1.4 Petrol cut out whilst driving

Hi all,
firstly, apologies for the long post, maybe it will be useful in the future though...
I have a 2004 A2 1.4 Sport, Petrol BBY engined . It has been very well behaved for >5 years and has ~65k on the clock. Regularly serviced at an Audi specialist. I love it.
Driving on a dual carriage way a few days ago, the low fuel warning light came on (justifiably) and then within a minute or so all sorts of lights came on (fuel, ABS, engine warning etc etc) and the engine cut out. Luckily I was able to coast on to a slip road...
The recovery guy thought it might be related to the fuel pump system as no pressure coming from the little valve at the end of the fuel rail. I think I've discounted this as if pressure is released and the ignition turned back on, the pump will run for a second or two and re-pressurise the system. I also directly successfully tested the fuel pump by disconnecting the battery and applying 12v at fuse #18 ( I *think* it was 18).
I've done the following as listed below and I'm fairly convinced at the moment that the ECU (ECM) has popped a clog. It would be fantastic to hear any thoughts, things that you think I should check or suggestions as to next steps. I am struggling to believe that (for example) a dodgy crank or cam sensor will knock out the whole ECU.
Unless I've missed something obvious, can anyone recommend an ecu supplier/repairer? There are ecus with this model and version number available on the bay for ~£200, but I think I'd rather buy one that I can be confident is working.
As an aside, there are also cheaper ones that look more recent. What would happen if fitting a newer ecu to this car? Maybe no-one has been mad enough to try it ;-)

I haven't checked any earths (yet?) - the body work is really sound and there's no damp in the passenger footwell. The battery is very healthy.
I have got a copy of VCDS-Lite, registered it, with a cable from Gendan.
When turning the ignition on:
ESP light stays on
ABS light stays on
The engine turns over willingly but will not fire. No revs shown on the speedo when turning over.
fuel gauge reads ok

During scanning, I was never able to connect to the ECU (ECM), address 01.
Tried setting Baud rate to 9600 as per https://www.a2oc.net/community/inde...o-communicate-with-engine-control-unit.28711/
Disconnected battery overnight as per
Disconnected ECU, reconnected
After all these actions, I got the same result:
[VCDS could not communicate with address 01]
The ECU is marked as Magneti Marelli 036 906 034 CA, 4MV 004 5949
Other scan results:

VCDS-Lite Version: Release 1.2 Saturday, 30 January 2021, 12:22:55:57640
Chassis Type: 8Z - Audi A2 Scan: 01,02,03,08,15,16,17,18,37,45,46,56,57,75,76,77
-------------------------------------------------------------------------------
Address 03: ABS Brakes Labels: None
Controller: 8Z0 907 379 D Component: ESP FRONT MK60 0101 Coding: 0007175 Shop #: WSC 00000 VCID: 49E53F0F60A1
2 Faults Found:
01314 - Engine Control Module (This went after clear codes)
013 - Check DTC Memory - Intermittent
01314 - Engine Control Module
004 - No Signal/Communication

-------------------------------------------------------------------------------
Address 08: Auto HVAC Labels: None Controller: 8Z0 820 043 D Component: A2-KLIMAVOLLAUTOMAT 0514 VCID: 370149F70655
No Faults Found or DTCs not supported by controller or a communication error ocurred
--Rescanned, not able to get faults
"if you have too many communication errors while try to read the aircon it's because of wrong setting in the usb port . the latancy timer is set to 16 you have to change it to 1 and then it works fine."
--I tried this later on
-------------------------------------------------------------------------------
Address 15: Airbags Labels: 8Z0-959-655.LBL Controller: 8Z0 959 655 G Component: Airbag Front+Seite 0005 Coding: 00102 Shop #: WSC 02152 VCID: 50D3126B83FB
No fault code found.
-------------------------------------------------------------------------------
Address 17: Instruments Labels: 8Z0-920-xxx-17.LBL Controller: 8Z0 920 980 R Component: KOMBI+WEGFAHRS. VDO D04 Coding: 04442 Shop #: WSC 00681 VCID: 50D3126B83FB WAUZZZ8Z45N003172 AUZ7Z0D2649277
3 Faults Found:
01177 - Engine Control Unit
49-00 - No Communications
01314 - Engine Control Module
49-00 - No Communications

01760 - Power Steering Control Module (J500) No comm. with Engine Control Module (ECM)
49-00 - No Communications
-------------------------------------------------------------------------------
Address 45: Inter. Monitor Labels: 8E0-951-177.LBL Controller: 8Z0 951 177 Component: Innenraumueberw. D04 Coding: 00001 Shop #: WSC 00000 VCID: 55DDE37FBC19
No fault code found.
-------------------------------------------------------------------------------
Address 46: Central Conv. Labels: None Controller: 8Z0 959 433 AF Component: Komfortgerát TBD 1210 Coding: 06859 Shop #: WSC 02138 VCID: 8A674003A527
No fault code found.
-------------------------------------------------------------------------------
Address 56: Radio Labels: 8E0-035-1xx-56.LBL Controller: 8Z0 035 152 D Component: chorus II GRO 0004 Coding: 02000 Shop #: WSC 02138 VCID: 361F44F3015F
No fault code found.
End --------------------------------------------------------------------------------------------------
I also did a control module finder scan (00 - 7F) and found the following addtional items:
Address 25: Immobilizer Protocol: KW1281 Controller: Bit teA dre ss Component: e 17 eingeben
Address 44: Steering Assist Protocol: KW1281 Controller: Bit teA dre ss Component: e 17 eingeben
Scanning these addresses gave:
25 - no no faults found or comms error
44 - no faults found or comms error.

