Engine runs for a few seconds then dies

[Limey]

Hi everyone
I need some help to diagnose some starting problems.

I have transplanted a previously perfectly operative 1uzfe from a 96 LS400 into my 50s Chevy pickup, and I excitedly tried to start it for the first time today.

It starts, but only runs for a few seconds. If I open the throttle during starting it revs up but still dies in seconds.
After it dies it will not start again until I turn the ignition off and back on again. If I don’t do that, and just turn to start from ignition on it won’t fire at all.

The engine was stood for nearly two years before this starting attempt.

I have got an external fuel pump, but still under control of the standard ECU. The pump remains powered for the whole process.

I suspect that the spark is being shutoff by the ecu for some reason, and is only re-established when the ECU is power cycled.

I am using the LS400 instruments, and the revcounter shows the idle at circa 1000 rpm for a second or so, it then rises to about 1300 then the engine dies.

Any ideas?

Thanks in anticipation

 

[gloverman]

Try running on Engine start. After sitting the injectors often seize. If so give them a gentle tap with a screwdriver or likes until the click again. Once running I would have them serviced.

This model the AFM signal issue could cause your issues. People often miss the 12 volt supply to the AFM when wiring. Check this is present.

Another option is check the ignitors are well earthed. They dont have a earth in the wiring. Then check the plugs onto the coils. If one is loose or damaged giving poor contact it will display your symptoms. Next thing I would do is replace the coils.

Good Luck.

 

[Limey]

Thanks for the advice Gloverman.
I have grounded the ignitors, and when it does run its on all eight, smooth and even, so I think I can eliminate the coil or injectors as being the issue.
What does the AFM do?

 

[gloverman]

If there is a fault with the feedback loop of the coils it will turn all eight injectors off - Its a built in feature of the ECU to save the converters if a misfire occurs.

The AFM gives a signal to the ECU of how much air (mass of air) is entering the engine and is the primary input for fuel calculation. Without it most of the time the Gen2 engines fire then stop a few moments later.

 

[Limey]

Answer my own question .... Air Flow Monitor.
I’ll check the connections.

 

[American Locomotive]

Hi, I know it's a little late, but we had the same problem. Turned out the timing jumped 1 tooth on one cylinder bank. Engine would start and die. If you unplugged one cam sensor, it would keep running. Verify your timing.

 

[Limey]

Back to this project at last....
I have checked the cam timing but it’s correct.
One issue I haven’t resolved is that I have a wiring diagram for a 96 North American spec LS400, that mostly matches my uk loom, but not quite.
There are some different coloured wires on ecu connector 11 for example.
Does anyone have a schematic for the engine wiring of a 96 uk car?

 

[gloverman]

Have you confirmed this ecu doesn't have an immobilizer? Some late gen 2 engines for the UK had immo fitted. Have you tried checking for fault codes? The U.S. diagrams show obd2 but the UK aren't so I'd guess you haven't wired a diagnostic port?

 

[Limey]

I have checked the (USA) circuit diagram and the anti-theft circuit has an output enables the starter solenoid, in other words, the engine will not start if immobilised.
My donor car hasn't got OBD2, but I have got the engine mounted Lexus diagnostic socket wired in.
I have run the diagnostics, but I don't get any fault codes, just a constant speed flash.
I'll try again in the next few days and check for codes again.
One more thing, I am not running the original Lexus fuel pump, having substituted an external pump instead, but still using the LS400 system, including a fuel return back to the tank.
I will check the fuel pressure, but even if it was too high I don't see how the ECU would know, and stop the motor until the ignition was turned off and on again.

 

[gloverman]

The engine mounted diagnostic port isn't the one I was talking about. I was referring to the one which enables a scan tool to be connected under the dash. There aren't the correct wires on a gen2 Under bonnet connector to establish comms to a scan tool. Flashing codes is a start though.

I'm not sure what diagrams you are looking at but I've never seen a USD spec car with immo but I have only wired about 20 of them as we generally get JDM over here. I use the ecu number to check against my listings as I have access to the Toyota parts database. I've only ever seen late uk spec cars in non vvti with Immo. Vvti have immo most of the time (post 97) And I've never seen one with the ecu that prevents starter from cranking. The start inhibit switch normal has a feed to the starting system with the ecu getting a neutral switch signal and start signal.

Also are you certain you have the correct ecu, loom and AFM combination.

As long as you get about the right amount of fuel pressure it will still run. I have seen a few with the lines connected in reverse. Easy to check. Remove return line and run fuel pump to confirm you have at least 100litres per hour flow. The start from about 50-60 l/h but will lack with driven. I run pumps that deliver 270l/h without issues so you have to be way too big to cause issues. Ecu doesn't see fuel pressure. Even wiring fuel circuit incorrectly to throw fuel pump circuit code won't stop it running.

Did it run on engine start?

 

[Limey]

I haven’t wired in the underdash diagnostics socket, reason being that I haven’t got access to an OBD1 reader. I still have the socket from the donor car, and could wire it in if needs be.

