Lift off -> stall prob, an update

[Matt Harwood]

After spending some time with Ian trying to help find this fault, I have to say, I feel sorry for him. It's a really strange problem, that seems to manifest itself in many different ways.

 

When I first arrived yesterday, the first thing we did was to swap ECU's. We put mine in his car and it coughed, but wouldn't start. I put his in mine, and it ran wiout any problems. So, his ECU is fine.

We then changed the plugs, (again), and after a bit af faffing around, the diagnostics told us there was a fault with airflow and the o2 sensor.

we found that the AFC's airflow wire had come disconnected during the ecu swap-outs, and that there was some dodgy pins in the ECU harness.

After correcting these, the car still wouldn't start. It was fuelling, but not sparking. We disconnected one of the coil pack plugs, to test, and it was fine, but strangly, during the test, the car decided it wanted to start!

 

So, we now had a running car, albeit with Ians previous stalling symptoms. (if anyones seen the film, Gone in 60 seconds, the clatter the car makes while it's on lift-off and trying to stall, sounds exactly like the Mustang does when it stalls in the film - I doubt that helps, but you never know).

When the car gets a stable idle, if you open the throttle a tiny bit, there is a definate click coming from the rear of the engine, that can be felt through the rear of the injector rail. As this click happens, the engine seems to stumble a bit, and as you lift off, the engine tries to stall again.

While this was going on, we disconnected the 6th injector while the car was running and it hardly made any difference to the cars idle. We then disconected the 2nd injector, (these injectors were only chosen for ease of access), and the 2nd did make a noticable difference.... But now, the car started idling at just over 1000 rpm.

We stopped the car, and when restarted, the idle was hunting from about 1300-2000ish!

I've given the ECU wiring a good tug around whilst it was running and it made no real difference, so I'm fairly sure there are no more duff joints down there, but short of tracing every wire, it's hard to be certain.

 

I left Ian at this point, still none the wiser. I posted this as another persons description might jog someones thoughts and hopefully if anyone else can see/suggest anything to try.

[Matt Harwood]

Originally posted by Ian C

I'm *sure* it must have been loose and he didn't simply break it

 

Hey, you did the first ECU swap that caused the car to not start!

[Ian C]

Cheers Matt I'm gonna yank the rail and injectors out and see what's what. It's a shit job but I've gotta find that click, I think it's part of the mechanism that's faulty, and I'll also see what's what regarding the injectors. Might get them flow tested and cleaned. There ain't much else left to test/swap/clean/hit with a spanner

 

-Ian

[Dragonball]

Very Large drink if it is this m8 - fingers crossed - sounds familiar!

 

Phantom Problem

Our shop Supra started to exhibit a strange problem one day after running flawlessly for over a year and a half. When started, it would idle very rough for several minutes like it was only running on 2 or 3 cylinders. After 2-6 minutes it would start cleaning up and finally run properly.

 

We tried various things to try to narrow down the problem. The symptoms were inconsistent at first. Sometimes the engine would start and run fine. Other times, it had the problem several times in a row. Engine temperature appeared to have nothing to do with it. Cold or hot, the engine would have the problem sometimes. Day by day, the engine was running rougher and it was taking up to 10 minutes for the engine to clean up. Finally, the problem was so bad, the car would hardly get off the line even with lots of clutch slipping. In neutral after starting, the engine would only rev to 3000 rpm with full throttle. Every time, the engine would eventually clean up and run fine.

 

We checked the coil pack to make sure that the problem was not ignition related. Everything there was fine. We scoped the injector outputs. We installed a fuel pressure gauge to make sure that we had proper pressure. We pulled off the return line to verify that we had good fuel volume. We did a compression check, checked the ignition wires, unhooked the cold start injector entirely, checked the injector ballast resistors, injector resistance, injector leakage and finally changed the ECU. (It's never the ECU). None of these produced any change. All exhaust ports were hot, indicating that all cylinders were firing sometimes but the mixture appeared to be incorrect.