"if you have too many communication errors while try to read the aircon it's because of wrong setting in the usb port . the latancy timer is set to 16 you have to change it to 1 and then it works fine."
changed latency in USB to 1
(In windows this is in the advanced settings of the usb device in device manager)
Testing now shows good latency

Address 08: Auto HVAC Labels: None Controller: 8Z0 820 043 D Component: A2-KLIMAVOLLAUTOMAT 0514 VCID: 370149F70655
2 Faults Found:
01271 - Positioning Motor for Temperature Flap (V68) (this went after clear codes)
41-10 - Blocked or No Voltage - Intermittent
01206 - Signal for Duration of Ignition Off Time (this stayed)
27-00 - Implausible Signal
01206 will not clear
 
A bit more reading suggests that if the ECU is dead, then I won't be able to replace as without the original secret codes I can't add them to the replacement controller. PS, I'm in the Bradford/Leeds area.
 

Joga

A2OC Donor
If you could borrow an engine ECU from someone and connect it, you could rule out any missing power supply to it, or damaged communication buses if you then get responses from that ECU in VCDS. Just a thought.
 

depronman

Member
You can get the skc read out from the instrument cluster
I’ve done this many times on tdi’s and have all the cables to do so but never tried on a petrol
It does sound a bit likely that the ecu is at fault if vagcom will not talk to it

Paul


Sent from my iPhone using Tapatalk
 

audifan

A2OC Donor
For the cost of a few pennies I would replace the following fuses. Numbers 36,38,39,42 and 45 ( if they come up in the lottery let me know ). Some directly effect the ECU some indirectly but for the cost before thinking it is the ECU got to be worth a try. Then try to start / scan with VCDS.
 

Howey

Member
Does it crank? If the engine earth has failed so will have the charging and could lead to a failure to run?
 

Robin_Cox

Member
Worth noting that VCDS lite will not speak to the Magnetti Marelli ECU in our 1.4i either despite working with our 1.4tdi cars and my A6 without issue. However, my ODB reader speaks to the 1.4 and clears errors as well as all the other basic functions (TB adaptation for instance) so I've managed without anything more sophisticated so far. @depronman has suggested probable Baud rate alteration with my PC laptop could fix it, but I've not tried changing that yet.
 

depronman

Member
Worth noting that VCDS lite will not speak to the Magnetti Marelli ECU in our 1.4i either despite working with our 1.4tdi cars and my A6 without issue. However, my ODB reader speaks to the 1.4 and clears errors as well as all the other basic functions (TB adaptation for instance) so I've managed without anything more sophisticated so far. @depronman has suggested probable Baud rate alteration with my PC laptop could fix it, but I've not tried changing that yet.

2400 baud seemed to work well. 9600 is std in windows. Adjust in control panel against the com port which windows assigned to the cable


Sent from my iPhone using Tapatalk
 
Hi all and thanks for all the advice & suggestions.
I have set the baud rate to 2400, checked and replaced all the fuses and re-scanned. Nothing appears to be any different - scans remain the same and ECU is still not contactable.
The battery is good and the engine cranks well. I have trickle charged the battery, but this is just to make up for the occasional starting attempt and endless VCDS scans.
@Robin_Cox , what is the OBD tool you use?
@depronman / Paul, what is your set up to get the skc out of the instrument panel? My understanding at the moment is that the SKC is one of the numbers that comes from the VCDS scan - ie VCID: 50D3126B83FB WAUZZZ8Z45N003172 AUZ7Z0D2649277
It feels like getting hold of another ECU is the next step.

It seems the car is Gen 3 immobiliser, http://wiki.ross-tech.com/wiki/index.php/Immobilizer, so I would need the pin code and/or a pin extraction tool. I certainly haven't got the PIN code.
 

Howey

Member
If it was the ecu then afaik get the same number ecu and plug it in, the car should then start and stop after about 2 seconds with the immo light flashing on the dash. Then you would need all keys pairing so clocks and ecu are all in harmony
 

Robin_Cox

Member
Hi all and thanks for all the advice & suggestions.
I have set the baud rate to 2400, checked and replaced all the fuses and re-scanned. Nothing appears to be any different - scans remain the same and ECU is still not contactable.
The battery is good and the engine cranks well. I have trickle charged the battery, but this is just to make up for the occasional starting attempt and endless VCDS scans.
@Robin_Cox , what is the OBD tool you use?

