After three days of business travel, last Saturday saw me return on the airport, tired after an overnight flight, only to find my green cat with a battery so dead, it wouldn't even blink a single light let alone open the doors to welcome me when I pressed the button on the key. After good 45 minutes waiting for the service guy from ANWB we got her running, but along the procedure measured that the car burns an 0.8 A (!) creeping current when the beast should be sleeping. I also realized that the car has never been standing without being moved for more than 2 days since I got it, so no wonder I never had this before.
After I took the whole afternoon today to find the leak unsuccessfully, I am not sure where to continue now and hope, someone here has a clue what might cause this. To feed you with all there is to know, this is what I have checked and found so far:

I hung an amperemeter between ground cable and battery pole.
To make the car think it's closed I closed all doors, pushed in the trunk lock to shut off the trunk lights and locked the whole thing with the remote. All went dark.
Initially there was a 1.9 A current for about a minute, then dropped to 1.2 A and further down to 0.9 and finally stabilized around 0.8 A after good 5 minutes, when all the little gizmos had fallen asleep one by one. Just the last 0.8 A remained, I waited for more than 15 minutes if this last gremlin shut off, but it wouldn't die.

some system still hungry 15 minutes after shutdown

I then took to the three megafuses in the trunk to see into which rough direction the electrons would disappear. It was all going into the circuit behind the 100 A megafuse on top. After disconnecting this cable the reading went to 0.

100 A upper megafuse taken out, amperes quiet at 0

According to the circuit plan that I found here in the forum, this 100A supplies to the passenger junction fuse box, but from then it also somehow supplies to the front and the rear fuse boxes. Well, that didn't exactly help, so I made my peace with the idea to pull every single fuse (and relais) and check for reactions on the amperemeter. (Who ever tried to squeeze around the passenger seat to stretch under the glove box and pull every single of these 5 rows of fuses can understand I wasn't happy.
Long story short, I pulled every single fuse and relais from the 3 fuse boxes and the only thing I found was:
rear fuse box #52 reduced the leak by around 0.2A to 0.6 .. 0.7A
rear fuse box #16 had no effect when pulling it, but leak increased to 1.2A when plugging back. Over a few minutes it dropped back to the usual 0.8 .. 0.87A again
rear fuse box #47 dropped by 0.1A
passenger junction fuse box #28 dropped the current by almost 0.1A
So altogether is detected approx. 0.4A distributed over 3 fuses, that's all. And even if I could eliminate these, remaining 0.4 A creep would still not allow to leave the car with its 100Ah battery standing somewhere for a week and come back to a living cat. Still a magnitude too big for my taste. (BTW, what have other fellow XJ8 drivers here measured as creep current on fully deactivated cars?)
I also unplugged all electric connectors to the LPG system, no effect
Found some hidden fuses in two little boxes on the port side of the engine bay (not sure what they are for, haven't found in documentation), none has an effect.

uncharted fuses in 2 sealed boxes on the port side engine bay. anybody know them?

So hail to the potent generator and my daily 20km distance to work which covered the problem quite well so far, but I really need to do something about this starvation after only a few days.
The only thing I installed so far was a USB charger / voltmeter combination in the place of the original cigarette lighter, but I connected that to the original cables for the cigarette lighter, and the respective fuse #7 in the passenger box had no reaction at all. Plus, there is no power on the USB ports when ignition is off and the voltmeter is dark, so it shouldn't be the culprit, I even checked before installation that the front cigarette lighter was switched by ignition for that reason.
In case it helps to draw conclusions, I already had a problem with the turn signal lamps twice, only thing electric that happened on the car before. Once the rear right bulb spontaneously blew when switching the signal. After change of bulb no repeat.
And recently the fuse #25 in the passenger box was toasted when signaling, leaving right front and side indicators dead. Changed the fuse, all works again.

I'm really hoping that someone here with a bit more first-hand experience of the electric flows of the X350 would take the time to point out some more ideas, what to look at.
Thanks for all efforts already in advance.

 

You need to wait about forty minutes (if I remember correctly) for the timed supplies to switch to zero volts.

 

An intermittent diode failure in the alternator?

 

Wow, 40 minutes, that's long. What would my car have to think about for so long after I park it? Thanks for the tip, I'll extend my check to 1 hour then, when sleeping beauty is dreaming so long ...

 

I finally managed to log the idle current after switching the car off, it all looks healthy, so I think I have to live with the uncertainty of not knowing the real reason for the empty battery. Curently, at temperatures below freezing, it starts like charm every morning.

Just in case anybody still wonders what idle current an XJ8 has, and especially what timing it has to shut down the little devices after parking the car, I attach my log. As I'm not willing to pay the amount that any real current logger would cost I used my digital camera in interval-picture mode, making one automatic picture of the connected amperemeter in every minute in the closed and locked trunk. The whole movie stretches over approx. 1 hour 20 minutes in reality, so from the timing of every change in current you can conclude what time has passed in real.
The sequence goes over 4.2A right after locking it, over 1.8 A then 1.2 A and after a few minutes to a longer time stable 0.7 to 0.8 A.
At 30 minutes the current drops to 0.2 A for some minutes, then after around 40 minutes it drops to the final level fluctuating between 0.01 and 0.09 A.

Maybe this helps someone in a similar situation to check if their car is healthy or not.

What I still don't understand hoqwever is, that the 0.8 A that are present for such a long time seem to not be routed over any fuse. I had pulled them all during my investigation and none of the fuses made it drop. Well, some things don't need to be understood as long as they work ...