Resolved 2019 A7 - Blower Output Reduction due to Energy Management

[Uwe]

when a load was placed on it (KOEO), it maintained ~12.4V output.

Yeah, but this car is a 48V mild hybrid, right? If so, it has a 48V starter/generator and the "main" electrical system is 48V. So I suspect what's going on there is the 12V sub-system is being maintained at 12.4V via the DC/DC Converter from the 48V battery while in a KOEO state.

-Uwe-

 

[Uwe]

Battery sensor does not measure voltage, it cannot, is connected to one pole only.

I'm near certain the battery sensor has a 12V supply going to it and it can report the voltage on that supply wire. It needs 12V to do LIN bus communications.

-Uwe-

 

[TTT]

Nice to know. Is module 19 reporting the voltage drop between charger and battery?

 

[davisev5225]

If so, why is the battery voltage still displayed?

It isn't when I unplug the 2-wire connector from the battery sensor. All values go to "N/A" when it is unplugged, including the "Battery Voltage" IE01834 value. The only one that still shows voltage is the IE00019 "Terminal 30" voltage, which is a separate measurement performed by each individual control module.

Nice to know. Is module 19 reporting the voltage drop between charger and battery?

You've got a full block map in post #23, you tell me? I'm not entirely sure what you're asking for here, since we can see the "Terminal 30" voltage as well as the supposed "Battery Voltage" value.

Yeah, but this car is a 48V mild hybrid, right? If so, it has a 48V starter/generator and the "main" electrical system is 48V. So I suspect what's going on there is the 12V sub-system is being maintained at 12.4V via the DC/DC Converter from the 48V battery while in a KOEO state.

-Uwe-

I don't believe the main electrical system is 48V. My understanding is the 48V is limited to vehicle starting (though there might also be a 12V starter according to the SSP?) and the mild hybrid belt generator for torque assistance. I don't believe any of the control modules run off of the 48V system, nor do I believe the 48V system charges the 12V battery - rather the reverse - the 48V battery is charged via the 12V system.

Edit: Nope, you were right. The alternator/generator puts out 48V into the 48V battery, which then steps down via the DC/DC converter to the 12V needed by the rest of the vehicle.

The A7 is equipped with both a 48 volt electrical system
and a 12 Volt electrical system. The 48 volt electrical
system is the main electrical system. The current is gener-
ated via the 48 volt belt starter-alternator when the engine
is running.

 

[NEtech]

Info and video.

 

[TTT]

0.5V voltage drop is too much. Anyhow, you seem to be knowing what you’re doing, so not much for me to do here.

 

[davisev5225]

I believe my battery sensor is faulty. Of the two wires that plug into it, the solid blue wire matches battery voltage exactly and remains constant, so that should be the battery feed. The purple w/ white stripe wire fluctuates between 9.6 and 9.8V, so that should be the LIN bus. Measurements taken when it was unplugged.

Since the 12V feed matches the battery voltage exactly, but the CAN Gateway is receiving bad data, I have to conclude that the sensor itself is faulty and reporting incorrect information.

 

[davisev5225]

0.5V voltage drop is too much. Anyhow, you seem to be knowing what you’re doing, so not much for me to do here.

I appreciate that you popped in and tried to help. More eyes are always good when diagnosing complicated electronics.

 

[davisev5225]

New battery sensor 4K0915181 installed. Now the sensor value and the multimeter value differ by approximately 0.02-0.03V.

Best guess as to what happened: the sensor appears to use a standard voltage divider to measure the 12V circuit at a lower effective value so as to not fry whatever micro-controller they used. (Thanks @NEtech for that wonderful schematic & link!) One or both of the resistors must have burned out or otherwise overheated to introduce too much resistance. It might have even happened when I swapped the batteries (though this is just speculation...) - I left the sensor cable plugged in when I attached the negative terminal, so that might have introduced too much surge current through the active/powered-up sensor when all the car electronics all came back online simultaneously. This time I unplugged the sensor cable before attaching the negative terminal, waited for all the electronics to finish making noise, then plugged the sensor cable in.



BEFORE:

AFTER:

(Sorry for the difficult-to-read laptop screen, was fighting direct sunlight for this one. Multimeter reads 12.18V, IDE01834 reads 12.209V)

I'll monitor this for a few more days and report back if it seems to be "permanent". Previous "fixes" didn't even last overnight.

 

[davisev5225]

Yep, I think it's resolved. Screenshot below after it's been sitting since Saturday, and taken key off. (Woke up CANBUS using the brake/hazards/high-beams method.)

Interestingly enough, it's also giving much better readings on other battery health metrics, such as the internal resistance value.

I think we can tag this thread as "Resolved" now.

 

[davisev5225]

Update time. I was still getting intermittent symptoms of this problem, so I pulled the battery and took it in for warranty. Their handheld battery tester said it was fine, but it tested so far above the rated CCA capacity that I insisted they use a different tester, which they did in the back of the store, then they sullenly gave me my warranty. Apparently it was a bit cooked after all.

That still didn't solve the problem, so I started digging into the battery harnesses a little, and I found something. Probing with my multimeter, I couldn't get a consistent solid tone when continuity checking between the battery sensor's 12V feed wire and the electrical manifold here inside the positive terminal clamp assembly.

THIS IDIOT LITTLE DINGUS SCREW might have been the cause of my A7's electrical problems all along!

That blue wire is the battery positive feed to the battery current sensor. It wasn't even finger tight - I could spin the screw freely with my fingers. So I pulled it off, cleaned up the ring terminal and mating surface, then reattached it using the German torque spec (guten tight). I'll be keeping an eye on this one, and if it comes loose again, I'm throwing orange Loctite at it.

 

[Uwe]

That blue wire is the battery positive feed to the battery current sensor.

So it does have a dedicated battery voltage sensing feed after all?

If I was designing that system, I'd want an accurate measure of actual battery voltage, meaning the most direct connection to the battery possible for the voltage-sensing wire.

-Uwe-

 

[davisev5225]

So it does have a dedicated battery voltage sensing feed after all?

Yep, I established that a while ago:

I believe my battery sensor is faulty. Of the two wires that plug into it, the solid blue wire matches battery voltage exactly and remains constant, so that should be the battery feed. The purple w/ white stripe wire fluctuates between 9.6 and 9.8V, so that should be the LIN bus.

 

[davisev5225]

Sorry for dragging up an old thread, but another post reminded me that I never followed up here with a conclusion. Ever since I tightened up that screw, I've not had any further battery issues. Go figure.