Howto: Duovalve maintainance

[television]

and here's the link I forgot to put abovve regarding the online manual shop

http://www.emanualonline.com/Cars/Mercedes/Mercedes-E_Class-Workshop-Repair-And-Service-Manual.html

Could be useful if you are keeping the car and like doing what you can.

 

[rodisi]

Center pin is supplied on the same circuit as the coolant circulation pump, if fitted, and comes from the control panel as far as I can see.

What voltage do you measure there?

Malcolm's point about making sure of your good ground is sound advice, and might be why the voltage seem to have changed.
But if it really was changing, for sure the outer pin voltages would have been dropping at the same time (obviously you were only measuring one at a time).
If this is the scenario, I will go for dry joints in the control panel.

I shall take readings again tomorrow and I shall look for the coolant circulation pump, wherever that is.

My negative lead for the readings was on a bolt and it was the same one used for all three readings, which were steady for the outer two and varying for the middle. But let's see what tomorrow brings.

 

[davidsw]

Reported

 

[rodisi]

OK, I was mistaken on one reading - my hand must have been shaking when taking the readings of the middle pin. For interest sake I also measured the same points with the engine OFF. Points 1, 2 and three were all 12.81V.

I've just gone out for a drive and with the engine at normal operating temperature. Full hot setting, pins 1 and 3 are at 13.9V while at full cold, only one drops to 0.2V. So again, as you gentlemen have said, I have to look in the controller section. Am I right in saying that do do that I have to start from lifting the lower console around the gear stick and then moving up to the hea control section?

BTW, in the below diagram, is item 83 the coolant circulation pump that you mentioned, Alex?

 

[rodisi]

I can't figure out why the parts breakdown lists ignore many items.

 

[Alex Crow]

I can't figure out why the parts breakdown lists ignore many items.

What is ignored?

If you measure 0.2v on one of the outer pins, this should correspond to cold air on one side of the car.
Thanks for verifying the center pin does have full voltage - the voltage you had first posted was impossible - and there is now no need to test at the circulation pump.

If we do have one valve being controlled, and one side of the car cold, it will be worth double checking those resistances of the solenoids when hot.

 

[rodisi]

What is ignored?

for example there are no part numbers for item 71, the duovalve or 83, whatever that is.

If we do have one valve being controlled, and one side of the car cold, it will be worth double checking those resistances of the solenoids when hot.

I shall do that next time I take a drive.

 

[television]

71 is the duo valve and 83 the pump.

An easy way to test the duo valve is to remove the plug and apply 12 volt across pins 1 and 2 and 2 and 3 and you should hear the valve clonk closed if all good.

 

[rodisi]

What is ignored?
If we do have one valve being controlled, and one side of the car cold, it will be worth double checking those resistances of the solenoids when hot.

Measured solenoids when hot, one was 16.5 the other 16.7 ohms. The solenoids actually look alright, there's no overheating colours.

While driving the car this time and being more discerning, I realised that even with the fan turned off, there's air coming out of the vents and I think I can hear the fan turning. I'm now really leaning towards the controller. Thinking that I may have to do that, I went to town to get a key for the radio and also bring the car up to temperature. I can't find the damned thing anywhere - looks like it's Mercedes shop.

 

[rodisi]

71 is the duo valve and 83 the pump.
An easy way to test the duo valve is to remove the plug and apply 12 volt across pins 1 and 2 and 2 and 3 and you should hear the valve clonk closed if all good.

do you know that through experience or does your PBL show those items?

 

[television]

do you know that through experience or does your PBL show those items?

That is the way that I test them.

 

[Alex Crow]

...While driving the car this time and being more discerning...

Did you note the individual left and right temperatures with both controls set to full cold?
Was one side cold and one side hot? (testing at the end dash vents with the palm of a hand will do)

 

[rodisi]

Did you note the individual left and right temperatures with both controls set to full cold?
Was one side cold and one side hot? (testing at the end dash vents with the palm of a hand will do)

Yes, I did. The side with the 0.2 reading was definitely colder.

 

[Alex Crow]

It will be worth measuring the resistance to ground of both outer pins of the socket (ie wiring back to ecu and ecu pcb components etc) with ignition off, to see if both the same.

It is looking like an output stage fault now, dry joints still a possibility.

If you could ground out both outer pins and go for a drive and confirm all cold both sides that would be good too.

 

[rodisi]

It will be worth measuring the resistance to ground of both outer pins of the socket (ie wiring back to ecu and ecu pcb components etc) with ignition off, to see if both the same.

It is looking like an output stage fault now, dry joints still a possibility.

If you could ground out both outer pins and go for a drive and confirm all cold both sides that would be good too.

This part scares me a bit. I tested the pins with the engine OFF. All pins had 12V, so I wouldn't like to put any to ground. In order to make my life easier with the measuring, I took the casing off the connector. Without the protection there, I mistakenly shorted one of the pins (don't know which one) and blew a fuse - hence my reluctance.

 

[Alex Crow]

I should have made it more clear, but you will obviously have to unplug the connector to do this measurement in a meaningful way.

Also, testing to ground with a meter is nothing like shorting to ground with a wire or screwdriver tip!

 

[television]

71 is the duo valve and 83 the pump.

An easy way to test the duo valve is to remove the plug and apply 12 volt across pins 1 and 2 and 2 and 3 and you should hear the valve clonk closed if all good.

I should have made it more clear, but you will obviously have to unplug the connector to do this measurement in a meaningful way.

Also, testing to ground with a meter is nothing like shorting to ground with a wire or screwdriver tip!

I did point that out Alex as in the above quote.

4mm female sockets work well on these pins to avoid shorts

 

[rodisi]

I'm sorry, I'm not ignoring you, something's come up and I have to attend to it. "I will be back."

 

[rodisi]

I'm sorry, I'm in the middle of decorating and today was my day for driving around picking up food for a food bank. Anyway, I've done a little drawing and trying to comprehend the electrics here.

The two solenoids have a common leg, which is always 12V so I'm figuring that depending on the setting of the heat knob, the controller puts a voltage on the other leg of the solenoid. The potential difference between the two legs is what regulates the position of the valve and in turn the water flow. In the drawing the left side has 12V therefore the potential difference is 0V on the solenoid. The right side has 0V so the potential difference is 12V and the valve is actuated.

Taking your suggestion Alex, when I can do it, tomorrow, I shall take apart the connector and plug just the middle leg in. I shall somehow connect the other two pins to ground and see what happens.

 

[television]

If the resistance is OK and near enough the same on both coils the problem lies within the control unit.

The side that had 0v should be cold in the car