False Crank Sensor Triggering

[alistai_Westy]

Hi,

I’ve been working on this issue for a long time and can’t solve it. Hopefully someone out there has had the same problem and solved it or has some advice.

Background.
Setup
2.0 Ford Duratec
DTA S40
Crank Sensor is a variable reluctance type Ford 2 Pin.
Shielded twisted pair cable used for this input into the S40.

Problem
When I turn the ignition on without turning over the engine the S40 thinks the engine is turning. In the DTA Swin Diagnostic display I can see this in the starting information.
I can stop the crank sensor false triggers by moving the +ve battery cable into a set position, generally nearer the engine block, however when I turn on a load like the hazard lights the injectors start to fire in time with the flasher relay. The same thing happens when I switch on and switch off the headlights.
I first thought this was an earth fault but I have been through everything, carrying out voltage drop tests, and found no issues.
I have swapped out the crank sensor for a new one, the issue is still present.
I have ensured that the crank sensor wiring is routed as far away from high current wiring as possible.
I have tried using capacitors across the alternator output to the alternator case and this doesn’t solve the problem.
The engine installation is in a Westfield kit car and as such the alternator is mounted in a non standard position. It is closer to the crank sensor than in a production road car.

I initially thought the root cause was electromagnetic radiation inducing a voltage in the crank sensor input but I’m not sure now.
The car battery is not in the best of health however I have connected another known good battery in parallel and the fault did not go away.

The engine starts and runs well but the problem is when the ignition is on it can fire the plugs and create a flame that scorches my throttle body valve (its plastic).
Any help and advice is much appreciated, thanks in advance.

Al

[Wally]

First off I would get the battery changed for a good correctly fitted one and then check again. You say the fault still occured when you had a second battery in parallel but if there was a fault in between the battery terminal and cable and you used jump leads onto the terminal you would still have the fault. Did you do the volt drop tests from the battery terninal or the battery post?

Moving the battery +ve cable should not cause the fault you have under a low load (not cranking) unless it is making and breaking a contact somewhere, and operating the flasher should not unless the battery voltage is fluctuating so much.

Is the Shielded twisted pair cable used for the crank sensor original ford fitment or one you have fitted?

Wally

[Alex DTA]

What firmware version is on the ECU? Please check using DTASwin, not what's on the lablel.

The only way the ECU can think the engine is turning is if it's receiving a signal on the crank pin.
To see what sort of signal the ECU is seeing, use the crank shaft oscilloscope.

Check the earth on the twisted pair shield. This needs to be connected to a sensor ground on the ECU end of the shield.

Another test to try would be to unplug the ECU, and turn on the hazard lights to see if the problem is still present. If it is, it's a simple wiring issue.

[alistai_Westy]

Ok thanks for the replies,

Responding to Wally first:

I have checked voltage drops on both battery cables and battery posts. I get the same reading of around 0.1Volts.

The crank sensor twisted pair is one I have fitted.

Response to Sparrow:

The firmware version is quite recent. I took the ECU into the office at Leamington in August this year. The guy reflashed it then.

The earth on the shield is on one end only, at the ECU plug and is good.

Unplugging the ECU makes the fault go away because there is no output to the coils or injectors.

I have attached some pics of the crank oscilloscope. One with the engine running and one with the engine not running and hazards on. It seems that when the engine is on the "interference" is not present.

Im about to give up on this one. Help before I go mad!

Cheers

Al

[Alex DTA]

That's a wiring problem, most likely an earth issue. It could be the battery is faulty, so start there, as you've already mentioned it being faulty.

The injectors are firing is because the ECU is seeing the spike as the engine starting to turn, so it fires the injectors for the amount of time in General Engine settings. As there's a large delay (as far as the ECU is concerned) between each hazard light flash, it's going to fire the injectors each time the hazard lights flash.

BTW, I'm the guy in Leamington.

[ignitionautosport]

I suspect your hazards are finding an earth back through another circuit, or back through the ECU earth if it doesn't have a good earth itself.
Maybe. Odd fault.

What voltage at the ECU when the hazards are doing that?

[ edit: sparrow posted while I wrote this, but I'll leave it anyway ]

[alistai_Westy]

Ok,

I've installed a new battery.

There are no earth issues. I have removed the whole car loom and installed 3 new earth leads to the chassis using appropriate gauge wire.

I have removed all the fuses except for the circuit that supplies the ECU relay and fuel pump. The fault is still present after the fuel pump has finished priming. So in essence the only load is the ECU.

I have installed a new 4mm positive wire to the ECU, the old one was a bit small.

Nothing above has solved the problem.

Sparrow - the engine is not turning on the starter while the issue is present.

That's it, I give in!

[Alex DTA]

Sparrow - the engine is not turning on the starter while the issue is present.

I know. However, the oscilloscope image you posted above with the hazards on shows the ECU seeing a signal. The signal is from the hazards, but the ECU doesn't know this, it thinks the engine is starting to turn.

I would try unplugging every connection from the engine to the ECU, apart from the crank sensor. Coils, injectors, temp sensors, everything. See if the problem persists.

Where is the earth on the engine wiring loom? Is it connected to the chassis, or the engine somewhere?

[gnutz2]

If the only problem is the fact it backfires and tries to melt the plastic airbox then I suggest setting the injector start pulse to zero in general engine settings.

[alistai_Westy]

I have unplugged all ECU inputs and outputs apart from the crank sensor and the problem still exists.

The injector start pulse is set to 1. Cant set it to zero.

Time to splash some petrol on it and drop a match in.