Explosion, now no start.

[vikenabm]

OK, had a spare ECU for an automatic transmission, same grey plugs. I have a M/T with a converted A/T harness, so I swapped the ecu anyway and it cranks fine. STA actually not checked once it cranked ok. Other voltmeter tests for the ECU and the "ignition on" were all correct, including all 3 injector channels, KSU signal, MREL, and ISC1-4.

Now the symptoms are back to before the explosion: With injector resistor connected I'm getting something. It chugs for a second and sputters, seems like it will almost go, but won't start all the way. When it was doing this before I managed to get it barely running a a few times and that's when it exploded and stopped, so I'm afraid to keep trying to start it without a clearer idea of what is wrong. Timing is set approximately according to the TSRM instructions for installing the CPS, which usually is enough to get it started.

Timing light didn't work in the injector wire. I put the magnetic pickup that normally goes on a plug wire over one of the two wires going to injector 1. Didn't trigger the light with old or new ECU.

Timing light does flash on plug wire #1, maybe not perfectly regularly, but better with the new ignitor than with the old ignitor, I think. STill not sure the spark is 100% regular. checked coil resistances - all ok. Plug wires (MSD) checked a month ago - some were close to 20K ohm, which is in spec but barely. I have some old stock plug wires left, as well as stock injectors and stock AFM, and an extra coil pack.

1)Should I focus on trying to get the timing more accurate by having someone crank while I check timing? Or do I need to double check anything else before going further. Not sure about these RC injectors, Lexus AFM, AFPR: haven't worked with this setup before.

2) Should I scrap the setup and go back to stock AFM, injectors, and plug wires just to test it? That's of course a couple of hours converting, and a couple more changing it all back later, but well worth it if I learn something from the exercise. Actually, I crimped the new injector pigtails and heat shrunk the connections, so I can't really go back to stock injectors very easily. I could make a adaptors of the old injector pigtails and some crimp on female terminals, I suppose. I'd do that if it would help.

3) should I go back to stock on just the plug wires, maybe swap the coil pack?

4) Should I continue with the A/T ECU? How much difference will it make?

At this point I don't want to shoot from the hip. I'd just be picking at random from the list.

Thanks.

[cre]

I would start with swapping the electronic part of the old AFM into the Lex AFM. The electronics are the same and you know the old one works I presume.

I don't understand your reply about the STA signal. This signal is important as it tells the ECU that the car is being started and despite the airflow being below the specified threshold for the emergency fuel cutoff that it should power the pump and injectors for a set period of time. At least this is my understanding of how it's applied in the 7M-GTE's case. The timing light on the injector wires is to ensure that they are getting a proper signal when cranking. If you know it's getting fuel then this probably doesn't matter.

As I reflect I feel that the explosion may have been either due to the ignition timing being wildly off from the mechanical timing OR the injectors are firing way out of order and fuel is pooling in the runners until that unfortunate spark lights one off. The injectors are low impedance right?

I wouldn't worry about keeping the other ECU in place. I don't think that's the issue... I could be wrong though.


I'll go back over this thread in its entirety tomorrow, think about it for a bit and try to see if I can think of anything specific which we've neglected. Have you verified that the injectors are connected in the proper order?

[vikenabm]

Originally Posted by cre

I would start with swapping the electronic part of the old AFM into the Lex AFM. The electronics are the same and you know the old one works I presume.

Yes, though the part number is different by one digit and might not be eletroncially identical. I have to supra part in the lexus body now, I will try that swap.

Originally Posted by cre

I don't understand your reply about the STA signal. This signal is important as it tells the ECU that the car is being started and despite the airflow being below the specified threshold for the emergency fuel cutoff that it should power the pump and injectors for a set period of time. At least this is my understanding of how it's applied in the 7M-GTE's case. The timing light on the injector wires is to ensure that they are getting a proper signal when cranking. If you know it's getting fuel then this probably doesn't matter.

Ok, I didn't understand the STA function. I will check it. My only proof that i'm getting fuel is that it almost starts when the injector resistor connector is connected and not otherwise. I believe this takes the injectors on and off line. With my lovely assistant, I will also verify that pressure is over 30psi WHEN cranking.

Originally Posted by cre

As I reflect I feel that the explosion may have been either due to the ignition timing being wildly off from the mechanical timing OR the injectors are firing way out of order and fuel is pooling in the runners until that unfortunate spark lights one off. The injectors are low impedance right?

both issues may have come to play, i think so. those are the two issues I considered: back fire or ignition timing. It sounded more like timing, too loud and crisp for a back fire, not a pop, but a boom. and a lot of protesting cranking, like the engine was kicking back at the starter. That isn't happening any more so I think the timing is at least better.

Originally Posted by cre

I wouldn't worry about keeping the other ECU in place. I don't think that's the issue... I could be wrong though.

I feel the symptoms are different with the new ECU: It definitely is much closer to starting now, chugs in it's own for a second, so I'm thinking the old ecu was damaged by jumping it incorrectly. there was smoke and an ozone smell in the passenger compartment... Just a matter of whether to buy a new M/T ecu. Plan was to wait until I get it running before worrying about that. I will take the old ECU apart and inspect just to see if the board is burned.

Originally Posted by cre

I'll go back over this thread in its entirety tomorrow, think about it for a bit and try to see if I can think of anything specific which we've neglected. Have you verified that the injectors are connected in the proper order?

I really appreciate your help. the injector plugs are long enough now that they could be connected out of order, but it would be pretty obvious. Only #5 and #6 are easy to mix up. I will verify continuity to the correct injector resistor pins per TEWD with a voltmeter. Also will verify that both leads to 1 and 4, 2 and 6, and 3 and 5 are in continuity with each other. Connection of each of the 3 circuits to the ECU was already verified with voltmeter yesterday.

