uncle hooked it up to his scan tool & we took it for a drive (he's a qualified aircraft engineer who now specialises in fixing euro cars). He agreed that something is not quite right on the Caldina and printed off values from the sensor while it was knocking on our test drive. The first print off was taken right at the end of the pinking and the second was taken mid way though it:


He suspects that when the the throttle is opened the car is taking too long to retard the advance of the ignition. Unfortunately he's rarely ever works on non-euros so he said wasn't sure of what the normal operating parameters of a GTT would be. Does anyone know if the advance looks like its too advanced while the car is on boost? His opinion was that it should be around 15 degrees at that point in image 1 (where its sitting at 37 degrees).

His scan tool could not pick up any stats from the knock sensor so we couldn't see what that was doing.

The boost is recorded in kpa (kilo-pascals).

Originally I thought it had to do with an air temp sensor which was broken when the car had a mechanical check, but that was (dodgily) fixed & appeared to be working correctly on our test (25deg's while driving, 40 while at idle).

I have posted about this a few times, had it in and out and in and out of Toyota for months now and they do not know what's causing it & don't believe its out of the ordinary for a GTT to do this. I have never driven or known another 3SGTE to do this and I suspect the car shouldn't be behaving like this.

Any help would be greatly appreciated.

[gepsk8]

u checked u cam timing?

[BZG Wagon]

u checked u cam timing?

It was doing it before & after I had the cambelt changed. A Toyota mechanic said the cam timing was checked on one other occasion.

[gepsk8]

has floged(don't know how to spell it) out the bottom crank pulley?

[fivebob]

37° is way too much for a 3S-GTE at 159kPa/3680 rpm, should only be about 26-28° according to the data I have. The 15° at 167kPa/4000rpm is probably in response to the know sensor, but I note these samples were obviously taken in a lower gear judging by the vehicle speed so they're not really a valid comparison of the symptoms.

You really need a realtime log of the ignition vs MAP and TPS to see what is happening, two isolated samples are not enough to diagnose why the ECU thinks 37° is the correct timing at this point, I doubt it's anything to do with "the car is taking too long to retard the advance of the ignition", ECU's don't work like that. It could be because of a faulty TPS or many other reasons, but I would need to see full logs to comment further.

[BZG Wagon]

37° is way too much for a 3S-GTE at 159kPa/3680 rpm, should only be about 26-28° according to the data I have. The 15° at 167kPa/4000rpm is probably in response to the know sensor, but I note these samples were obviously taken in a lower gear judging by the vehicle speed so they're not really a valid comparison of the symptoms.

You really need a realtime log of the ignition vs MAP and TPS to see what is happening, two isolated samples are not enough to diagnose why the ECU thinks 37° is the correct timing at this point, I doubt it's anything to do with "the car is taking too long to retard the advance of the ignition", ECU's don't work like that. It could be because of a faulty TPS or many other reasons, but I would need to see full logs to comment further.

From memory, 1. was taken in 2nd gear and 2. in 4th. Unfortunately I don't have a copy of the log we recorded (only the two printouts), but its good to know 37° is too much at that point in time.

Do you know of anyone in Auckland who would likely be able to solve this? Does it sound like something that's actually solvable - do mechanics encounter this type of problem often?

I've had the Toyota service department look into this 4 times (on one occasion they had the car for 4 days), a local auto electrician twice, a local mechanic once and obviously a quick look in Christchurch - I'm wondering if its just too complicated to solve or there is too much ambiguity in finding the problem.

Someone suggested finding an auto-electrician who can put it on a dyno and assess it while its running. Your thoughts fivebob?

[Trls250s]

I doubt the problem is unsolvable.

Question you have to ask is how does your ignition timing change when it comes on boost, and why does it change that much?

It maybe how it changes the ignition timing that its going wrong.

What generation 3s is it? 4th Gen?

Do these have a wasted spark system or do they run off a dizzy like the gen 3's?

[BZG Wagon]

I doubt the problem is unsolvable.

Question you have to ask is how does your ignition timing change when it comes on boost, and why does it change that much?

It maybe how it changes the ignition timing that its going wrong.

What generation 3s is it? 4th Gen?

Do these have a wasted spark system or do they run off a dizzy like the gen 3's?

Its 4th generation; '97 ST215 Caldina GTT.

It has coil packs so I assume it doesn't have a dizzy .

The issue I have is that I'm not a mechanic, I don't have the knowledge, skills or tools to resolve this and I have been unable to find a mechanic to diagnose the problem. This is costs me money every time I drop it off to get it diagnosed and its becoming quite frustrating.
I'm doing my best to narrow the problem down (searching the net, posting / browsing forums etc) in the hope I can get enough knowledge to discuss with them what specifically needs to be checked. So far dropping the car off and waiting for an answer doesn't seem to work.

The only other alternative is flicking it off to someone cheaply who might have the time and knowledge to fix it. It took me six months of hard searching to find this car, I paid a little bit over what they normally go for because it was in such good condition & I absolutely love everything about it but its really starting to get the better of me. I've spent 9 months of banging my head against a brick wall.

[Trls250s]

Have you tried resetting the ecu and getting it to relearn everything again?

I had a chat to my old man about this and we think that for the ignition timing to jump around (which will cause the pinking) either your base timing isnt set right, or the sensors that tell the ecu to advance/retard the timing are giving the wrong information.

