P38 EAS constantly adjusting

There are 12hrs between this and previous photo.

bstein said:

I do not think there is a leak at the spings, but I attached some tape earlier today and we will see the result tomorrow.

Valve block was brand new. I have not opened it so I do not know the colour of the seals.

Click to expand...

The ONLY reason the compressor will keep running is a loss of air pressure. The fact that it also keeps adjusting means either there is an air leak or that air is being lost because the ECU is adjusting unnecessarily due to an inability to determine a stable height.
From your photo, it looks as though you have the car on wading height, fully extended. Airspring leaks often seal themselves when extended. There could also be a leak in the valve block itself, have you checked for air escaping from the exhaust?

The suspension set at standard ride height will drop to motorway height after 30 seconds at 50 MPH. If the speed then drops below 35 MPH for 30 seconds it will return to standard. That is normal. Suspension constantly correcting other than as aforementioned whilst being driven is not normal and would indicate leaks or a height sensor that is none linear giving false readings. Go through the checks in my EAS post in the technical archive.

I will do some more research and find out if there is any detectable leaks.

In the picture the car is in «standard» height, (calibrated to: 116,116,115,115). Do you think it is too high? What is the measurement from level ground to wheelarch on your car?

bstein said:

I will do some more research and find out if there is any detectable leaks.

In the picture the car is in «standard» height, (calibrated to: 116,116,115,115). Do you think it is too high? What is the measurement from level ground to wheelarch on your car?

Click to expand...

As said look at my EAS post in tech archive. Can't help if you won't listen. 25 MM or so drop in 24 hours is acceptable.

wammers said:

As said look at my EAS post in tech archive. Can't help if you won't listen. 25 MM or so drop in 24 hours is acceptable.

Click to expand...

I am sorry for the misunderstanding, I will read your post, do the tests and report back.

These are the parameters for setting heights. If the photo is at standard as you say it appears set to high. Set each height within these parameters but not if possible within 10 beats of upper and lower limits. Set according to measurements arch to wheel centre given in my tech archive post.
...................Front.............................. Rear.
Extended 120-180........................ 105-150.
Standard... 95-150.......................... 85-130.
Low........... 75-140.......................... 75-120.
Access.......50-110...........................50-100.

Better to get some dowel and make up the blocks for setting the heights. Cannot remember if the measurements are in Wammer's guide.

- Checked the calibration. Seems pretty good, previous owner calibrated with blocks. The car has 16" wheels so that might be why it seems tall on "closeup".
- Did as the post in "Technical archive" said the checked compressor. Seems good as assumed, since i just rebuilt it
- Unscrewed the silencer at the valve block and checked for leaks as described in "Technical archive" post. Could not detect any.
- Taped all 4 corners and removed timer delay. Will see tomorrow if any leaks appear.

bstein said:

- Checked the calibration. Seems pretty good, previous owner calibrated with blocks. The car has 16" wheels so that might be why it seems tall on "closeup".
- Did as the post in "Technical archive" said the checked compressor. Seems good as assumed, since i just rebuilt it
- Unscrewed the silencer at the valve block and checked for leaks as described in "Technical archive" post. Could not detect any.
- Taped all 4 corners and removed timer delay. Will see tomorrow if any leaks appear.

Click to expand...

16" rims with the correct tyres have the same rolling radius as 18" wheels/tyres. The car looks high, did you re-calibrate after fitting the new sensors?

bstein said:

- Checked the calibration. Seems pretty good, previous owner calibrated with blocks. The car has 16" wheels so that might be why it seems tall on "closeup".
- Did as the post in "Technical archive" said the checked compressor. Seems good as assumed, since i just rebuilt it
- Unscrewed the silencer at the valve block and checked for leaks as described in "Technical archive" post. Could not detect any.
- Taped all 4 corners and removed timer delay. Will see tomorrow if any leaks appear.

Click to expand...

It does not matter about the wheels, even if you fit larger wheels providing the correct tyres are used, the distance from wheel centre to arch should be the same. On your photo if that is set at standard, the gap from the top of the tyre to the lip of the arch appears to large. Measure with a tape the distance from centre of wheel to lip of arch when set at standard and post the result. It should be 470 mm plus or minus 7 mm. The plus or minus is so you can set the fronts within two beats side to side to prevent dancing.

No need to rename to .zip. Just put it in a zipped file. In Windows just right-click and send to zip, or use Winzip. Not sure if Lanyzone can handle zipx though. WinRar and 7zip are free. You can rename extensions in Windows Explorer but your file properties need to be set to show file extensions. Semi-colon is the european standard for a CSV file. As others have said, text to columns in Excel will split it (ALT then D then E is the shortcut.)

Attached are your CSV logs together with an analysis in Excel which has a summary trace from my car for comparison. Your pump is running 75% of the time (compared to 7% for me). The max duty cycle of that pump is about 35% so it will burn the piston seal out in short order if you keep driving it. Compared to mine yours seems to spend a lot of time filling the front bags so I'd suspect a leak there. If you're sure the bags are sound then check where the lines go into the bags and the valve block.

