On Tuesday, November 19, 2019 at 12:35:12 PM UTC-5, danny burstein wrote:
In trader_4 writes:

Any diode worth using will have so little 'backflow' it would take
years to drain a car battery.

+1

I guess things have improved since my days working the
lab with George Steinmetz...

The other thing Danny thinks the battery is draining over many days,
while the OP clearly said it happens overnight.

My impression was more that it started off (for illustraion)
on Sunday with 100 pct charge. Not driven at all. So Sun eve
it's at 95 pct, Mon morning at 90.

(Using "20 percent" as the cutoff needed for starting)

Driven a bit Monday, so Mon eve 92 pct, Tues morn 85.

Drops down a couple more percent each day.

Finally, the next week.. Mon morning 21 pct, mon eve 23,
Tuesday morning... 18 percent. So... nada...


That's true if you don't drive the car enough to recharge it.
It shouldn't take much driving to take a battery from 90 back to 100.
They have 90a, 120a alternators. If the parasitic current is 1A,
with just 30A charging current, you could replace 12 hours of drain
in 1/30th that time, ie ~24 minutes of driving time. Of course if
the current is 3A, then it gets worse. But in my experience, they've
typically not been very large.

But the right solution here is to find the fault. I gave him some
tips on how to find it. It can be time consuming and if you have
some basic skills, you can do the time consuming part, to find which
fuse circuit it is, yourself. At that point, if you can't proceed
yourself, the dealer should be able to find the fault without too
much time. Could turn a long repair time into just an hour or two.