You tried looking at your MAP sensor, and your CAS sensors to see if they are operating correctly?

Who set your base ignition timing? Have you tried resetting it to factory specifications?

Does your ECU throw any errors?

I know this is a gnarly problem and this might be a bit off the wall but it cant hurt to have a look.

[BZG Wagon]

Have you tried resetting the ecu and getting it to relearn everything again?

I had a chat to my old man about this and we think that for the ignition timing to jump around (which will cause the pinking) either your base timing isnt set right, or the sensors that tell the ecu to advance/retard the timing are giving the wrong information.

You tried looking at your MAP sensor, and your CAS sensors to see if they are operating correctly?

Who set your base ignition timing? Have you tried resetting it to factory specifications?

Does your ECU throw any errors?

I know this is a gnarly problem and this might be a bit off the wall but it cant hurt to have a look.

I reset the ECU yesterday & it didn't solve the problem.

I was told that the timing couldn't be set on 4th gen 3sgte, although a Toyota mechanic said he "checked the timing".

The ECU does not produce any errors (only ever the air temp sensor which was fixed months and months ago & Toyota erased the error from the ECU).

I'm not sure if my GF's uncle looked at the MAP or CAS sensors. He checked a few things on the car but went waaaay past anything I understood. Is there some way I can check the MAP or CAS sensor or do I need to hook it up to a scanner tool?

[BZG Wagon]

I should probably add: If I go 1/4 throttle through the rev range which causes the problem & then go full after 4000rpm it generally doesn't knock (or at least, I can't hear it).

[Trls250s]

Its a lot of work to check the MAP sensor i believe. You will need to find the resistance values that its ment to produce, take a tape of the resistance of it as the engine is running and then see if this resistance changes as it should as the engine goes through its rev range.

CAS sensor is easier, just have to make sure ur tacho reads what is mechanically true.

Have you tried borrowing another ECU to try?
I know ecus dont commonly break but its worth a shot.

I just read something interesting about the 4th Gen 3s and that is the Knock sensor should retard your timing if it senses pinking etc. Have you checked your knock sensor is running?

[BZG Wagon]

Its a lot of work to check the MAP sensor i believe. You will need to find the resistance values that its ment to produce, take a tape of the resistance of it as the engine is running and then see if this resistance changes as it should as the engine goes through its rev range.

CAS sensor is easier, just have to make sure ur tacho reads what is mechanically true.

Have you tried borrowing another ECU to try?
I know ecus dont commonly break but its worth a shot.

I just read something interesting about the 4th Gen 3s and that is the Knock sensor should retard your timing if it senses pinking etc. Have you checked your knock sensor is running?

The scan tool we used in Christchurch didn't register the knock sensor so we couldn't check it (not sure why we couldn't see it though).
My GF's uncle thought it could have something to do with (or associated with) the knock sensor. I would have thought that would produce an error code though.

Checking the ECU & MAP would probably be the thing to look at after checking the knock sensor? I've never heard of an ECU or MAP sensors actually fail? Does it really happen?

[fivebob]

either your base timing isnt set right,

Wouldn't matter what the base timing is, the ign timing as seen by the scan tool is relative to the base timing anyway, and 37° is too high under these conditions.

or the sensors that tell the ecu to advance/retard the timing are giving the wrong information.

This is the only possible cause unless the ECU is faulty or programmed that way.

You tried looking at your MAP sensor, and your CAS sensors to see if they are operating correctly?

There's probably no point in checking the sensor, the ign timing values that the ECU is generating for the boost/RPM level are wrong, so there must be something else that is causing the problem, either another sensor like TPS has an invalid value or the ECU is faulty.

It also does not matter if the CAS is wrong, the ECU wouldn't generate the wrong ign timing value if the CAS was incorrect, as all values would be relative to the measured crank angle.

Have you tried borrowing another ECU to try?
I know ecus dont commonly break but its worth a shot.

That would be my next test if you can get hold of another manual ECU.

I just read something interesting about the 4th Gen 3s and that is the Knock sensor should retard your timing if it senses pinking etc. Have you checked your knock sensor is running?

All 3S-GTEs will do this, but they shouldn't advance the timing to 37° under those conditions regardless of the knock sensor signal. Stock timing should be around 26-28° under these conditions.

This problem is solvable, try swapping ECUs first, but if you can't do that, or it makes no difference, then to solve it you need information from all the relevant sensors. The only way to determine if it's a sensor fault is to log all the sensors that have an input into timing from 500rpm before the problem occurs, to 500rpm after it stops. You would need to do this several times to get good info and also need a run(s) that didn't exhibit the problem to compare the sensor values with.

[Trls250s]

Good input Five bob, i dont know much about 3sgte's im a 4age kinda guy.

I was just trying to think through what could happen and why things would be happening.

[gleem]

i have a spare ecu you could borrow but its tiptronic

[cedwards]

BZG in my Dash post you say yours is doing similar things, now mine never did pink badly but I did on the odd occasion think I heard it for a fraction of a second, now I know nothing of the ecu layout or how the sensors run from engine to ecu but is it possible an intermittent dash fault or possibly increased resistance on the dry joint could be mis-guiding your ecu into advancing the timing excessively?? (Long shot I know)

[solitaire]

My mate had a manual beams.... did exactly the same thing, took it to two different toyota shops who could not figure it out... ended up flicking it on.