Intermittent Fault - No One Knows What It Is!

[benro2]

Hey guys, this is kind of a last ditch effort to try and diagnose an intermittent problem I've been having for the last couple of weeks.

It all started when I was having problems with the car stalling randomly. I was able to start the car again straight away, then one day it wouldn't start. Turned out to be a short in the wiring which was blowing the fuse to the fuel pump. Got that fixed, then I noticed another problem:

After driving the car for about 20 mins, not hard or anything, mostly normally in 80km/h highway conditions, I noticed the idle speed dropping to about 500rpm for a few seconds, then returning to normal. After this started happening, I would have a partial loss of power for a few seconds, then would return to normal. A minute or two later, same thing, while accelerating lightly. Came to a stop, car idles fine for a while, then idle speed drops really low for a few seconds, then back to normal.

I have had the car at the auto electrician for a few days now and he's brought in other people to have a look at it and no one can find what it is!

Incidentally, I have an ECUTalk unit hooked up and the injector duty cycle doesn't work anymore. The rest of the display is fine, it's *just* the IDC, and obviously L/100km as this is calculated from IDC. This used to work before I took it to the auto electrician's, and he is aware of it and has been trying to fix it but with no luck. He did hook up his scan tool but I don't think it's talking to the ECU properly - however I was in the car the first time he did this and the IDC was displaying fine straight afterwards, so I doubt it's that.

I know it's extremely hard to diagnose a car from a bit of text but am just hoping that someone out there may have had the same problem and happens to know what it may be? Any help is much appreciated!

[benro2]

I found another clue! I think it's the AFM...

The car did the same crap tonight. Only this time I'd been driving it for 2 mins, wasn't even up to temp. I was pulling away from the lights and as soon as I was in about 3rd gear I just lost pretty much all power. I held the throttle in the same position and the boost gauge was displaying 0 psi and crept up to 2 psi (pretty sure my foot didn't move). I didn't get any more power - in fact, I think the car was pretty much not accelerating at all!

It was OK for most of the journey home on the M4 but just before I got home I jabbed the throttle slightly as this is what seems to set the problem off. I noticed that after I'd stopped the AFM voltage was about 1.2 - 1.3V (correct at idle), but then the car started dropping revs and almost stalled - but as it did this the AFM readout went up to around 1.8V. I couldn't see out the back of the car but apparently last time it did this it was blowing a lot of black smoke.

So this leads me to believe that the AFM is telling the ECU it's running a lot more air than it really is which is richening up the mixture a LOT and killing power. Sounds about right?

I actually replaced the AFM a while ago with another 2nd hand one and it was running fine for a long time. However, both this problem *and* the stalling problem (blowing fuses) happened around the same time. Is it possible there's still a short somewhere, or possibly the short from before has damaged the AFM? How can I test this?

Also, would my broken injector duty cycle readout have anything to do with all this?

Edited April 7, 2011 by benro2

[DVS JEZ]

Try another afm fro a known working car. See how that goes

[mosquitocoils]

Could you be losing fuel delivery at some point? Maybe bad electrics to the injectors or in that area?

Probably not it, but worth investigating. If your fuel supply is dropping out, your car will splutter and die, won't it?

[benro2]

Try another afm fro a known working car. See how that goes

Yeah, just gotta find one first Taking the car back to the auto electrician's on monday (3rd time LOL) and he's gonna run some tests on my AFM. Hopefully he will be able to tell if it's stuffed or if it's the wiring somewhere.

Could you be losing fuel delivery at some point? Maybe bad electrics to the injectors or in that area?

Probably not it, but worth investigating. If your fuel supply is dropping out, your car will splutter and die, won't it?

Yep, it could be this, however I would have thought AFM because of the increased voltage when the problem occurs. I don't think the fuel pump failing would cause a change in the AFM voltage...?

And yes, when the fuel supply stops, so does the car I did it the other night in the Supra by accident when the tank was nearly empty - started boosting up a steep hill and suddenly my head nearly went through the windscreen! Fuel pickup must be near the front of the tank

[EXEMPT]

The AFM will be tripping out due to the lack of fuel supply...

Fuel Pump Relay

Fuel Pump

Injectors/Injector Power Supply

Fuel Filter

O2 Sensor

Edited April 13, 2011 by EXEMPT

[3LGODZILA]

check basic sh1t, could be a bad earth?

[EXEMPT]

Yeah, just gotta find one first Taking the car back to the auto electrician's on monday (3rd time LOL) and he's gonna run some tests on my AFM. Hopefully he will be able to tell if it's stuffed or if it's the wiring somewhere.

