Thanks for the explanation, and what does the "passive immobiliser" do ? (the one which can be deactivated vie the Nanocom ?

So, just for my understanding, the P38 does not have the "immobiliser" which consists of a small transponder in the key case and a pickup coil around the ingnition key slot ?

Yeah, right, but the lights go on when I open any of the other doors.
Just not the driver (LH) door....

Yes, I agree in that it looks like water condensing on the cold roof metal.
But the "wetness" of that area is so dynamic in that it will be totally dry when there has been no rain for a while and then soaking wet immediately after strong rain. And the more it rains the further the wetness creeps towards the middle of the roof, strting from the corner. And why would it only condense only in that small area and nowhere else ?
I have now removed the A-pillar trim and the windscreen top finisher (which would impossible to figure out how, if one didn´t know the trick) and have put a bead of silicone (ran out of caravan sealant) around the corner of the windscreen.
Let´s see if that changes anything, lots of rain forecast for the next days....

I also like the idea of having a spare key hidden somewhere on the car but I never did it because it would be missing the transponder for the immobiliser...did you get one and attach it to the spare one or did you disable the immobiliser....?

Is there a way to remove the top windscreen and A-pillar trim without breaking it to maybe reseal that area ?

as Brian said, if you cut all the wires you would not have the petrol injectors connected to the engine ECU any more since they get wired through the LPG ECU when installed. That is how the ECU measures the petrol injection time.
I would just remove the cable going into the fusebox (+ve feed) and leave the ECU and cables as they are. You can of course remove the cables leading to the hardware lpg components that you are removing.

Hello,

another problem which is gettings worse as the rainy season in northern germany is gettings stronger:

The headlining (re-clothed by be about 1 year ago) is wet at the front left (driver) corner, from door side of A-pillat upte the middle where the lamp is. So the whole strip behind the sunvisor. Sometimes dry (when weather good) sometimes damp, or when really raining, soaking wet with occasional drips.

I checked the sunroof drain, it are free of debris, transport water and is tightly attached. All tested.
I lifted up the cover of the foor rack attachment points, all good there.
I stuck the rubber door seal (in the corner between A-Pillar and roof) which one sees when door open back onto the metal body.

Couls it be a bead sealing of the windscreen ? I have no other idea.

Here some photos I just made looseing the headliner in that corner:

greetings, Max.

Hello,
a sidekick of my EAS problems is this:

Nanocom tells me this:

in EAS/Inputs Menu: it does not say that the driver (LH) door opens. All others are recorded correctly. Therefore EAS continues to level with that door open.
in BECM/Inputs Menu: it records correctly that the driver door is opened (12V->GND). All others too. But the interior lights do not go on. But with the other doors.

How can that be?

Microswitch in door must be OK. And EAS ECU gets that info from the BECM, but the information is lost....?

greetings, Max.

One thing I would like to add:

The ride state "levelling" which in my case occurs while driving and sends the car to the bump stops (preceeded by the stae "ECU wakeup"), is the state the EAS system is in when ignition is ON but engine OFF.
So would that be the state of "self-levelling" which occurs every couple of hours?

That could mean only two things:

Either there is some impulse from the delay timer relay to the ECU while driving (which would be some kind of fault in the timer circuit) OR the EAS ECU has some kind of fault going into that state uncommanded.

I still have the faint theory that the ECU is behaving erraticly and is therefore faulty.
But of course I am hoping tightening the 13-way connector in the EAS box cured the problem.
As long as the system works fine I don´t care how weirdly the ECU is behaving internally...sounds very human to me :-.)

@Pete reassuring to know the system can work perfectly with the "invalid fault code"

Hi just wanted to give a small update:

I tightened the connector pin sockets in the 13-Pin Connector which is in the EAS Box.
I didn´t mechanically check if they were loose, but I could see that the sockets were not too tight.

The drove for about an hour, logging all the data, no umcommanded height changes, system operating as intended.
Then went through all the log thoroughly and found the actual heights to be much closer to the target heights than before, also the rear left (which was mostly too high in all of the states) is now reigned in and much closer to the other corners.

There is still, though, the "invalid fault code", (which remains after clearing) but if that is what it needs to function properly then the system may happily keep it :-)

I am still only cautiously optimistic as I have had periods of error-free driving before but it would be great if that trick worked. I will continue to monitor the situation and give feedback here.

