ITB`s with EFI won`t start. Need ideas

[Cheshire Cat]

Ok, 3.2SS, ITB`s with Ecumaster EMU Black ECU. The car was starting, idling, drivable. I was in a process of tuning\bits fixing etc.
The last time car started about over a week ago, I had a few tuning related issues (like problems with start when hot etc), nothing major. Got back home killed the engine.
Next weekend starting the car - nothing. Starter works, cranking but she won`t start.
Thought - run out of fuel. Refuelled, started cranking - nothing. The ECU`s scope shows nothing at all like I don`t have sensors yet the RPM`s in software shows progress.
Changed the crank sensor - didn`t change anything. Cam, sensor acts dead.
Cam sensor wiring is fine. Crank sensor shows errors and rpms. But the car wont start.
I don`t know what to look at. I`m frustrated.
I didn`t do anything after she was working.
So to me it`s a mystery I need to solve. I need your wisdom gents. Please.
You`re my only hope.
Cheers

[JakobM]

I would leave the ECU for a moment

1) Pull a plug - what do you see/smell? fuel or no fuel?

2) Earth the plug, and have someone crank the engine - do you have spark?

I would guess, one of the two is where your problem is, then narrow down the reason

[mikedsilva]

I had similar issue last year with my AEM Infinity. Turned out that the power lead to the ecu had corroded over time, where it was connected to the terminal on the car. Once stripped bare, and the terminal cleaned, it was perfect.

[Cheshire Cat]

Well, since I`ve posted I`ve changed:
crank and cam sensors - nothing.
Scope for crank sensor shows random. Scope for cam sensor shows nothing. And I have 3 sensors now.
Spark is there, fuel is there. How I know? When testing output channels for ignitors the motor bangs.
There is an idea that the ecu might be faulty. What makes me think that? Filtering for crank sensor signal set to "none" makes ecu think the car has started and it sparks as it supposed to with no key turning involved. So either signal shielding (grounded one one end) or the ECU. Will test tomorrow.
p.s. funny thing, I didn`t see my post "live" till today...

[Jonny042]

Originally Posted by Cheshire Cat

Well, since I`ve posted I`ve changed:
crank and cam sensors - nothing.
Scope for crank sensor shows random. Scope for cam sensor shows nothing. And I have 3 sensors now.
Spark is there, fuel is there. How I know? When testing output channels for ignitors the motor bangs.
There is an idea that the ecu might be faulty. What makes me think that? Filtering for crank sensor signal set to "none" makes ecu think the car has started and it sparks as it supposed to with no key turning involved. So either signal shielding (grounded one one end) or the ECU. Will test tomorrow.
p.s. funny thing, I didn`t see my post "live" till today...

There have been some oddities with the forum lately. There was a complete outage for a whole weekend not long ago.

[dannobee]

Back to basics. Spray some carb spray down the intakes and see if it starts and runs for a few seconds. If it starts, the ignition system is working. Check fuel pressure and check for an injector pulse with a "noid light" if available. If it doesn't start, start checking on why the ignition system isn't working. Check fuses, basic wiring (B+ and grounds). Check the resistance of the crank and cam sensor circuits at the ECU. And of course, check for fouled spark plugs.

[john walker's workshop]

Love me some carbs that don't make me want to rip my remaining hair out.

[mikedsilva]

Originally Posted by john walker's workshop

Love me some carbs that don't make me want to rip my remaining hair out.

I hear you!

[Knockdown]

There is a lot of current from the battery when cranking the engine, make sure you are not getting a large voltage drop from the engine to the chassis. This will play havoc with the sensors on the motor. This is best tested with an analog volt meter and some long test leads. I had to clean the ground strap connection points really good located up by the transaxle when I installed my EFI/ITB’s.

[stownsen914]

Originally Posted by Knockdown

There is a lot of current from the battery when cranking the engine, make sure you are not getting a large voltage drop from the engine to the chassis. This will play havoc with the sensors on the motor. This is best tested with an analog volt meter and some long test leads.

This was one idea I had as well. Much less than 9 volts during cranking and your ECU may get unhappy. Had this exact issue when I first got my current racecar running. In my case I needed a more substantial battery. Charging the battery, or a jumper pack may give it some additional oomph to start.

[pampadori]

The ecu should know when you're cranking. Most require a signal wire for that so they can account for low voltage. Are your temp values within reason when power is on?

[Jonny042]

If you're showing RPM from crank sensor with random pulses and cam synch completely dead....

- Are you using VR or Hall? Shielded wires? Both would be wired to sensor ground from what I can tell - other sensors working ok on that ground?

Both types of sensors use the same pin (B7?) to connect - is there a software selectable setting for hall vs vr that could be set wrong?

Can you wire cam synch #2 up and reconfigure to use it instead as a test?

