[HugoCountsto7]

I'm lost.

[galahad05]

Those Voltage spikes imply something wrong with the alternator or voltage regulator.
Yet, both of these guys were replaced.

Is the alternator pulley and serpentine belt new as well?

[galahad05]

Still....it also feels like an intermittent ground or positive connection around the alternator. Like a flaky ground. These things are so hard to find sometimes....
I suppose you could wait until the car has the problem you're trying to track down then get a multimeter to see if you have connectivity from the negative/ground terminal of the alternator and actual chassis ground. (multimeter in continuity mode)


I'm definitely reaching for straws here.

[galahad05]

Well, let's be explicit here: I'm not familiar with your particular model's alternator. Was the voltage regulator replaced? I ask this, because, seriously, the symptoms have all along best matched a bad voltage regulator. I went off onto other possibilities (however unusual), because I just assumed that the regulator was swapped out for a new one when you said they replaced the alternator, and the odds of two different voltage regulators/alternators in a row being bad in exactly the same way were pretty low....

[PhaseP]

Check your alternator pulley, it is an overrunning clutch interesting type pulley. If you remove the belt and block the alternator shaft, the pulley should spin one way freely and smoothly, but lock up the other way. I know you replaced the alternator, but if it is refurbished they might not have replaced the pulley. It is either the alternator causing the 18 V spikes, or you have some big current going through an inductor in some module that gets its current interrupted (sudden open circuit) causing voltage spike. Don't know what module could do that.
When you parked your car it is idling, so less load on the belt and pulley.
Just something to check.
With the BDS from ECU connected, ECU was trying to correct the 18 V spike by telling alternator to reduce its output voltage, and that was causing 8V output until ECU tells it again to get it back up to normal.

[HugoCountsto7]

Thread closed

[PhaseP]

You have a 320d, it is not common in US. I saw at realoem your engine has a mechanical water pump. Since your problem started with coolant leaking and also only experienced after car warms up, you should look at there. Did you bleed the coolant good after the leaking line replacement. Wonder if water pump is air locking intermittently.
For the voltage to spike up to 18 V like that, you need a voltage source. Battery cannot spike its voltage. What is left is alternator as source.
A big inductor with high current flowing which current cut off suddenly can cause voltage peaks also, but I don't know what module could have that.
Is it possible with your logging tool to log other variable so you can correlate what else happens around the time voltage spikes?

[Levanime]

interesting thread and interesting posts,

following,
keep in mind that battery is stabilizing voltage too, so to have that kind of spike, source should be really strong.

DME can be replaced with used one and programmed, u need to find someone in Germany,
if u cant msg me and ill help to find someone, they will clone ur dme.

also visual inspection of DME is not enough, there may be micro crack that u can not see by eye.

fact that u'r getting lights is because modules are turning off when they see voltage off limits.

my 2 cents

[galahad05]

Ah, Phase is talking about Load Dumps. Could be.

Lev is also right.

[PhaseP]

Another possibility there is no voltage spike but ECU is measuring it wrong intermittently.

[galahad05]

The voltage spikes w/o the control line in place strongly suggests that the problem isn't the DME. At least not directly.

I suppose it could be the water pump like Phase said--if it's mechanical I don't know that the DME can directly tell what's going on with it (i.e. no shadow codes).

I could look at my bentley manual again, but I don't think it covers that model.

[galahad05]

Originally Posted by PhaseP

Another possibility there is no voltage spike but ECU is measuring it wrong intermittently.

I suppose that's possible. You'd have to measure it at the battery with a multimeter that could store peak measured voltages. Or some sort of oscilloscope.
EDIT: Looks like Amazon has one (multimeter) that could do the job for about $35.
https://www.amazon.com/Digital-Multi...eter+peak+hold

I'm sure you could find one locally for something like that price.



It really feels like bad wiring though. Especially with the bit about the problem starting soon after the coolant leak (corrosion on wire contacts).

[HugoCountsto7]

closed

[HugoCountsto7]

Originally Posted by PhaseP

Another possibility there is no voltage spike but ECU is measuring it wrong intermittently.

I can measure the voltage swings with a multimeter directly on the Battery terminals and at the Alternator Positive terminal plus a grounding point. I believe the voltage swings are real. I think with the ECU plugged in with the BSD cable, that the ECU see's this abnormal voltage and attempts to instruct the Alternator's Regulator to step down the voltage. Then they get into a fight, pulling down and pushing up the voltage. As long as I remove the ECU from the equation, the alternator is not getting interference, and I see the root cause of the issue.

[HugoCountsto7]

Originally Posted by galahad05

It really feels like bad wiring though. Especially with the bit about the problem starting soon after the coolant leak (corrosion on wire contacts).

So to this point what all could be affected by the leak? What I see in the near to the leak position:
Starter Motor is not directly under the leak, but is below and to the rear (near firewall)
Alternator is absolutely directly below the leak spot (keep in mind new alternator has not been leaked on)
Engine to Chassis Ground strap is not directly under the leak, but is below - to the back - and more toward the wheel.

Beyond these major points of concern I am not sure.

What about adding a direct ground wire for the alternator instead of leaving it to the alternator case and engine block to conduct the ground?

[HugoCountsto7]

closed

[PhaseP]

The air in coolant causing this probably low possibility. I think it is a loose connection carrying high current. Check the connections at battery or any other junction that will be carrying high current. Road bumps may be shaking the loose connection making it more obvious while driving, where ever it is.

[PhaseP]

Also any chance these coincide with ac usage?

[HugoCountsto7]

Originally Posted by PhaseP

Also any chance these coincide with ac usage?

No, the problem occurs with or without AC. Also, bumps or sudden stops do not have any impact on starting or stopping an event. The battery cables are tight at their terminals. Ground strap was good, and a temporary was even added in parallel (didn't help). Wiggling Main positive and ground wires in the engine bay or at the battery in the trunk have not been able to start or stop an event.

[galahad05]

Electric fan I assume?

[maxoutput]

Might also be worth checking out the B+ cable behind the fuse box as they are known to come loose and either cause gremlins, the car to shut down or turn the cars dash into a Christmas tree. Have you checked the wires from a alternator plug for continuity leaks between them ?

[HugoCountsto7]

closed