Yep, it could be this, however I would have thought AFM because of the increased voltage when the problem occurs. I don't think the fuel pump failing would cause a change in the AFM voltage...?

And yes, when the fuel supply stops, so does the car I did it the other night in the Supra by accident when the tank was nearly empty - started boosting up a steep hill and suddenly my head nearly went through the windscreen! Fuel pickup must be near the front of the tank

"I don't think the fuel pump failing would cause a change in the AFM voltage...?"

How sure are you coz the AFM is attempting to balance the air with the lack of fuel and consequently tripping the gosh darn out and the increase in voltage duty could be attributed to it working harder to attempt to conrol air intake?

I reckon your fuel pump relay has shat itself. Clean your IAC valve and AFM to eliminate variables with the air intake system.

Check your plugs, coils and ignitor as well as all your intake and intercooler piping, check your O2 sensor and the other shit I listed in my last post...

[benro2]

"I don't think the fuel pump failing would cause a change in the AFM voltage...?"

How sure are you coz the AFM is attempting to balance the air with the lack of fuel and consequently tripping the gosh darn out and the increase in voltage duty could be attributed to it working harder to attempt to conrol air intake?

I reckon your fuel pump relay has shat itself. Clean your IAC valve and AFM to eliminate variables with the air intake system.

Check your plugs, coils and ignitor as well as all your intake and intercooler piping, check your O2 sensor and the other shit I listed in my last post...

OK, you may have a point with the AFM voltage and fuel pump there, but I don't know how the ECU reacts if a fuel pump is failing. Also, judging from my last experience, my OEM pump just died suddenly, it didn't let go gradually. So I'm on my 2nd fuel pump, nothing fancy but not OEM, and that was probably 1.5 years ago, so it's highly unlikely to be the fuel pump itself failing. Don't forget, the AFM can't actually adjust the amount of air I'm receiving - all the ECU can do is try and richen the mixture, so if it's seeing what it thinks is too much air, it will try and add fuel. This is why it's running rich - it's seeing higher voltages (ie: what it thinks is more air), and adding extra fuel.

Fuel pump relay: I had a short in the fuel system and this has been fixed. They would have found this at the time - the problem I am having now is something separate. I have already cleaned the IAC valve and AFM!

Plugs were changed just before this happened. Coils are Splitfires and the series 2's run coil-on-plug ignition so no ignitor. There are no obvious air leaks but I'll leave that until last since it's so hard to check O2 sensor was replaced a while ago, I would say under 20k though, and it didn't make any diff to fuel consumption at all even though the original was > 100k old! I know some people say theirs are only good for 60k and they get heaps better fuel consumption as soon as they change it, but I keep a very close eye on my fuel economy and it didn't change at all

The other stuff you said to check:

Injectors/Injector Power Supply: I believe my injectors were flow tested about a year ago and they came back fine. Haven't looked at the injector power supply, might be worth keeping in mind.

Fuel Filter: this was changed when I changed the fuel pump, about 1.5 yrs ago. Should be fine!

Due to some family difficulties I haven't been able to take the car back to the autoelectrician's but in the meantime I was able to do a bit of testing of my own with the aid of my ECUTalk (damn that thing has come in handy!)... I think I must have some "bridged" (?) wires in the AFM circuit, as even when I disconnected the AFM plug I could see a voltage (sometimes) - maybe 0.1 - 0.2V . I also watched the voltage change anywhere from 0.2 - 0.6V while the car was off, but with ignition on and AFM plugged in. And as mentioned before, at idle it should be about 0.2 - 0.3V and I've seen up to about 0.6V. It probably goes a bit above this and at that point, the car stalls. Hopefully will get it looked at soon and will let you guys know what the problem was!

[zebra]

Have you checked every join in the cooler piping? even a tiny airleak plays havoc with the RB ECU and gets worse under load

At Idle you should see anywhere between 1.0 and 1.7V depending on the car

You will see a tiny amount of voltage even with the AFM disconnected.

Have you tried a different ecu? the short that was blowing fuses may have friend something on the board

[EXEMPT]

OK, you may have a point with the AFM voltage and fuel pump there, but I don't know how the ECU reacts if a fuel pump is failing. Also, judging from my last experience, my OEM pump just died suddenly, it didn't let go gradually. So I'm on my 2nd fuel pump, nothing fancy but not OEM, and that was probably 1.5 years ago, so it's highly unlikely to be the fuel pump itself failing. Don't forget, the AFM can't actually adjust the amount of air I'm receiving - all the ECU can do is try and richen the mixture, so if it's seeing what it thinks is too much air, it will try and add fuel. This is why it's running rich - it's seeing higher voltages (ie: what it thinks is more air), and adding extra fuel.