I will start another thread for the "door ajar" mismatch between EAS ECU and BECM, (and other woes), I guess for future reference it is best not to mix up topics too much in one thread....

greetings, Max.

I have now done several test drives with the Nanocom logging the data and have ploughed through numerous log files with endless number columns and many many graphs, comparing all data and how height changes ate influenced by them....
The cars I had until now mainly needed a set of spanners and a welder, but with this one I am staring at excel spreadsheets for hours...^^ Well, that´s evolution for you....

There are more ride states aside from Extended, High Profile, Standard, Low Profile and Access, namely "ECU Wakeup", "Levelling" and "Dormant". They are mentioned in the Blackbox Solutions Manual, but for example not what they really mean.

What I found out is that "ECU Wakeup" can occur at any time while driving, always after (in my case the rear left) one corner is too high, the corresponding valve and inlet valve have opened, but no change in height has occurred. (This can of course also happen without resulting in the ECU Wakeup Mode, so I am not sure how the ECU decides to do so.) When the ECU goes into ECU Wakeup, all valves are closed. This Mode is always followed be "Levelling", which causes all corner valves and the exhaust valve to open, the ECU sets Target height to 20, and the car goes all the way down to the bump stops (hard fault). This will stay. As fault I will then always have the "invalid fault code".

The other state is "Dormant", which can also occur at any time, but when one corner (again, rear left) is too low, the corresponding valve and inlet valve are open for some time and no height change has occurred. But in this state the ECU continues to function "normally", i.e. opening and closing valves according to target height, and no sinking to the bump stops.

Another, maybe not connected, thing is that when I open the driver (left) door, the inner light does not come on and the opening of the door is not recorded by the EAS ECU, so no inhibiting of function. But in the BECM Menu, I can see that the "Door ajar" switch is functioning properly. As far as I know the BECM send that signal to the EAS ECU....could that be related and point toward some kind of intermittent electrical problem or is that something else...?

BTW, I have not checked/tightened the Multiconnector in the EAS box, that is still to be done....

hope I am not boring anyone but I am trying to sort this all in my head and writing it helps too :-)

greetings, Max.

P.S. The forum response time seems extremely slow since yesterday or is that just me?

I marked the spreadsheet starting at the left top corner, marking the first couple of columns down to the end, then right-clicked "fast analasys (?)" (Schnellanalyse in german) and then one can choose what kind of graph excel shall produce....
But the Nanocom csv viewer is a lot better to analyse the data because you have many more viewing options. You just have to convert all words (on/off, open/close, etc.) to numbers. But that is easy with Ctrl-H (find and substitute).

So I spent some hours going into the details of all the logs in excel and converted them so one can see them graphically in the csv Viewer.

The trick is to convert any "words" in the spreadsheet (like on/off, closed/open, standard/high pro, etc.) into numeric values (e.g. on=1, off=0). Then you can open the csv file with the Nanocom csv viewer which is great to see every detail although the handling of the program is pretty clumsy.

Here are the logs directly from the Nanocom (the ones I created for Post #5) Here one can read all the switches and valve positions:
https://c.web.de/@337510546125364120/VGPX1iUQQ0yyaWKdqnPx3g

As they are not images, one can not upload them to imgur....

And, in case anyone is interested, the csvs converted from "words" to numbers:
https://c.web.de/@337510546125364120/G8MjRHxSRq6vZhwm2agvLQ

My conclusion from going through all the data:

• Some of the time the system works as is intended, user inputs become the correct valve actions which lead to correct heights. (log EAS1)
• Some of the time (usually resulting in an EAS error) the user inputs become the correct valve actions but do not lead to the heights changing as expected. The only explanation that comes to my mind is that the valve actions of the ECU do not actually take place as intended, therefore the target height is not achieved. That would lead me to believe that the driver pack (or cable connections between ECU and valves) does not drive the valves properly. Since the ECU has no feedback if the valve is actually opening/closing or not.
• Another thing is that the on/off of the compressor is not in sync with the opening/closing of the pressure switch. Does that have to be?

Sorry to have dug this deep into this EAS hole, but I would like to know what is happening and found the log function of the Nanocom to be invaluable and new to me. Throwing new parts at the car and in the end not knowing which actually cured the problem is very unsatisfying...
Tomorrow I will be doing a longer drive with trailer, I will log the whole trip, there are bound to be surprises then, so I might catch something interesting on the log then.

greetings, Max.