[3rd_gear_Ted]

Your in Australia, get a Motec

[Jonny042]

Originally Posted by stownsen914

This was one idea I had as well. Much less than 9 volts during cranking and your ECU may get unhappy. Had this exact issue when I first got my current racecar running. In my case I needed a more substantial battery. Charging the battery, or a jumper pack may give it some additional oomph to start.

He did say it sat for a week.....

[al lkosmal]

If your cam and crank sensor signals are flaky.....and you cannot synch ....you will not start.
If you have verified that your batt voltage is not too low and the cam and crank sensors are good, I'd check the ignition setup in the program, to ensure that it is still correct...and if that does not reveal the problem...it is possible your ECU needs to be looked at.

Also, EcuMaster seems to have very good tech help. Have you contacted them? They can connect to your system, remotely, and help diagnose the problem. It will be better to go direct to the specialized tech help.

I have a friend (blee on Pelican) who elected to go with PMO carbs and has an EMU Black for sale, if interested....



regards,
al

[Tea Tray]

^^^^ I agree with al.
…signals are flaky….
Check connectors for displaced pins.
Check wire to wire for insulation integrity.
Ring out each wire with a meter and wiggle the wires like they’re a boa.
Disconnect/bypass the sensors and feed voltage and ground directly and scope the signal while cranking.

[Cheshire Cat]

Ok, thank you gentlemen, for all the suggestions.
Here we go.
Yes, the car sat for a week or so yet the battery almost constantly on the charger (CTEK 3.8 just in case).
I definitely can see the voltage drop on cranking and could so when she started before, say the last time I have a working log when she started the voltage dropped from 12.6 down to 7.84 and yet she started, with having the charging voltage up to 12.2V on 1300 sort of RPM.
Crank has VR sensor (have another one, brand new) and cam has hall sensor from clewitt (have 3 of them by now, 2 brand new).
Shielded cables on both crank and cam sensors. Curious thing is, trying to fix this current problem I`ve realised that my shields aren`t grounded, fixed it (the ground point is\was on the same spot as ignitors and injectors relays, under the passenger seat) and it gave me no help as for the start issue but gave me some troubles on "filtering" issue, this one "Filtering for crank sensor signal set to "none" makes ecu think the car has started and it sparks as it supposed to with no key turning involved".
So, I will relocate the shielding grounds to whatever it needs to be (so far there are a few options - sensor grounds, power grounds or ECU grounds), so far I have them ungrounded.
But, I did NOTHING in between those times when she ran and when she refused to start. Like LITERALLY nothing.
I can\will try to hook up cam sensor to secondary trigger 2 pin, to make sure if it works. That is not a problem.
Both crucial sensors are grounded to one of sensor ground pins on the ECU, so to check that pin I can relocate it to some other sensor ground pin. But grounds are there and sit very tight and on their places on the block. The battery ground is tight and fixed, the car lives in the garage and no water involved.
And there are a couple of people who trying to help me, in this case Ecumaster is very good, but one of them in Germany and another one in Sydney. None seems to be here in Melbourne to look at the car.
I`ve managed to wire old EMS ECU, she started, I drove her etc, now this - started, drove etc and now - nothing.
And this is literally the first time I`m jealous I`m not on carbs...
Pins are on their places, but I will check them once again just because I became paranoid and don`t trust myself any more.

[Cheshire Cat]

Sort of an update, possible a solution...
I was trying to find/fix the problem and started to check the inputs if they are failed, changed the cam sensor input #1 to cam sensor to #2 - didn`t help. So it`s not the ECU`s inputs are failing.
Started to check the grounds - all the inputs give good signals on mustimeter, nothing wrong on the ECU side.
Decided to check the wiring to the sensors on the motor side, to do so I had to cut the zip ties that kept everything organized, checked the grounds - all good, checked the gap between the tooth wheel and the crank sensor, adjusted it to 1mm, and tried to see the scope signals. And it`s there. The cam sensor is there and the crank now shows almost what it has to show, now I`ll charge the battery to the max level and will keep going. Have to sort those shields asap.

[Cheshire Cat]

Aaand another update. You know what?
Bloody ground strap from body to gearbox... The nut was tiny bit loose (not like loose-loose), tighten it up - almost started. Put some shielding grounds - here we go - she starts!
But... She dies after some time, for some reason I have many errors on trigger when she runs... Have to figure that one too...

[Jonny042]

Yay!

Isn't it ALWAYS a ground?

Suspect you might want to add some grounding between the chassis and block if you are grounding sensors to the block. IMO if one loose nut is enough to cause that issue you want to beef up the ground system overall.

EDIT - if you have poor grounding, tying the shields to poor grounds could make things worse, not better. So you can get confusing results from your efforts, and the harder you try the worse it gets.

At least you're on the right track.