164 Q4 ABS problem (warning light on)

[Alfan]

I went for a short test drive in the Q4 yesterday and the ABS error code reappears after being erased. I will probably try and put the other unit back in later, to see what that will bring.

I took photos of the error codes from the Marelli tester.

ABS error 1

ABS error 2

ABS error 3

ABS error 4

ABS error 5

[Alfamuppet]

Looks like the front right and rear left solenoid valves have a problem ? These would be part of the ABS unit probably controlled by relays

[Alfan]

As mentioned earlier, I have had the welded relays replaced - before that it gave *pump motor' failure. I think this is a problem within the mechanical unit, so I will try and replace that with another one and see what comes out of that.

[AlfaWhiz]

Long time has passed but I will be posting an extensive update on the thread very soon.

I did a lot of measurements which will be helpful for everyone here (for example I've mapped all components from the schematics to the actual physical unit, so I know what is where exactly and how to measure it from the top).

At this point I'm pretty sure that my pump motor is shorted to ground thus causing issues on both relays, but I will ask you guys for your analysis here as well, so stay put, I'll post the details soon.

[AlfaWhiz]

So here are the mappings I've done between the N51 group and schematic. You can see which diode is where in reality, diode labeled by me as 4 and 7 are not visible. There is also my crude hand drawing, where you actually can see what is connected to what, I will try to digitize it when I have time.

Hopefully this is helpful to everyone.

I'll be posting my measurements soon, as something does not add up in my case unfortunately.

[AlfaWhiz]

Results:
- I am measuring ground when the screw leading power to the pump motor is screwed down (this I suppose is short to ground, it's not supposed to be shorted, but I may be wrong)
- Because of this the measurements are split into 2: with the screw in place (would be normal operation, but I fear it's shorting to ground), and without the pump motor connected at all (no screw in place)
- After 10 or 15 seconds the solenoid valve relay disengages on it's own, I assume N51 group CPU is ordering it to shut off, this is measured as +10,8V on both pins 85 and 86
- I've marked in orange what I think is wrong on the measurements
- Battery voltage is measured at ca. 11.2V, drop to 10.8V on relays may be due to loss on the diodes
- I don't understand why the voltage on pin 87, 30 and 86 (as they are all connected) drops from ca. 10,8V to 0,7V when the relay is engaged (this is only on key ON)
- I don't understand why pin 30 of pump relay is getting 10,6V after 10-15 seconds when solenoid valve relay disengages, it should get it's current through pin 30 of said relay
- Even with the pump motor (and potential short to ground) disconnected, I don't understand why voltage on pin 87, 30 and 86 drops to 0,7V as it's taken directly from the battery via pin 87

Will be extremely grateful if anyone more experienced in electric circuits could help out here and diagnose why am I seeing what I am seeing here.

Many many thanks.

[AlfaWhiz]

I was not expecting to say this, but... I FIXED IT

Finally, after many years my ABS light is going off after initial checkup! I could not believe it when it first worked.

The diagnosis turned out to be somewhat misleading. What was ultimately the culprit was relatively straightforward: bad connection on the G416 60A main supply fuse for the ABS.



What was misleading, was:
a) Diagnosis reporting error as "faulty solenoid valve relay" which was indeed not the case, but after one understands how this is verified (looking at the diagram you can see where the ECU takes it's inputs from)
b) No G416 fuse on the main ABS schematic, it's only included in the "hot wiring" section of the manuals
c) Loss of voltage only after the relay has been engaged

But nevertheless, it was suggested to me that I should check the supply voltage on pin 2 of G272 and go upstream from there, checking if it reads reduced voltage all the way. I had a look at the schematics to see where this wire leads to and ultimately found the "hot wiring" diagram, where G416 is mentioned.

The fuse (at least in Q4) is located right on the firewall, next to the coolant reservoir and next to the G56 junction (sort of screwed in and sticking out to the side of it). The two connectors holding the 60A fuse were completely covered in oxidation...

Cleaned it up and presto, ABS light goes off, voltage measures as it should everywhere!

ABS - solved. For the first time I have no warning lights on my dashboard!

[Alfan]

Awesome!!!

I need to investigate this on my Q4, when I get time to start working on it again. Hopefully, it will have the same impact...

[alfa east]

Wow, we all rejoice because you were on to this for quite a while!

[AlfaWhiz]

It's tears of joy let me tell you.

I've just been testing the system with my Marelli tester, ran the pump motor test and the solenoid valve actuation test a couple of times to flush the old fluid out, and the system is blocking and releasing the wheels and pulsing as it should. Very satisfying

Fingers crossed that you can sort out your issues as well, the main fuse should definitely be the first on the list.

[Alfamuppet]

Great news and it’s brilliant when the fix is that simple!

[Alfan]

So I finally got to drive the Q4 and having only replaced the 60A fuse (and only driving outside the garage), the ABS light did not come on!!!

So I have hopefully confirmed your observations

I did not care much that the engine started to stutter when parking it, as I think it is low on fuel, though...

So thanks a lot for finding this very simple solution

Now I should only have a few other things to fix

[AlfaWhiz]

Wow, that's wonderful news!

Looks like the ABS unit itself proves to be quite reliable, and Alfa engineers should have included this simple step in the debugging instructions too! Would save everyone a LOT of trouble.

Thinking about it in retrospective, this fuse looks like a total afterthought. Literally every single other fuse is located in some predesignated place, while this one is stuck on the firewall without any cover.

But it's good at lest that we have a simple fix for this apparently complex problem, hopefully we will only need to tend to the fuse again in about 25 years from now.

[Alfan]

Yes, the location of it does not look very integrated. Another small test run this weekend was fine, no ABS light coming on unexpected - and putting in some fuel made the car run smoother again

It was on display at the 'open garage' as well, together with another project - not mine!

Alfa on show

[Alfan]

https://www.alfaromeo164register.com/vi ... f=2&t=2039

This links to another thread with the details of how it was found that checking the ABS fuse is a good idea, before getting other ideas....