 

Finally we unplugged the injector harness to one set of 3 injectors and the engine ran much smoother even though only 3 cylinders were firing. We then plugged in one injector from the other bank of 3 and the engine ran rougher but picked up about 100 rpm. We plugged in one more and the engine picked up another 50 rpm but got even rougher. We then plugged the last one in and the engine lost 400 rpm and shook like mad. With the injector output plugged into the other 3 injectors, the engine ran OK on 3 cylinders again. Something was dragging down the injectors when the second output was connected. This would cause the injectors to be lazy in opening, making the engine run lean.

 

After a bit of head scratching, we decided that the only possible explanation was a poor injector ground. On the EM-3, the injector ground is connected to the white plug on the ECU. We checked the pins and connector and saw no problem. We disconnected the injector ground wire from the chassis connection which was a bolt coming through the firewall, near the transmission tunnel. We re-grounded this wire with an alligator clip to bare metal. The problem cleared up immediately. The ground bolt in question was the same one used for the last 18 months on 2 different systems. When checked with an ohmmeter, the resistance varied between 4-5 ohms. It should have read 0 ohms.

 

The bolt held part of the heater assembly on to the chassis and looked like a good ground. It WAS a good ground for a while. The failure appeared to be caused by salt water corrosion under the bolt washer. The roads are heavily salted in winter here. I have seen a lot problems in 20 years of working on cars, but never one with these inconsistent symptoms. I cannot explain why the ground would always eventually become good enough to make the engine run after a while, yet 10 minutes later, would break down again and cause the problem. This took a lot of time to solve. There is a good lesson here. Grounds are easy things to check if you have a problem. Even it it was working and suddenly you have a problem, grounds should be one of the first things checked

[Ian C]

Heh - my plaintive searches of the internet got that story too Paul

 

I'm not convinced that it's a MkIV Supra they are talking about, from their singular use of "coil pack" and the two sets of injector wiring harnesses, plus the MkIV switches the injectors by earthing them through the ECU - this means each injector has it's own earth wire independant of the others. Of course, they are all switched internally to the same ECU ground but that ground is fine. I've even cleaned it up despite it being clean already

 

Might check their power supply is a clean 12v with no high resistance though...

 

-Ian

 

PS re-reading it this is talking about an aftermarket complete ECU, the EM-3, so it might have been a MkIV.

[eyefi]

is all yr moddified injector loom ok including the resistor pack?

[Ian C]

As well as I can be sure of!

 

I'm gonna check continuity from battery positive to injector feed and see if it's the expected resistance (if only I could remember that resistance ) I can also unplug the resistor pack and check that there isn't any resistance there either.

 

-Ian

[Timwildman]

It might be that the injector is getting 12volts but not enough current to make it work correctly, Like Pauls story. You can sometimes find a multimeter will register 12volts, but the item just wont work. A good test is to connect a lamp of simular current draw across the output, and if it lights up its definatley good. Or you can just carry out a curry draw test.

 

We have this sometimes at work, everything looks like it should work but theres just 1 bad crimp pulling the whole lot down.

[Ian C]

But the only time the car runs OK is under load, booting it is still mad enough, and a gammy fuel injector would soon manifest itself under those conditions. I'm not convinced it's a dodgy injector, I'm pretty much taking the fuel system apart because I've checked everything else more obvious, not because I think I know what the problem is

 

-Ian

[jagman]

just a thought on your idle mixture problem,is it possible the ecu is getting a false engine cranking signal from the starting circuit,so even though tne starter is out/engine started ,the ecu thinks its still cranking,hence increasing the injector pulses-overfuelling? I think the signal comes from the start relay,could this be the click you are hearing? can you just remove the wire to the ecu and try it -this problem is starting to bug me, its like an unfinished crossword

[Ian C]

You mean it constantly thinks it's starting? Possible - I'll check that circuit. I'm charging the battery again at the moment, I think all this has nailed it's coffin up

 

I found a ground which doesn't appear to be documented, goes from the rearmost bolt on the rear engine hook to a random hole in the firewall:

 

http://www.ian.chisholm.clara.net/pix/randomearth.jpg

 

Anyone else got this? I've outlined in yellow the basic path of the wire and the outline of the bracket that is bolted (yellow X, one bolt) to the firewall. I ask because the resistance of that bracket is all over the shop so I've taken it out (nothing else was on it ), cleaned all the white powder off the ring terminal crimped onto the earth wire, and bolted it directly to the firewall where the yellow X is. Haven't tried anything yet as the battery is charging, the car just clicked the starter motor.