It is an Xtool VAG401 that I think came from Amazon around 6 years ago. For me it works brilliantly - I originally bought it to save myself £50 every time I had to go to the independent Audi guy locally when the EML light went on in my 2003 A6 (eventually fixed with a new fuel cap!), but have subsequently used it with my A2s for checking & clearing error lights, adapting new throttle body on the 1.4i, calibrating the fuel gauge on my AMF tdi car (fuel light was coming on with 11L remaining, now got it down to 6), and switching between maps on the Tdi cars from winter to performance and vice versa, Webasto diagnostics etc. There are a few tutorials on youtube from a guy with a MkI TT using the same VAG401 for various adaptations (the fuel gauge calibration on his car works in the opposite numerical direction from my BHC car fuel gauge - a well-known sender model change foible) but useful nonetheless. I've just searched youtube with Xtool VAG401 and there are actually a lot more videos now that may be useful (obviously, not all A2).

I need to qualify this : other people here have had mixed or negative experiences with the same piece of kit, so it is plausible that I have been lucky on many levels with this exact scanner and my exact cars - but it actually works better for me than VCDS lite for the things that I need to do (mostly because of not needing to gather extension cables and power supply for laptop, wait for it to boot up, open the software, then try to handshake with car etc.) - I am usually not at home, and outdoors, so just being able to plug something powered out of the ODB port by my knee is a godsend. On our AUA car this is what I have used all through the project with sorting out the engine lights, EGR, TB, misfires, defective coil, leaky oil filler cap and so on. Proper VCDS should be better overall, particularly with trickier diagnostics, but for around 32 quid original purchase price it has been a good buy.
 
I managed to get a bit more time on this today. Decided to look at earth points, so I started with the engine to frame one first. Cable was intact and I cleaned the connections. I was certainly reminded of how tightly this stuff is packed in...
Then I removed the passenger kick panel to look at the earth point there and found a 10a fuse that has blown. I have spent a while searching on here and more broadly but either there's not much out there, or I'm rubbish at search. Does any one know what this fuse protects?


DSC_0810.JPG
 

Proghound

Admin Team
I know the circuit for the seat lumbar adjuster motors has a 10A fuse and it's in that location. Might not be the only one though.
 
Thanks Proghound. Searching on Lumbar adjust confirms that. I didn't even know it was a thing. Definitely something wrong there - I put a new fuse in and with the ignition on it blew fairly quickly. I'll add it to the to do sometime if ever get the car working again list. Still slowly working through earthing points at the moment and wondering what next.
 
I've done some more investigating. ElsaWin tells me that pins 1 & 2 on the ecu connector (J537) are earth and pin 3 looks like the 12v supply. 1,2,4 & 5 all have earth continuity and pin 3 shows 12v when the ignition is on. I've swapped out for another ecu and this has made no difference - ecu cannot be contacted by VCDS. I got too cold to look for continuity on the CAN bus but I can try that. I could even connect a CAN analyser to the car. I won't know what the CAN messages mean, but I could maybe listen at different locations and see if the message stream changes. It all feels a bit desperate at the moment ;-)
 

Robin_Cox

Member

could it be related to part 6 or part 23 in this diagram? There are various models depending on the engine, but there is a fuse as part of the unit in each case, and they appear to carry relays for either the ECU (6) or the fuel pump (23) respectively - so I would imagine the fuse going would impact on function of components powered by the relays?
 
Part 23 isn't there, the is nothing apart from a fuse (which his fine) in this carrier.
Part 6 is in the larger carrier, but not in the place implied by the diagram above. - An 8Z0 906 383, the fuel pump relay. It also appears to supply the injectors and charcoal filter. Also has some ancillary switched contacts feeding the ECU and air-bag. There also 1 fuse in this carrier which is ok. And an intermittent wiper relay.
When I test this, the main and all ancilliary feeds go to 12v, but after about 2 seconds, the main feed is switched off and will not come back on again until the primary 12v is reset. From the wiring, I cannot see a way that a reset could occur without switching off the ignition, so this seems like a good candidate for a replacement. This was a good 'lead' Robin, thank you, with no apologies for the pun.
 
I have solved the problem. Feel a bit dumb because a fuse *had* blown (fuse 20). My only excuse, your honour, is that it isn't marked on the fuse layout panel in the car. I now see that it's plastered all over the internet though. Anyway, cylinder 1's coil had died, taking the fuse with it. I know it's cylinder one because when I plugged a new fuse in, smoke started coming out of that coil. Live and learn as they say. I found this by following the fuel pump relay and ICU check processes in Elsawin. The fuel pump relay cuts out after 2 seconds if there is no signal from the icu on one of the small relay pins..
So 4 new coils I think. Thanks for your help everyone!
 
Top