Peugeot 308 alternator not charging?

[GiveMeABreak]

Excellent

[xantia_v6]

A few days later...

The charging is still not right. The alternator voltage settles down to 12.5V and stays there regardless of how flat the battery is, but even at 12.5V it will produce 80A to balance the electrical load. There are no fault codes present.

But... If I perform a BSI reset, the next time the engine is started the charging voltage jumps up to 14.1V or 14.4V (depending how flat the battery is) and over an hour or so driving slowly reduces back to 12.5V as the battery charges. If the engine is stopped during that time, it reverts to whatever the charging voltage was when the engine was stopped.

I connected Diagbox again this morning and here are a couple of BSI screen shots (when the charging voltage was low).

Note the "Status of the electrical network at generator operating" is Downgraded. WTF does that mean?

Note the "Controlled alternator regulation voltage" is 23.5 V. This voltage varies with the actual alternator output voltage, but is always about 10V higher than the correct value.

The Engine ECU does not seem to show any unusual parameters, but does indicate that battery charging is not required.

[RichardW]

Last one looks odd, regulated voltage 23.5V? Is the BSI seeing the measured voltage - wondering if there might be a high resistance somewhere?

[Huskyxantia]

Isnt there a regulator on the alternator ? Maybe that is breaking down ?

[xantia_v6]

Last one looks odd, regulated voltage 23.5V? Is the BSI seeing the measured voltage - wondering if there might be a high resistance somewhere?

The footnote in diagbox says that that is the voltage that the BSI thinks the Engine ECU is asking the alternator to produce, but the engine ECU values in diagbox don't reflect that.

Isnt there a regulator on the alternator ? Maybe that is breaking down ?

There is a regulator in the alternator, but it is controlled by the engine ECU that decides what voltage it should be producing. The BSI and Battery Status ECUs also get involved, but I have no information on the actual algorithms involved. I did try another alternator, but it did not change the behaviour.

[Huskyxantia]

This is a can of worms type of thing, would poor earth points cause problems ?

[xantia_v6]

Time for an update...

It is not fixed.

The behaviour is quite consistent, if the battery is disconnected and reconnected before starting the engine, everything works correctly for that run, with the charging voltage slowly decaying from 14.5V down to about 12.5V over about 300 km of running. If the car is left for more than about 4 hours, the charging voltage starts at 12.5V and stays there, causing the battery to eventually go flat.

What I discovered today is that pulling and replacing F1 in the engine fuse box has the same effect of restoring the charging. I don't understand the mechanism, is it that the engine ECU needs an extra reset, or is it that a relay in the fuse box is not cutting power when it should? The lack of any documentation of how the system is designed to work is very frustrating.

What exactly does F1 isolate?

[RichardW]

You'll need a wiring diagram... What's the RP no? Marc might have one, otherwise you'll need to pay the €5 (or whatever it costs now!) for an hour's access and download the diagrams. Does it go down to 300km even if this is over a series of runs (thus ruling out possible heat soak in the ECU)?

[xantia_v6]

I have an old SEDRE wiring diagram, but it doesn't give enough detail of the internal routing inside the BSM. There are also inconsistencies because on the charging diagram it shows an ECU with a single 55 way connector but this car has an ECU with 28 way connectors. Maybe the online diagrams are better?

I had not thought about heat soak, but I doubt that it is a primary issue. The voltage sag occurs on multiple short runs or a single long run.

I have checked for moisture or corrosion in all the under bonnet connectors, but they are spotlessly clean (no salted roads here and the car is seldom used in winter).

[Huskyxantia]

What sort of mileage are you driving , is there a peak of charge at a certain mileage ? Maybe just a weird theory , part of the loom could be breaking down , gets hot ? Then possbly un able to run current ? , im only recalling a car owned by a family memeber which the wipers blew fuses the windows blew fuses and the heater never worked , the loom to the heater was half shot so that got chopped out and replaced , to which the bad part was half way in the loom and only found when it was all stripped of its outer case. Earths can be the same , dim head lights for example not always an alternator issue ive found an earth corrided the full lenght , i replaced that lights were brilliant after. Hope you get to the end of your problem

[GiveMeABreak]

The engine Fusebox (PSF1) shows F1 as:

F1, 20A, +batterie, Engine ECU main relay - cooling fan slow speed relay.

Circuit

[RichardW]

Appears to be a fairly simple circuit - assuming 1320 is the engine ECU? What does it report on live data for the system voltage? Can you scope the output from the ECU to the alternator - presume it's a pulse width signal, and the duty cycle drops off over time? Can't help but think the ECU has developed an internal fault....

Edit - although if other systems are involved (e.g. the BSI) who knows what's asking for what, and where it has gone squiffy!

[GiveMeABreak]

It is, Richard.

1320 Engine ECU
1020 is the alternator
1010 is the starter
BFDB is the fuse box at the battery outlet

[RichardW]

Trouble is computers are involved....

Actually thinking about it, some of the data above showed the BSI voltage to be 23V or something, so if the BSI is 'seeing' a high voltage it might be directing the ECU to cut output.

[xantia_v6]

The control line from the engine ECU to the alternator is LIN bus, a digital signal. It produces an error if it is unplugged. I also tried connecting to to a spare alternator (not mechanically coupled to the engine) and no errors until I started the engine, and it then threw a "mechanical error" because the alternator was not reporting a sensible speed.

The battery voltage reported by diagbox (which I think is sensed in the BSI) always reports about 300mV above the real battery voltage, but I think that is just a calibration error, not a fault. The 23-25V reported value is what the BSI thinks the engine ECU is requesting of the alternator, but the engine ECU reports a sensible value that matches the actual battery voltage.

I will do some more tinkering today and see if I get anywhere. Unfortunately there are no 308s (or PSA vehicles with the same architecture) in local breakers yards, and the nearest I know of is over an hours drive away, and they seem to have a random number generator for pricing.