 

With the battery out it also means I can't test the voltages etc going to the injectors. I get 680ohms between the positive battery terminal and the injector power feed, which climbs to 708ohms if I hold the probes on long enough. I guess this is due to various permanent live connections.

 

jagman - it feels more like a murder mystery* to me, and as I said to Matt I can't shake the feeling that there is one vital connecting clue I've overlooked...

 

-Ian

 

*or toothache at times

[eyefi]

ive got this one, connects at the lift hook then goes back to a tab on the bulkhead.

[Ian C]

That's the beastie. I've got a piccy of Terry's engine bay that just happens to have this bit in shot as well so it must be a stock fitting and yet I can't find a reference to it in the grounding points in the US service manual. Hmm.

 

That tab thing, is it just a bit of metal bolted in one place by a 10mm bolt on the firewall? Seems a bit pointless to me...

 

-Ian

[jagman]

pin 77 on ecu-is from starter relay,to give fuel enrichment during cranking,it would kinda explain the symptoms youve got,as it would have almost no effect at wot,and max effect at idle or during lift off, - the answers easier in a murder mystery you just start from the last page , oh and when youve found your problem,every one on here will say "oh I would have checked that first":D

[eyefi]

Originally posted by Ian C

That tab thing, is it just a bit of metal bolted in one place by a 10mm bolt on the firewall? Seems a bit pointless to me...

 

it is, its also got a weird foam pad stuck under it, holding most of it away from the body, just making contact at the bolt.

[Guest Terry S]

Big C have you looked to see what the fuel pressure is doing?

[Ian C]

I just started the car and it ran at 1600 to 1800 rpm. After some poking around and a couple of retries it reached a peak of 3000rpm with the throttle shut.

 

This is insane, and also officially taking the piss.

 

Any clues?

 

-Ian

[Dragonball]

Originally posted by Ian C

I just started the car and it ran at 1600 to 1800 rpm. After some poking around and a couple of retries it reached a peak of 3000rpm with the throttle shut.

 

This is insane, and also officially taking the piss.

 

Any clues?

 

-Ian

 

Talking out loud here

 

Well it cant be mechanical - you would have signs - you know your way round!

 

the clicking must be some sort of solenoid opening?

 

The ECU works fine!

 

Have you taken the AFC out of the loop?

 

Regulator working?

 

This has to be 'electrical'?

 

What about running some grounding leads from the battery?

 

The closed loop system is not working - the O2 sensor then warms up and tries to compensate

 

What makes this closed loop system work - shuts off all the normal O2 readings until sensor can differentiate before and after readings and start making legit adjustments

 

Why is this not working?

 

This has to be electrical!

 

Makes ben Hur look like an epic!

 

 

 

 

 

[Ian C]

Thanks for all that Paul I took the earth off the battery and left it for some time, and repositioned the TPS according to an old closeup picture I had. Bastard thing fired up normally and has done since. I've no idea what went on there but it survived the ECU fuse pull which was interesting.

 

So, since then it's fired up normally each time, that particular episode seems to be over, but by "normally" I mean of course "tries to die in cracking the throttle and on lifting off".

 

That clicking noise is strongest around injectors 3 & 4 on the fuel rail. There is nothing else for it now but to strip it out and visually inspect the rail and injectors. That's a job for tomorrow. However I would like to check my fuel pressure, how can I do that?

 

-Ian

[Adam W]

I would have though that there is no way the engine can be revving that high with a closed throttle unless you have a pretty large air leak somewhere; even with the idle valve jammed fully open it wouldn't rev that high would it? No matter how much fuel the ECU is pumping in it needs air to mix with unless your car is doing something VERY weird!

