Car starts then cuts out immediately

[Gavin Waddell 18]

Second update:

 

went to to the car today and carried out a short to ground. 

I unplugged everything set ohm meter to highest level ( that’s what scanner d’année on you tube said to do) and went through the loom plugs as per telfs diagram, and nothing shook the loom moved the loom about nothing I checked my earth connection before every time I checked a plug to make sur it was good.

 

 

[PhilNW 44]

On 5/28/2020 at 2:38 PM, Gavin Waddell said:

Went and plugged it all back in tried to start and nothing. I jumped the tach relay to rule that out and nothing. Unplugged the AFM and it started. Not good.

How well does it run when no AFM connected? Think you need to focus on this, sounds like a bad signal is getting back to the ECU killing it rather that no signal which will allow some running.

 

Does it do the same thing every time with the AFM unplugged?

 

Does it run continuously with the AFM unplugged however badly?

 

From an old sketch I made a while ago when I was having problems with no start, (sorry did not note cable numbers at time)

AFM connections

terminal (5) nearest bulkhead  - green should be +ve (from tachy relay)

next terminal (9)                            white should be an earth

next terminal (E)                            no connection

next terminal (7)                           yellow to ECU (signal wire? voltage?) 

terminal (8) nearest front            grey to ECU (signal wire? voltage?) 

 

Assume all gti's have the same cable colours and numbers is this what you have got?

 

 

 

Edited June 2, 2020 by PhilNW

[welshpug 1,661]

On 1/9/2020 at 11:40 AM, Gavin Waddell said:

 

what I don’t get is why it’s runs when the temp sensor is un plugged or the AFM. I looked through Haynes wiring diagrams and I can’t really see what could go wrong apart from maybe écu or AFM?

 

the injectors won’t fire when everything is plugged in but once I unplug one of the two mentioned above it runs albeit very rich.

 

 

 

 

 

So I had a quick scan of the whole thread again and this is what jumps out at me, when the injectors don't fire is the 12v still present?  if it is, check for a grounding pulse from the ecu.

 

The AFM< SAD<TPS share the same 12v switched supply from the ecu, the CTS shares the same earth but gets its voltage from the ecu.

 

[Gavin Waddell 18]

Thanks for that, now that I know that the loom has good continuity and no shorts to earth, I’m going to start checking voltages on the components. Just waiting for somebody new crimping pliers to arrive so that I can make some nice splices, been watching dave from Haltech loom blog, got me very inspired to do things right the first time.

[Nobbly 22 2 Cars]

Hi Gavin,

 

I've just read through this (as my tach. relay is playing up!!) and can't help but think there must be something obvious that you have missed - apologies if I'm going over old ground. 

 

Am I correct in assuming that you haven't yet changed the tach. relay?  My understanding of your symptoms:

• Car starts then stops
• Car runs (but very rich) with either AFM or CTS disconnected.
• Possible low voltage at the injector +ve

Pin 50 from the starter will supply fuel (i.e. tell the pump to run when cranking) hence your car starts initially.  When you release the key from the start position the relay relies on the speed signal to keep the pump running, this is a safety feature.

 

So, if you have either a fault with the speed signal, associated wiring or (more likely imho) the relay then this could be the issue.  Perhaps the disconnecting of the AFM and/or CTS (you said you have no SAD) is affecting the relay just enough - they all share the same pin on the relay - to power the injectors/pump. The rich running once started could be due to the ECU assuming the temperature of the coolant is very cold or the air flow is very high (as the sensor is disconnected).  Both will cause the ECU to add more fuel than is needed.

 

Hope this helps, good luck with your fault finding.

 

David

[Gavin Waddell 18]

Thanks for that, now that I know that the loom has good continuity and no shorts to earth, I’m going to start checking voltages on the components. Just waiting for somebody new crimping pliers to arrive so that I can make some nice splices, been watching dave from Haltech loom blog, got me very inspired to do things right the first time.