OK, I finally managed it, tried it on the phone but wouldn´t work, but did on the computer.

Now updated Post #5 with the graphs.

OK, I tried doing that via imgur, but I don't understand how imgur works, it doesn't seem to be a filehosting Service but some Kind of "instagram" type thing, or?
So I create an account, post the picture and theb share the link or how does it work...?

Hm, if I copy the entire Link and paste it into a new browser tab it will show a thumbnail of the graph and give the options "herunterladen " (download) or "öffnen" (open/show).
Is there a better way to integrate pictures into this forum ?

I had the log files "closed" for quite a bit as I left the Naocom on until I was back home from the test drive....Maybe it didn´t work because I have the 1.37 beta version on the Nanocom ? Changed back to 1.36 and will see if it works then....

I am amazingly bad at excel but making a graph from the numbers was surprisingly easy :-)

OK, so calibration would be necessary, will have to make up some blocks then...

Thanks, Max.

OK, I did a test drive with several height settings and logged them on the Nanocom (didn´t know that that worked :-)

I tried the CSV Viewer Program from BBS but the program would not open the csv files, saying they contain invalid charactes and no log data, (any ideas why???) but I managed to convert the raw height sensor data to a graph with Excel. Here are three test drives:



The bottom four colours in the legend correspond to the height sensors described on the left hand side.
On the second drive the EAS went into Error mode and in the log the status went from "high profile" to "dormant" which is probably when the actual heights do not change as the EAS expects them to...but it didn´t go to extended mode though...

What I make of all this is that sometimes the system just does not react to the target heights and sometimes it does.
The front left sensor mostly sems to read too high and the rear left mostly too low.
There are also some peaks (mostly on the left side) that may show the sensor sending a wrong signal. But it was a bumpy road so I don´t know as to how much the signals fluctuate normally.....

I guess next would be to swap the sensors from side to side....?

Oh yeah, the connector from driver pack to solenoids looks clean, nothing obvious there....

greetings and thanks, Max.

What a conicidence, today a new fault came up: "Rear left height sensor out of range" so you might be right there, thanks....:-)

I still have some questions though:

When I drove today it did all kinds of weird stuff up and down. and even though the rear left was very high most of the time, the height sensor reported this correctly. The system war very erratic but the reported height at each corner always corresponded to the actual heights. Only the system didn´t react when one side was very high.

Also when the car goes to high on its own, I suspect it is going in the "super-high" mode, or is this the normal "high" height in this picture:
https://c.web.de/@337510546125364120/NhK4NfUiShKmRFb1PF9QDQ
I measured the height on the rear from mid-hub to wheel arch and its 57 cm (normal high is supposed to be 51 cm).

Also when I open the driver (left) door, the Becm will report door open, but the EAS will not detect an open door. Other doors report correctly.

Maybe I´ll also look for a used EAS ECU....?

greetings, Max.

Hello,
My EAS is giving me problems and I don´t know how to proceed further. I would be great if I could get some hints....I have read about every manual, document tand internet post there is for the EAS but I am at a loss now....

Since I have the car (2 years) it will rise to high without me commanding it. Sometimes every couple of minutes, sometimes once a month. Also when the inhibit switch is on. I could usually tell it to go down again, It would also give me faults (usually "pressure switch open") with the same regularity. I have a Nanocom.

I removed both multi-connectors behind the left and right kick panels and soldered them. They still look dry and fine. No change in behaviour.
Car will not lose height overnight, but will often be too low at rear left (new airbag) when I park. I replaced both left airbags when I got the car, of the remaining ones 1 looks like new, the other OK.

Since a couple of months the usual fault changed to "invalid fault code" which I can clear but will come back regularly, sometimes doing nothing, sometimes leading to soft fault, sometimes to hard fault. Usually when the fault comes the car will be high at the back and low at the front...

Compressor was loud for some time but is quiet again now and works well (fills empty system in under 5 minutes). The green/purple cable leading to the compressor was burnt at the connector but still worked, fixed that.
I just refurbished the valve block but that made no difference other than now the rear left, which was usually to low when i park, is now too high.

Other new observations:
Compressor will run even when engine is off.
Height will change even when door is open ! That is new. But Nanocom confirms that door is open, so no lock switch fault...

I can switch on/off all valves with nanocom and hear them engage.

All that comes to my mind now would be the ECU or the driver pack.....any ideas or further questions would be very welcome before I lose faith in this wonderful car....

greetings, Max.