heh.. sooooo ECU repairs?

  • Thread starter Thread starter GeX
  • Start date Start date

GeX

GeX

GeX

Soldato
Joined
17 Dec 2002
Posts
7,008
Location
Manchester
The French electric gremilins have struck me already :(

It's not firing an injector pulse on number 2. Tested using a noid light. I have check for +ve at the injector and it is present. I have checked continuity back to the relevant pin on the ECU connector and that is there..

.. so i guess it can only really be the ECU

I've had one quote so far of £230 to repair it for me, anyone know of anywhere cheaper, or anywhere that anyone here has used and would recommend?

I'm going to check the output with a scope when I dig out - just to make sure there's nothing / something bad going on from the ECU.
 
they are coded to the body computer and key, so that won't help unless i then take it to a dealers and pay £LOL to get it coded.

edit; that one on ebay.. i'm not sure it would work as my car is a phase 2
 
Sent some info to your email, but probably nothing you didn't know. Does sound like a coil driver from the description. De-coded reconditioned ECU's available? It's that or you need the other coded bits and ECU from a runner, you can change the transponders in the keys to save fitting the locks too.
 
They use Magneti Marelli ECU's don't they? So Italian electronics in a French car - not a good combination!

Crack it open and look for anything obvious - dry joints on the connector or injector drivers. If you can get hold of a similar ECU you could swap injector drivers - I've done this with a Bosch Motronic (I used an old BMW ECU to fix my Fiat one).

Probably worth checking the impedance of all the injectors - if the dead one is significantly lower it may have an internal short which could have fried the driver.
 
Last edited:
BigglesPiP said:
Sent some info to your email, but probably nothing you didn't know. Does sound like a coil driver from the description. De-coded reconditioned ECU's available? It's that or you need the other coded bits and ECU from a runner, you can change the transponders in the keys to save fitting the locks too.
Click to expand...

cheers for the email :)

There are decoded reconditioned ECUs for the phase 1 (or gti6 etc) but not for the phase 2 VTS. The ECU is similar on the two cars, but the phase 2 has 3 plugs instead of the one and some more internal changes for OBDII compliance.

I've looked getting second hand ECU, body computer, key etc - but by the time i've spent all that and faffed about - £230 to repair mine isn't really that bad.

Dogbreath said:
They use Magneti Marelli ECU's don't they? So Italian electronics in a French car - not a good combination!

Crack it open and look for anything obvious - dry joints on the connector or injector drivers. If you can get hold of a similar ECU you could swap injector drivers - I've done this with a Bosch Motronic (I used an old BMW ECU to fix my Fiat one).

Probably worth checking the impedance of all the injectors - if the dead one is significantly lower it may have an internal short which could have fried the driver.
Click to expand...

hehe, i know - in my defence i did assume the Phase 2 used a bosch ECU :D

If I hadn't bought this car to replace one that constantly needed faffing about with then i would certainly look at swapping the driver out myself. but tbh, i just need this thing to work.

I have checked the injectors, and they all come in around 13ohm - so nothing untoward there.

I can't find my scope, so might just send the ECU off and be done with it - they throughly test them before any work, and if they find it to be fault free it's only a small fee for testing.

End of the day, if it costs me £230 to fix - it brings the cost of the car up to £1160 - which is still within what I was willing to spend on buying a VTS (but i'll have a 2 year warranty on the ECU :) )
 
Are you sure it's the ECU? Would be gutting to find afterwards that the same problem exists... have you run a full diagnostic scan?
 
it wouldn't be gutting really, as if it's not faulty it only cost around ~£40 and then i'll know 100% that the ECU is fine.

What do you class as a 'full diagnostic scan' ?
 
GeX said:
cheers for the email :)

There are decoded reconditioned ECUs for the phase 1 (or gti6 etc) but not for the phase 2 VTS. The ECU is similar on the two cars, but the phase 2 has 3 plugs instead of the one and some more internal changes for OBDII compliance.

I've looked getting second hand ECU, body computer, key etc - but by the time i've spent all that and faffed about - £230 to repair mine isn't really that bad.



hehe, i know - in my defence i did assume the Phase 2 used a bosch ECU :D

If I hadn't bought this car to replace one that constantly needed faffing about with then i would certainly look at swapping the driver out myself. but tbh, i just need this thing to work.

I have checked the injectors, and they all come in around 13ohm - so nothing untoward there.

I can't find my scope, so might just send the ECU off and be done with it - they throughly test them before any work, and if they find it to be fault free it's only a small fee for testing.

End of the day, if it costs me £230 to fix - it brings the cost of the car up to £1160 - which is still within what I was willing to spend on buying a VTS (but i'll have a 2 year warranty on the ECU :) )
Click to expand...

IME the BOSCH kit is much much worse, the sensors last about 5 minutes.
 
ok.. so.. yeah.

The ECU tested fine, no errors with it. So they sent back.

Someone suggested it could be a coil pack fault, and so the ECU is cutting off fuel to the cylinder effected by the fault to prevent it chucking fuel into the cat (makes sense i guess)

So I cleared the fault codes, swapped the coils around and then fired it up to see if the fault moved. It did not. So it's not that.

In my (basic) ODBII software it's telling me that the Fuel System 1 is in Open Loop due to a system fault.