 

Good point on the tach relay. The thing is that i bridges the contacts on it, so the injectors, AFM, fuel pump etc where on a constant feed and it still has the same problem. 

[Dazzer 1969 0]

Hi

new to the forum Ohave a very similar problem with my GTI, overfueling and missing, if we disconnect the fuel pump, car runs fine, until it runs out of fuel.

any help would be great

[welshpug 1,661]

1 hour ago, Dazzer 1969 said:

Hi

new to the forum Ohave a very similar problem with my GTI, overfueling and missing, if we disconnect the fuel pump, car runs fine, until it runs out of fuel.

any help would be great

better with your own thread than confusing posters with information not pertaining to the original posters vehicle.

[Gavin Waddell 18]

Little update.

 

checked voltage at the injectors and 12v then rechecked with AFM unplugged and 12v.

 

checked the CTS and it had 5v

 

unplugged the CTS started the car and with my probe light you could see the écu was grounding the injectors. 

 

So all norma there, wanted to see if the injectors were ground with everything plugged in but my battery died.

 

allof this was done with the tach relay removed and the terminals bridged that way it rules out a dodgy tach.

 

going to go and have another look over the weekend once the battery is charged.

 

gavin 

[Gavin Waddell 18]

i went back to the car and had another crack, with everything plugged in tried to start the car and nothing, i was unbale to see if the injectors where being grounded as the car wont even idle.

 

i also retrid a new afm and nothing.

 

i am gong to bridge the tach rlay again this time with out bridging the fuel pump and will try to start the car and see what voltage readinds i get at the differnt components. im guessing that everything should be getting 12v.

 

thanks

 

Gavin

[Gavin Waddell 18]

Little update

 

had a few hours on the car, I redid the injector wiring with new source and and ground wires. 

 

Plugged everything in tried to start it and zero. Checked voltage at the injectors and they get 12v when cranking? Checked the ground with a probe and there was ground. Also checked AFM and good ground and 12v +.

 

unplugged AFM car started so measured 12 v at the injectors and checked they were grounding with probe light which it was the probe light  was flashing 

 

So like everyone suggested the écu is possible not grounding the injectors ? 

 

But why does it work with Cts unpluged or AFM ?

 

the grounds our all good.

 

even if I bridge the tach relay it only runs unplugged?

 

If you have an idea as c razy as it might be let me know ands I will carry out the check on the car to see what happens .

Really want to find out what The problem is.

 

 

thanks Gavin 

[PhilNW 44]

Did the ground at the injectors  vary from about zero ohms to infinity when cranking , if the ecu is working properly it it should do. It happens that fast that you should get some form of variation in reading on your probe/meter (as it cant react that quickly).

 

Does the same thing happen when running with AFM disconnected? If its running then ECU is grounding in some way.

 

As I suggested in earlier post sounds like the AFM is giving a rogue signal killing the ecu and/or ground. 

 

Sounds like the fault is repeatable so sounds like problem is consistent (rather than an intermittent fault)

 

Edited August 1, 2020 by PhilNW

[Gavin Waddell 18]

I’m going’s to try that next, I’m guessing I can see a discrepancy if I carry out the test twice in the 2 different conditions.

 

any one have valued for the different plugs to check against?

 

thanks

[Tom Fenton 1,542]

Have you tried the engine on easy start? Disconnect hoses and get someone to spray while you try to start it.

If it runs on the easy start you can rule out a mechanical problem such as timing head gasket bent valves or whatever. It’s a useful

fault finding method.

[Gavin Waddell 18]

Hey tom 

tried easy start and it starts up fine, meaning it’s the injectors not firing. 

Like Phil posted above a dodgy signal from one of the components. 

 

Im trying to find some base values for the components so that I can cross reference and find what’s not working properly.

 

gavin 

[PhilNW 44]

Cant remember but does the car start (however badly) with the AFM disconnected on its own ,  the cts disconnected on its own or with both disconnected?