[Ian C]

It'll rev to 2000rpm with the idle valve wired up but physically unbolted But then some sort of failsafe kicks in and it oscillates between 1500 and 2000rpm. This time however it started at 2000+ and got faster until I switched it off.

 

Why it did it I don't know. I've wanged the ECU loom around and wiggled the injector wires like Matt did when it first kicked off like this on Sunday and got no change now it's back to "normal".

 

The throttle was not open, it had to have been the ISCV wide open.

 

Is the fuel pulsation dampener bolted to the block just by the starter motor?

 

-Ian

[dude]

Originally posted by Ian C

It'll rev to 2000rpm with the idle valve wired up but physically unbolted But then some sort of failsafe kicks in and it oscillates between 1500 and 2000rpm. This time however it started at 2000+ and got faster until I switched it off.

 

Why it did it I don't know. I've wanged the ECU loom around and wiggled the injector wires like Matt did when it first kicked off like this on Sunday and got no change now it's back to "normal".

 

The throttle was not open, it had to have been the ISCV wide open.

 

Is the fuel pulsation dampener bolted to the block just by the starter motor?

 

-Ian

 

Yup ugly looking thing fuel hose either side of it , ive got a by pass hose you can borrow to try if you want :thumbs:

 

Dude:flame Dev

[Dragonball]

Derivability Symptoms

A lambda sensor's normal life span is 30,000 to 50,000 miles. But the sensor may fail prematurely if it becomes clogged with carbon, or is contaminated by lead from leaded petrol or silicone from an antifreeze leak or from silicone sealer.

 

As the sensor ages, it becomes sluggish. Eventually it produces an unchanging signal or no signal at all. When this happens, the Check Engine Light may come on, and the engine may experience derivability problems caused by an overly rich fuel condition. Poor fuel economy, elevated CO and HC emissions, poor idle, and/or hesitation during acceleration are typical complaints.

 

If the average voltage from the lambda sensor is running high (more than 0.50v), it indicates a rich condition, possibly due to a bad MAP, MAF or Air Flow sensor or leaky injector. If the average voltage reading is running low (less than .45v), the mixture is running lean possibly due to a vacuum leak or because the sensor itself is bad.

 

If the lambda sensor continually reads high (rich), it will cause the engine computer to lean out the fuel mixture in an attempt to compensate for the rich reading. This can cause lean misfire, hesitation, stumbling, poor idle and high hydrocarbon emissions (from misfiring).

 

If the lambda sensor continually reads low (lean), it will cause the engine computer to richen the fuel mixture. Injector pulse width will increase causing fuel consumption and carbon monoxide emissions to go up. Constant rich fuel mixture can also cause the catalytic converter to overheat and it may be damaged.

 

If the lambda sensor's output is sluggish and does not change (low cross counts & long transition times), the engine computer will not be able to maintain a properly balanced fuel mixture. The engine may run too rich or too lean, depending on the operating conditions. This, in turn, may cause derivability problems such as misfiring, surging, poor idle, and high emissions.

 

If a heated sensor has a faulty heating circuit or element, the sensor can cool off at idle causing the system to go into open loop. This usually results in a fixed, rich fuel mixture that will increase emissions.

 

Sometimes an apparent lambda sensor problem is not really a faulty sensor. An air leak in the intake or exhaust manifold or even a fouled spark plug, for example, will cause the lambda sensor to give a false lean indication. The sensor reacts only to the presence or absence of oxygen in the exhaust. It has no way of knowing where the extra oxygen came from. So keep that in mind when diagnosing oxygen sensor problems.

 

The lambda sensor is also grounded through the exhaust manifold. If rust and corrosion of the manifold gaskets and bolts is creating resistance, it may affect the sensor's output. To rule out a bad ground, use a digital volt meter to check for a voltage drop between the sensor shell and the engine block. More than 0.1v can cause a problem.

 

 

 

 

Keeps pointing to the good old sensor m8

[dude]

I think Ians allready tried the O2 sensor !!!

 

Dude:flame Dev

[Dragonball]

http://www.picotech.com/auto/lambda_sensor.html