FWIW - I wonder if air/fuel mixture could be way too rich. I have AFPR, A/F adjust screw on the AFM, and a bosch 'BOV' (recirculating discharge with stock plumbing). I have checked for boost leaks carefully, so I don't think that is a problem, but I might have been fixing boost leaks when the problem starting first occurred, before I screwed up the timing. Just another factor to consider.
Thanks again.

[cre]

Ok, I didn't notice you had mentioned the smoke and burnt electrical smell previously... Wow, did I actually miss that? Check wiring too, more may have burned than than the ECU. The ECU is pretty well closed up, actual smoke is more likely to present from wiring or a relay. Do you know that you jumped the wrong pins? If so which ones? One big issue with burning wire harnesses is that the wire doesn't always burn to the point that continuity is interrupted, you often get shorts across other wires where the bundle is wrapped tightly together... not fun to track unless you can narrow it down by locating two devices which are not both operating and share the same harness.

No need to buy another M/T ECU, the A/T ECU should function the same. I'm pretty sure there aren't even any wiring changes to be made.

The fuel mixture would have to be obscenely rich to prevent the engine from at least chugging for a minute after a cold start. Especially if it's been allowed to sit long enough for the excess fuel to evacuate. I'm talking rich to the point that the injectors are leaking. Now, that said, it might behoove you to pull them and test them as specified in the ECU to be sure they're not leaking and are opening when voltage is applied. The shock of the combustion in the intake could damage injectors but I don't think it's very likely. If you think the cylinders are flooded then pull the plugs, crank the engine for several seconds and let it rest for 24 hours to ensure that the vapor has had time to evacuate the cylinders.

The electrical components in the two AFMs are identical and present the same scaling when tested in the same housing. They are completely interchangeable. The housing swap generally requires tuning. I do not recommend using the air bypass screw for this purpose. Its scope is full spectrum and the airflow signal needs to be tuned selectively for the best performance and to ensure that you're not creating any lean spots or washing a cylinder. Do yourself a big favor and invest in a fuel controller and a wideband controller. The wideband is an invaluable diagnostic tool. Innovate Motorsports makes a couple very inexpensive controllers that are very well rated and easy to use.

A boost leak isn't going to cause a problem like this... You'd have to have a 1" or larger hole in the piping to see something this bad. A leaking BOV won't have any affect on the operation of the engine when not boosting.

You haven't made any modifications to the AFM or sprayed any type of cleaner on the sensor, have you?

I haven't gone back through the thread yet and it's been a long day. I will try to make time tomorrow. Hopefully we can get this sorted out in the next couple days and have you on your way. Open and inspect the old ECU as you said. If any of the traces or components are burnt see if you can identify which pin(s) on the harness are related and thus limit the amount of time you may need to spend diagnosing the wire harness.

[vikenabm]

checked wiring - I pulled the dome fuse because there was a question of a minor short in that circuit. Otherwise, no short to ground. and so far haven't found any wiring faults doing voltmeter tests on the ECU. Here's what I did: jumpered the car with positive to positive and negative of the other car to the positive alternator output. Don't ask me why, in do know better. blew a fusible link and a a fuse, replaced these and had headlights coming on but not popping up, and not turning off with the headlight switch off. When I finished testing all the circuits for shorts and replaced fuses the lights worked normally and it was behaving pretty much normally, minus the dome light fuse, except the ECU, which I replaced. Now seems okay except the starting issue. Really no idea why the lights went haywire or how they resolved.

Not flooded - tried it several days, same result. Also, I usually disconnect injectors or injector resistor and then crank with throttle open to clear a flooded engine and that isn't helping with this problem.

I have a wideband
Fuel controller is the next purchase. once it starts.

No mods to the AFB and no cleaner on the sensor.

Don't sweat the time frame. I'm a weekend warrior mostly, but your rosey prognostication gives me hope, thanks. Couldn't work tonight, it's day by day.

[cre]

I've been through this with a couple MKIIIs and I know a few others who've been there too. It's a royal pain and typically is due to more than one point of failure or something which the ECU and the TSRM don't know how to diagnose (such as a short in the wire harness across two lines such as Vc and Ks.... been there, rebuilt the whole harness). With diligence you'll find it it just takes patience and a lot of deep breaths.

Now that you have an ECU that you're more confident in in place you should check for error codes regularly to see if it sees something we haven't.

[vikenabm]

Okay, I think I've covered all your questions so far.

I verified fuel pressure, set it to 35. Only issues is that may AFPR leaks pressure a little too fast when the ignition is off. I don't think that is keeping it from running.

My injectors are correctly wired to both the ECU on one side and the injector resistor on the other side. THe injector resistor has a good ground.

My ignitor works, it is wired to the correct coil packs and plugs, and the plugs all fire in pairs as described in my write up on the first page of this thread. I hope what I write up is the correct wiring for the 7MGTE - it was the best I could figure from available sources.

I pulled all my plugs - one had a cracked ceramic insulator, don't know if it was firing. others were intact but ALL PLUGS WERE both WET with gas and black with soot, and there was a thin layer of gas on each piston.

My intake air pipes were sooty from sporadic intake runner ignitions.

My original ECU still doesn't run the car, though it cranks. It had no evident burned circuit board components when i took it apart. The new ECU still almost starts the car, it chugs along very rough for a second or two.

I swapped the electronics on the AFM. I'm using the one that came with the Lexus throttle body.

Any thoughts now?