You can tee a pressure guage off the reservoir tank feed. It occurs to me that a faulty pressure switch could cause the pump to run all the time too.

-I taped off all the wheels yesterday and removed the delay timer relay. I see not significant drop at any corner, max 5mm in about 24hrs. I guess this would suggest that the system from valve block is leak free.
-Previous owner calibrated with blocks after replacing sensors. Yesterday I measured "Standard" from centre of wheel to wheelarch to be 47cm (+/-7mm) so I do belive the calibration is still good.
-Grrrrrr: Impressive work in .Excel, thanks!


In "standard", "highway" and "access" all 4 corners of the car dance up and down, but in "high" it seems OK.

bstein said:

-Grrrrrr: Impressive work in .Excel, thanks!

Click to expand...

Cheers.Might be worth filtering where the distance from target is at extremes and understanding what is going on at the time (like a height change request).

bstein said:

-I taped off all the wheels yesterday and removed the delay timer relay. I see not significant drop at any corner, max 5mm in about 24hrs. I guess this would suggest that the system from valve block is leak free.

Click to expand...

Taped off? Were the pipes left connected to the valve block?

bstein said:

-Previous owner calibrated with blocks after replacing sensors. Yesterday I measured "Standard" from centre of wheel to wheelarch to be 47cm (+/-7mm) so I do belive the calibration is still good.

In "standard", "highway" and "access" all 4 corners of the car dance up and down, but in "high" it seems OK.

Click to expand...

Just theorising here, you understand. I do not know the inner workings of the EAS computer. Theory: each sensor reads from 0 - 255. That has to handle Access, Motorway, Standard, High and extended high, plus some run out above and below for suspension travel and to stop the contact running off the track in the sensor. Therefore the EAS computer MIGHT have a rough idea of where the start and end of each height should be in extremis. If the sensors were fitted in such a way that 1 or more sensors dropped outside that expected range it might cause issues and start it dancing.

However, that pump running all the time is more urgent than dancing. It has to be a dodgy pressure switch or a leak somewhere, either in the valve-block or the air-dryer or the reservoir underneath the car itself. Spray all connections with soapy water (washing up liquid will do) and check for bubbles.

Grrrrrr said:

You can tee a pressure guage off the reservoir tank feed. It occurs to me that a faulty pressure switch could cause the pump to run all the time too.

Click to expand...

but would not account for the "dancing" Teeing in a gauge just provides 2 more potential leaks, it will prove nothing.
Pressure switches are not a common failure item.

bstein said:

-I taped off all the wheels yesterday and removed the delay timer relay. I see not significant drop at any corner, max 5mm in about 24hrs. I guess this would suggest that the system from valve block is leak free.
-Previous owner calibrated with blocks after replacing sensors. Yesterday I measured "Standard" from centre of wheel to wheelarch to be 47cm (+/-7mm) so I do belive the calibration is still good.
-Grrrrrr: Impressive work in .Excel, thanks!


In "standard", "highway" and "access" all 4 corners of the car dance up and down, but in "high" it seems OK.

Click to expand...

Have you checked the height sensor connectors for corrosion?

Grrrrrr said:

Cheers.Might be worth filtering where the distance from target is at extremes and understanding what is going on at the time (like a height change request).



Taped off? Were the pipes left connected to the valve block?

Click to expand...

Just regular tape from wheel arch to wheel in order to detect drop. The pipes were left connected to the valve block.

Grrrrrr said:

Just theorising here, you understand. I do not know the inner workings of the EAS computer. Theory: each sensor reads from 0 - 255. That has to handle Access, Motorway, Standard, High and extended high, plus some run out above and below for suspension travel and to stop the contact running off the track in the sensor. Therefore the EAS computer MIGHT have a rough idea of where the start and end of each height should be in extremis. If the sensors were fitted in such a way that 1 or more sensors dropped outside that expected range it might cause issues and start it dancing.

However, that pump running all the time is more urgent than dancing. It has to be a dodgy pressure switch or a leak somewhere, either in the valve-block or the air-dryer or the reservoir underneath the car itself. Spray all connections with soapy water (washing up liquid will do) and check for bubbles.

Click to expand...

Maybe I should get a T and a manometer in order to see what is going on the the tank pressure.
I did pressure test at the connection with dishwashing liquid yesterday but I could not detect any bubbles.

I reckon there is a dirty connection on a sensor which cannot settle but as it changes height it upsets all the rest which then try to get back to where they want to be

RawData tab in my spreadsheet.

Row 697. Look at the live reading in columns H. Why so low all of a sudden? Engine speed changes up, went down a gear?
Row 1839. Look at the live reading in columns E and F. Why so low all of a sudden? Engine speed changes up, went down a gear?
Row 2990, give or take a bit: Look at columns E to H. Again, low actuals. The very low value in H corresponds to the foot-brake being applied and the battery volts dropping. Revs drop down to about 550 as well which is a surprise as I would think it would stall.
Row 3065. Again, low values in E and F. Engine speed is dropping and volts do a weird thing around then.

Might be chasing data glitches but maybe not?