I have about 20k miles on my current airbags. The compressor failed (it ran consistently and became louder and louder) so I had it replaced. Unfortunately, they had some mounting issues with the replacement compressor so it's painfully obvious when it's running (they will be replacing the mounts). So I'm able to tell that it turns on for about 30 seconds every minute or so when I'm driving (in other words, consistently). The independent MB shop insists that's because there is a leak in the system. Via whatever computer tool they have, they've ruled out the airbags, compressor, reservoir - and now believe it's the hose between the compressor and the reservoir. That's a long hose that runs in and out of the cabin and will take some time to open it up.

My question is this - the car never loses height when I leave it for 2-3 days in the garage. If there was a leak anywhere in the system, wouldn't it drop? Is it more likely that one or more of the ride height sensors is sending faulty signals to the pump about when to turn on (i.e. all the time)?

Is there any part of the system that would NOT be under pressure when the car was turned off (such that a leak wouldn't be noticed)?

I'm trying to figure out whether I seek a second opinion (about 90 miles away) before I let them tear out the seats to find the hose they think is leaking (not sure how a leak develops in the middle of a hose that's well protected).

 

They are right, you have a leak but don’t tear into the car just yet.

There is no hose from the compressor direct to reservoir. There is one from compressor to valve block, and it is possible to test the integrity of it using the proprietary MB star scanner tool. Completing the line from the pressure source to the storage tank there is also a hose from valve block to reservoir. It is also possible to test the i integrity of this line and the tank as a unit using Mbstar. It is also possible to test the pump output.

That being said, these automated tests merely tell you where to hook your audiovisual meat sensors on and go looking and listening for leaks. I would concentrate on the circuit from compressor to valve block - it’s all right there on the passenger front side wheel well and look for a leak there using leak detector spray with demand set to the system - so hose it down with stoner spit and press the raise button. Since you’ve monkeyed with the compressor it is likely to be a connection or the line, or possibly the compressor itself.

then go check the next likely culprit, the rear air bags.

 

Max Blast - thanks for taking the time to respond.

They did use MBstar and they believe it isolated the leak between the valve block and the reservoir. That's why they want to pull out the seats to expose the full hose and look for leaks there. I was a little suspicious as they couldn't find any guidance from MB as to how long it would take to pull out that hose (suggesting that it's a very rare occurrence). They also tested the pump output and the reservoirs ability to hold a charge.

My main question is why, if there is a leak, would the car not sink when left for 24 hours? Since the compressor turns on every 30 seconds or so when running - I'm assuming it's a significant leak. Unless the leaking portion is not under pressure (able to leak) when the car is off?

 

The reason your car doesn’t sink perceptively is that the air pressure in the struts is held by the valve block, and your bags might be fine.

that being said you wanna absolutely rule a leaky bag out by observing sag over a period of time and I believe the limit is 10 cm per day which is very generous. It’s also possible that you have a leak in a bag somewhere that just hasn’t been uncovered by the way you park your car on a given surface. Bags have folds that sometimes cover minor pinhole leaks and beginning cracks and when you park the car in a different orientation those are uncovered in the car will sag.

Reason why your compressor is running all the time and will burn itself out in very short order, is that there’s a demand signal. That demand signal comes from one of five places. It is either from the central Reservoir pressure being too low (the central reservoir and it’s supply/output line are considered one system). The other four signal sources are each wheels airbag and it’s line, triggered by that wheels level sensor. Iirc there is a pressure sensor in each wheel bag line, but it doesn’t trigger the valve block to release reservoir air into it, wheel position does though.

All the MBstar guided tests on the airmatic does is tell you where to go physically look and listen. So they are in fact telling you the right things in accordance with process. Now the first thing I would do before I start tearing anything apart is to do a smoke check, or a pneumatic pressure check on the line running from the valve block to the Reservoir the other way this can happen is disconnecting the line at the Reservoir and doing a smoke check from that end. The intent would be to see if smoke fills the cabin then you know there’s a leak and there is caused to disassemble the interior and patch that part of the line that may be leaking.

 

Max,

Thanks for the clarification. That makes sense. I'll let you know what they find.

 

Here is a schematic of the system, drawn up in Paint when I was bored when I first got to Idaho and was trying to track down my own air system failure.




Certainly before ripping apart the interior to inspect the hose, I would put soapy water on the valve block connection and the connection to the reservoir.

The reservoir is in a somewhat vulnerable location, aft of the spare tire well just fore of the rear bumper.

Max - I believe there is one pressure sensor, which can tell the pressure in a particular zone by opening valves as appropriate. For example, to read the reservoir pressure, you open the valve to the reservoir and thus the manifold inside the valve block becomes pressurized to the reservoir pressure.

 

Yep.
That is a better and cleaner schematic than I was working on.

makes sense that there is only one shared pressure sensor.

 

They pulled the seats out and replaced the hose - long story short, no change. The system still registers a leak, but they can't find it with a smoke test or any other method. That makes me suspect the pressure monitor itself. The system is telling them there is a leak between the reservoir and the valve block - BUT, the car never lowers. The only way I know there is a problem is that the compressor turns on every minute or so when driving (and since that destroyed the previous compressor, it got very loud).

That makes me suspect that something faulty is telling the compressor to turn on. Would that be the pressure sensor? Or possibly one of the ride height sensors? How would I troubleshoot that?

 

The error codes are cryptic. I recall when I was dealing with a failing pump that the system seemed to think there was a leak at the reservoir - so I looked and looked. Turned out it was just that the pump was not getting the reservoir pressure up fast enough.

Honestly the most expeditious way to get this fixed would probably be to change the valve block and compressor together. I don't understand that the system is continuously monitoring the reservor pressure, so it does not make sense that anything is causing the pump to cycle on and off. The only possiblity I can think of is the pump is failing and so is running - overheating - stopping and cooling - running etc. But that would be a cycle time longer than a minute, and would yield a dash error message.

And no, ride height sensors won't cause any of this.

If the pump you installed is not an AMK, you may as well replace it anyway. The consensus has been that Chinesium pumps (a) are loud, and (b) last about 20k miles.

 

All,

I've been a long time delayed in thanking you for your input and indicating the final cause. The diagnostics suggested there a leak, but the smoke test couldn't find any leak - the pump kept coming on (summary of the issues described above). They noticed a kinked hose (I believe between the compressor and the valve block - so they opted to replace it. I can't say why that might fix the issue - but it's been 9 months and it's been fine ever since.

I really do appreciate Eric and MaxBlast for the time you took to help me sort through it. Very generous of you.

In other news - we're opting to replace the car (got a 2nd cylinder misfire that hobbled it for a moment, all well now - but it freaked my wife out). At $175k miles - and with two adult disabled children, we can't afford for her to get stranded somewhere. It's been a great (albeit expensive to maintain) car.

 

A kinked air hose. Never heard of that before. Those hoses are so stiff it is difficult to imagine them getting kinked.

It was most gracious of you to check back in. You are welcome; glad to help.