Fuel pump relay: I had a short in the fuel system and this has been fixed. They would have found this at the time - the problem I am having now is something separate. I have already cleaned the IAC valve and AFM!

Plugs were changed just before this happened. Coils are Splitfires and the series 2's run coil-on-plug ignition so no ignitor. There are no obvious air leaks but I'll leave that until last since it's so hard to check O2 sensor was replaced a while ago, I would say under 20k though, and it didn't make any diff to fuel consumption at all even though the original was > 100k old! I know some people say theirs are only good for 60k and they get heaps better fuel consumption as soon as they change it, but I keep a very close eye on my fuel economy and it didn't change at all

The other stuff you said to check:

Injectors/Injector Power Supply: I believe my injectors were flow tested about a year ago and they came back fine. Haven't looked at the injector power supply, might be worth keeping in mind.

Fuel Filter: this was changed when I changed the fuel pump, about 1.5 yrs ago. Should be fine!

Due to some family difficulties I haven't been able to take the car back to the autoelectrician's but in the meantime I was able to do a bit of testing of my own with the aid of my ECUTalk (damn that thing has come in handy!)... I think I must have some "bridged" (?) wires in the AFM circuit, as even when I disconnected the AFM plug I could see a voltage (sometimes) - maybe 0.1 - 0.2V . I also watched the voltage change anywhere from 0.2 - 0.6V while the car was off, but with ignition on and AFM plugged in. And as mentioned before, at idle it should be about 0.2 - 0.3V and I've seen up to about 0.6V. It probably goes a bit above this and at that point, the car stalls. Hopefully will get it looked at soon and will let you guys know what the problem was!

Hokayy,

Are you running a pump that requires 12V instead of 6V? Or maybe it was the other way round? I can remember reading something on here about a bloke that had same issue as you and it was because wires to his fuel pump needed to be bridged to give him the correct voltage to the fuel pump? You'll have to check that shit. I'd pull your fuel line off and let it prime and see if you have pressure if you haven't already. If its a continuous stream of fuel then you'll know the problem is closer to the engine.

Theres no reason why a 1.5 yr old pump can't die. Just like theres no reason that your pump wont suck some serious shit from your tank and spooge it into your filter blocking it.

Check voltage output from the O2 sensor. Also check voltage to the sensor.

Swap in another AFM and check that?

Check your BOV piping too.

Swap in a different battery (clean up your terminals as well )

And then finally check every motherf**king hose and pipe connection you have.

If nothing changes after all of this. Swap in another ecu like zebra says

[benro2]

Have you checked every join in the cooler piping? even a tiny airleak plays havoc with the RB ECU and gets worse under load

At Idle you should see anywhere between 1.0 and 1.7V depending on the car

You will see a tiny amount of voltage even with the AFM disconnected.

Have you tried a different ecu? the short that was blowing fuses may have friend something on the board

Taking the car to the autoelectrician's tomorrow so if he doesn't come up with anything, I'll have to go through the piping with a fine tooth comb. The reason why I'm putting this off is because these problems happened straight after I had changed the plugs and coolant. I was a bit rough with the main intake pipe that runs over the top of the engine, and a few of the things around it, which is why I suspect it's something I've done. I've obviously checked the main join/clamps and they're fine. Obviously it's not the main intake pipe itself as it's metal, and I have checked many times all the air pipes and electrical conn's that I have moved. I suspect that I may have yanked a bit too hard on an electrical wire!

How come there's still a voltage even with the AFM d/c'd? What sort of voltage are we talking here? And should it fluctuate? Mine was all over the place, seemed a bit too much for my liking.

And yes, I'm on my 2nd ECU My old one was partially fried (well before these problems) and it was a REAL pain to find another one (S2, auto), so REALLY hope it's not that!!

Hokayy,

Are you running a pump that requires 12V instead of 6V? Or maybe it was the other way round? I can remember reading something on here about a bloke that had same issue as you and it was because wires to his fuel pump needed to be bridged to give him the correct voltage to the fuel pump? You'll have to check that shit. I'd pull your fuel line off and let it prime and see if you have pressure if you haven't already. If its a continuous stream of fuel then you'll know the problem is closer to the engine.

Theres no reason why a 1.5 yr old pump can't die. Just like theres no reason that your pump wont suck some serious shit from your tank and spooge it into your filter blocking it.

Check voltage output from the O2 sensor. Also check voltage to the sensor.

Swap in another AFM and check that?

Check your BOV piping too.

Swap in a different battery (clean up your terminals as well )

And then finally check every motherf**king hose and pipe connection you have.