I don't know if it's correct in saying that, maybe it's just in open loop because the engine is cold? Either way - i'm out of ideas. Help :(
 
So:
-Removing one of the injector plugs makes no difference to the running (on 3), removing the others takes it down to 2?
-Exactly the same with the HT leads? (or is it pencil coils)
-Swapping the injectors doesn't move the fault?
-Swapping the coils doesn't move the fault?
-Injector drivers are in the ECU, are the coil drivers in the ECU or a separate part?
-How did you get on with my email, did that test show one of the drivers not conducting to ground or were they all working?

I still think the ECU is the fault going by the description so far.
 
-Removing one of the injector plugs makes no difference to the running (on 3), removing the others takes it down to 2?

Correct

-Exactly the same with the HT leads? (or is it pencil coils)

It's COP

-Swapping the injectors doesn't move the fault?
Not tried, but all injectors show ~13ohm when poked

-Swapping the coils doesn't move the fault?
no it doesn't

-Injector drivers are in the ECU, are the coil drivers in the ECU or a separate part?
presume it's all in the ECU

-How did you get on with my email, did that test show one of the drivers not conducting to ground or were they all working?
I couldn't find my scope (think it got put into storage somewhere last time i moved - so just sent the ECU off instead.
 
so you've swapped coils, they're all OK, sorry to be an idiot, but the plugs are OK? Taking the injector plug off a piston with a bad plug can make you think "ah, injector fault" out of turn. If the theory about it turning the fuel off for safety is true then it might be an ignition fault after all, coil driver maybe? If they're not in the ECU you're looking for an "ignition amplifier" or "amp pack".


Got a test meter?
This is unlikely, but check all the ECU's ground pins (in the loom) for continuity to the battery -ve pole.

That all done I'd love to see some scope traces of the injector and coil pins on that ECU in situe.




Edit:
One more, you've checked that there's Battery voltage supplied to each injector, have you checked the continuity of that? If one of the wires is a very poor conductor due to a break/corrosion it will still show up as 12V/14V whatever... But when the injector is turned on and current begins to flow, almost no power gets through to the injector.
 
Last edited:
One more, you've checked that there's Battery voltage supplied to each injector, have you checked the continuity of that? If one of the wires is a very poor conductor due to a break/corrosion it will still show up as 12V/14V whatever... But when the injector is turned on and current begins to flow, almost no power gets through to the injector.
Click to expand...

this is what i intend to try tonight :)

looking at the wiring diagrams, all the drivers are inside the ECU - so in theory have all been tested and are working as they should do.

tonight i'll connect the test light up to the trigger pin and the other side to a direct feed to the battery and see what happens. If the 12v to the injector is poor, it'll show a pulse with a new feed in place. If that doesn't work, i will try connecting the light to the 12v feed on the plug and the test light to ground on the other side. If it lights then that's telling me there is a fault on the ECU/trigger side.

edit; although i checked between the +ve pin on one injector to another and got ~0.01ohm so actually.. the +ve wiring must be fine and oh my god this car is a heap of junk :(
 
GeX said:
edit; although i checked between the +ve pin on one injector to another and got ~0.01ohm so actually.. the +ve wiring must be fine and oh my god this car is a heap of junk :(
Click to expand...

Not necessarily proof, use the light to ground on each supply, then you're loading each conduction path.
 
ok, so this evening i've done a few more bits of testing.

From pin2 of the injector (+ve) i ran a wire to a test light and too ground with the engine running. The light lit up.
From pin1 of the injector (pulse/ecu) i ran a wire to the test light. I ran a wire from G2 (or G3, i forget) of the ECU plug (this is the other end of the pin1 wire from the injector) to ground. I then touched the other wire of the test light onto the +ve terminal. The bulb lit up.

This proves that the connections on the injector wire are good enough to carry enough current to power the test light, so SHOULD work when the test light is in place of the injector. BUT it doesn't.

I assume the firing order is 1, 3, 4, 2. No2 is the faulty one. So maybe the grounding on the ECU is poor and with 3 injectors going, there isn't enough left to power the last injector to fire. (very long shot, but hey tried everything else). So injector number 3 unplugged. Injector wire 2 plugged into the test light. Started the car on 2 cylinders. Light did not flash. Theory disproved.

I tried running a wire from the ECU casing to ground. Nothing.

Oddly here.. with the engine running i measured the voltage over the injector plug and got 12.43v. Tried another and got 12.57v. I do not understand how with power there, the test light won't work. But it further rules out a wiring fault, and points back at the ECU.

The people who tested my ECU (http://www.ecutesting.com/) have not responded to my e-mail yet. So I will call them tomorrow.
 
ok, so now it is sorted. Finally got round to taking it to an autoelectrican who reset it with a lexia.. and now it runs fine.

I guess my ELM327 OBD2 reader wasn't resetting the code properly, and that 'fuel system fault' was a left over from when it initially stalled in that car park many weeks ago.

Big thanks to Steve at Autotronix (http://www.autotronix.co.uk/) Sheffield for sorting it for me :)
 
So it was the ECU imposing a safety measure on an injector? I wonder how that other company managed to prove all the coil drivers were in working order with one disabled?
 
You must log in or register to reply here.
Back
Top Bottom