[Gavin Waddell 18]

Starts with either one dissconnected I have tried to start it with both dissconnected . Which is strange as the CTS and AFM dont share the same power only the ground is the same. The AFM also has the 2 reference wires which run to pin 8/9 on the écu ? Which is not shared with the CTS.

 

so it could be a ground but I’ve double checked them for continuity and shorts and there is nothing, I also moved the ground connection around and it changes nothing. 

 

Changed afm nothing and also the écu.

[PhilNW 44]

The reference wires you mention I think are the signal cables, one sends flap position opening and the other sends the inlet temperature if I remember correctly

 

If you pull back the rubber cover to the AFM plug (still plugged in) what voltages do to get on the signal wire pins and the 12v supply and earth?

Not sure if it is a permanent 12v supply to the AFM or only when starting or running.

 

In particular does one of the signal cables vary as you move the throttle flap with a screwdriver or similar?  The other one should have a voltage but pretty static as the inlet temp varies only slightly.

[Gavin Waddell 18]

Hey Phil 

 

cant get to the car right now I’m going to go and have a look tomorrow I will let you know.

 

any chance you could measur your AFM that way I can compare findings?

[PhilNW 44]

Have you gone through the Injection/AFM bench test procedure in Haynes manual (Chapter 3/25)? This gives some values

 

When testing as suggested in previous post above with everything connected just see what is happening on the signal wires to the ecu/12v supply/ground in-situ.

 

The CTS (in theory) is a very basic sensor which should just give a voltage to the ecu based on water temperature as you have swapped this I cant see the CTS being the problem.

 

Suggest concentrate on the AFM and its wiring at the moment.

[Gavin Waddell 18]

I did bench test it and retracted it also the values where pretty much what was stated in Haynes/ auto data so static test good.

[Gavin Waddell 18]

so i went to the car and did a bit of a test to see what values i got with the AFM inplugged and the Afm plugged in

 

AFM plugged in

 

wire 1 (signal from ecu) =7.4v

wire 2 (signal from ecu=1.6v

wire 3 (signal from tach)=12v

wire 4 ground =2.2V?

 

 

Afm unplugged

 

wire 1 (signal from ecu) =2.2v

wire 2 (signal from ecu)0.06

wire 3 (signal from tach) =12v

wire 4 ground = -0.0 (so ground)

 

looking  at the above figures when the AFM is plugged in all the values are ok, apart from the ground which seems not to ground?

 

when i unplug the afm the ground seem to be ok but all the other vaules are all over the shop.

 

i might make up a fly lead so i can try to ground the AFM some where indepedantly , or any one got any other ideas.

 

gavin

 

[PhilNW 44]

Couple of extra checks

 

With battery disconnected, what is the resistance reading (ohms) for the earth wire with AFM disconnected?  should be less than 1 ohm according to Haynes.

 

With battery and afm connected does the voltage vary on terminal 7 ? as you move the AFM flap? Should mimic the variation in resistance in Ch3/25 para 8 in Haynes.

 

I think the direct earth fly lead to the AFM is the next move depending on the 2 answers to the above.

 

Since I had my 205 I have always had a couple of very long fly leads with crocodiles clips at one end to connect to the battery to get a good 12 v and a good earth anywhere so I can bypass any possible wiring issues when testing. 

Edited August 5, 2020 by PhilNW

[Gavin Waddell 18]

So the voltage on pin 7 varies with the battery connected .

 

i set my multimeter on 200 ohm and I get a reading of 124 ? 

 

Dodgy ground?

[PhilNW 44]

If the 124 ohms is with the battery disconnected then looks like dodgy (high resistance) ground.

 

If you add 124 ohms to the internal resistances of the AFM and CTS it probably puts the overall resistances out of the range the ECU can cope with.

 

Check the resistance of the ground to the CTS in a similar manner just for a cross check.

 

Then try fly leads to both AFM and CTS grounds or find the source of high resistance.

 

In my case it was the terminals in the connector to the AFM , replaced all 4 and problems went away

 

 

Edited August 5, 2020 by PhilNW