If nothing changes after all of this. Swap in another ecu like zebra says

Better not be the damn pump Will definitely look into this if it comes back broken from the autoelectrician's. I have no idea whether it's a 6V or 12V pump - all I know is that it should be as close as possible to the stock pump.

I'll have to ask around for a spare AFM if it comes to that as I don't have a spare lying around

BOV itself and piping has already been checked!

Battery isn't the best, but I can see the voltage readout as I'm driving and it's nothing bad. Starts at around 14.2 - 14.3V and after driving for 15 mins it drops to about 13.7 - 13.8V and doesn't go any lower. With the car off it's a bit under 12V which leads me to believe one of the cells is stuffed, so a new battery is coming soon!

As mentioned above, I'm really hoping it's not the ECU (and doubt it just by how this all happened) as they really aren't easy (or cheap) to find. If it was a manual, no probs, but being an S2 auto, they're like $300+ on ebay and they aren't around all the time!

Thanks for the help guys. Some very useful stuff to check out there. My gut feeling is that it's electrical. After that, I'd say AFM itself, then maybe air leak. Wish me luck!

Edited April 28, 2011 by benro2

[Grimi]

Sounds like you have a very similar problem to me and i cant work it out either... if i find out anything ill post it up...

[V8or4X4]

Between us all hopefully we'l get this sussed. This morning I got in my 32GTR, started it, let it warm(for about 8mins like I always do) and drove about 1k around the corner. Cruising at low revvs my car started to shudder and then began to decelerate as if I had taken my foot completely off the gas. For a few seconds it sounded and felt like it was 5 cyls short then died. It would start but still running like a truck so I gave up not wanting to do(more) damage. Round 3 hours I returned with tools and a fresh tank of patience only to have the car start and slowly drive me home.

Benro2, I apologize for encroaching on your topic but I feel as if these are similar probs.

Does anyone have any suggestions or are we at least in agreement that I should go down the same path as Benro2?

Cheers in advance all.

[DVS JEZ]

Take the afm off and check the solder joints i side it. Theres a diy thread on SAU.

[benro2]

Sounds like you have a very similar problem to me and i cant work it out either... if i find out anything ill post it up...

FINALLY figured it out! Yep, it was the AFM Bought another one off ebay, plugged it in, runs like new!

Between us all hopefully we'l get this sussed. This morning I got in my 32GTR, started it, let it warm(for about 8mins like I always do) and drove about 1k around the corner. Cruising at low revvs my car started to shudder and then began to decelerate as if I had taken my foot completely off the gas. For a few seconds it sounded and felt like it was 5 cyls short then died. It would start but still running like a truck so I gave up not wanting to do(more) damage. Round 3 hours I returned with tools and a fresh tank of patience only to have the car start and slowly drive me home.

Benro2, I apologize for encroaching on your topic but I feel as if these are similar probs.

Does anyone have any suggestions or are we at least in agreement that I should go down the same path as Benro2?

Cheers in advance all.

No probs, I know exactly how you feel Turns out I had about 3 problems all at once which makes it VERY hard to diagnose, but from what you've said above it just sounds like your AFM has had it. Easiest and probably cheapest thing to do in your situation is just to replace it with a known working 2nd hand one (ebay will have a few 2nd hand parts suppliers who will give it usually at least a 2 week DOA warranty).

If that doesn't work, I'd be looking at wiring to ensure you don't have a short somewhere like I did. If you do have a short it's more than likely not going to be in the same place as mine (being a totally different car as well LOL) but mine was right on top of the exhaust manifold heat shield. There is a little bracket there probably to tidy up the wiring and the cables around there were loose and causing problems.

Also, because of the above problem, it kept on blowing the fuse for the fuel pump. This must have been getting worse and worse because the car stalled a couple of times but would start straight afterwards, but on the 3rd time it just wouldn't start (obviously blown fuse at that point), and from then on, every time you replaced the fuse and attempted to restart, the fuse would blow again straight away. So once the short was fixed, replacing the fuse made it all good again.

After all that, VERY happy it drives like new now!!

[nobby]

afm's that have a intermittent problem will sometimes show the fault if you simply tap it's caseing/body with a finger. the vibration can bring fault on.

the soldering inside afm can sometimes also fail with age and be a cause of poor afm.

This can sometimes be fixed, but you have to be lucky for that remedy and able to solder skillfully without damaging electronics inside afm.

NOT recomended if you missdiagnose afm and end out with a dud.

You realy want to be sure it's the fault and needs replacing before you try thisone.

I TAKE NO RESPONSIBLITY IF YOU TRY THIS AND FAIL OR DAMAGE IT ETC ETC.