I’m fairy certain that the immobiliser does not connected to the ECU.

The ECU, loom and the AFM are all from the donor car, which drove with no problems before I stripped apart.

You have confirmed my thoughts on the fuel pump, in that the ECU is not pressure aware, but I was concerned that after many test starts and momentary runs, the plugs got sooty then wet, to the point where the motor wouldn’t start at all. I couldn’t decide whether this was because it was overfueling by too much pressure, or that the ECU richens the mixture a start. I’m sure I haven’t got the feed and return mixed up.

I still suspect reason is to do with a miss connection to the ECU. I can’t get to the diagrams just now, but in a few days I’ll list the differences between mine and the US schematics.

Thanks for the advice so far.

 

[Bushwhacker]

What is your ECU number? 89661-?????
Or, does the ECU have white or dark grey sockets? If they're white then the ECU is immobilised. No problem if you still have the Lexus ignition key and key reader/transponder though.

 

[Limey]

Buswhacker
Thanks for chipping in... The ECU number is 89661-50310, and it has dark grey plugs and sockets.

More info:
On my ECU the "E11" 28 Pin connector has a Red/White wire on Pin 11, and I don't know what its for. On the US wiring schematic it shows a Brown wire on Pin11 that is engine 0v.
On E11 Pin12, the wire colour is Green/Pink, not White as in the US version.

Also, on E11, there is a wire on Pin 23, which I also don't know the use of, on the US wiring schematic there is no connection.
Similarly, on engine ECU "E14" 34 pin connector, there is a grey wire in Pin 19 that is shown on the US version.

Also: I have deleted the EGR pipe from the right exhaust bank to the back of the inlet manifold, and plated over the respective holes.

I have not fitted the vapour sensor, nor the catalytic converters or the after cat. oxygen sensors, all of which I don't think should cause the problems I am getting.

One thing haven't checked yet is whether the ignition signal is getting back to the ECU from the ignitors, but I suspect that it will be fine since I haven't altered the loom in that area since it left the donor car.

 

[Limey]

It’s been a while, but I back on this problem now.
I found that the capacitors in the ECU were leaking, so I have now replaced them all.
Tried to start today, same thing. The endings fires, runs smoothly for a few seconds, at about 1000rpm, then the revs build to about 1400 and the motor stops. If I turn the key back completely to off, then the engine will do the same again. If I don’t turn the key to off, and try to start from ign on,the no starting!

 

[Limey]

More info, and some questions for you good people to answer please......

Engine Fault light now working, and showing following codes:

21 - Main oxygen sensor signal (on left bank)
27 - Sub-oxygen sensor signal (on left bank)
28 - Main oxygen sensor signal (on right bank)
29 - Sub-oxygen sensor signal (on right bank)

99 I can’t find out what this means

The “ECT PWR” green light also pulses steadily when the diagnostic link is fitted.

I have no sub oxygen sensors fitted, so codes 27 & 28 make sense.
21 and 28 are concerning since I haven’t tampered with wiring, or had the connectors apart. Is it possible that both sensors would fail through lack of use after the engine standing for several years?

Would the lack of both main sensors stop the motor running?

What does code 99 mean?

I suspect the Ecu being the cause of the problems, does anyone know which alternative ecus I could try if I can’t get another 89961 - 50310?

 

[gloverman]

99 - immobilizer fault.

 

[gloverman]

With the oxygen sensors it's possible both have failed. I often have vehicles come in that need both replaced however the ecu can only monitor the heater circuit when the engine isnt running so it might be worth checking you have wired the 12 volt supply to the sensors correctly. It also goes elsewhere in the loom which may cause operational issues. Won't stop the engine running though.

 

[Limey]

Gloverman, thanks for the advice.
I’ll check the oxygen sensor’s wiring shortly.
However, most intriguing is the code99. If the ECU is showing an immobiliser fault, would that cause my engine problems?
How does the ECU know there is an immobiliser fault?
On the (USA) electrical schematics I have I see no connections into the ECU from the immobiliser circuits.
In the USA schematics it looks like the immobiliser stops the engine starting at all by interrupting the key start signal to the starter.

 

[gloverman]

I'm not sure what diagrams you are using as the USA spec cars didn't have an immobilizer to my knowledge. As I said earlier the only non vvti engines with immobilizer function were UK spec cars at the end of the model. Hence no connection of immobilizer to ecu as they didn't have it.

How does the ecu know any fault exists?

Do you know how the immobilizer functions?

 

[Limey]

I’m using US diagrams because I can’t get hold of UK ones.
My donor vehicle was April 96, which I presume is a late non VVTI. The car is long gone, so I can’t trace how the immobiliser was connected, I think it must be more complex than just interrupting the starter circuit, and the extra connections that I have on my ECU that don’t match the US one are to do with immobilisation.
Confusingly, one internet search I did showed that my 50310 ecu was from ‘94, which is a long time before my vehicle was built.

If I just guess what the spare ECU wires do I may damage the unit, so it sounds like my best bet is to fit an alternative, non immobiliser ECU, any ideas what ones will suit my engine?