DavidI

My driver's side servo will only work to cold - when I swap left and right servos the right hand side works fine, the left only ever goes to cold. I've worked my way through Adrian's book and Roly's very helpful article on www.porche964.co.uk and assume (please feel free to correct me here!) that my left-hand wiring is damaged somewhere; when power needs to be applied to 'reverse' the motor (i.e. drive the left servo arm clockwise) the necessary power is not getting through. This appears to be confirmed with a multimeter where for the most part pins 4-5 on the block connecting the servo to power show no reading. Connecting the servo to 12v out of the car I can drive it both ways, as expected.

Is it possible to work out where the wires going in to the servo power block come from or go to? The wires from the block get wrapped in insulation tape and disappear into the bowels of the trunk - I'm hoping they make a reappearance with another fat chunk of cabling at the right side of the heater gubbins area but that is pure wishful thinking....

Are wiring diagrams / locations available anywhere or is this one for the OPC or specialist?

Many thanks,

springer3

The factory manuals are available, and the electrical book contains well-executed wiring schematics that document every wire and connector in the car. There is a locator grid on the wiring diagrams for finding modules and connectors.

I'll bet you can track down the suspect wire to each connector and isolate the break location.

Good luck

Jeff Curtis

Just a S.W.A.G. at this point, but I would suspect the climate control unit...it sends the needed signals to the servos to open/close the mixing chambers and it's not unlikely that this unit has gone South in your case.

BUT, if you're like ME...in this case, one may want to prove the trouble vice just "Easter egging" and replacing suspect parts.

Springer has decent expectations with his troubleshooting suggestion.

I'd be interested to see if in fact you locate a wiring/connector trouble as that suggests either corrosion or there has been some work in that area in the past??

mojorizing

Hi David,
I'm not an expert in porsche HVAC....

But when you said you swapped servos (good to suspect) did the servo cycle to the cold position or did it not move at all? Could it be that the temp. sensor on that mixing chamber is faulty? It could be that the faulty sensor is telling the CCU that it's too hot and positioning that mixing chamber servo to full cold. Remember that there's a feedback potentiometer that tells the CCU the position, therefore it should regulate between full closed and full open as needed.(connections 1,2,&3) I think simply repositioning the suspected servos arm to a different attitude, then powering up the car and watching the servo cycle to close could help here. If it moves to cold, then the CCU is not suspect and I'd look at the corresponding temp. sensor. If it's dead, then suspect the driver circuit in the CCU or wires leading to it. I got the schematics on CD if you need them in PDF format. The servo leads directly back to the CCU, but I think the temp. sensors have a common leg that must go to a connection somewhere ( haven't really looked yet) . The pinouts can help to do continuity checks. BTW, Randall G. is the resident expert in this area. Do a search and you'll see his input on a lot of HVAC problems.

The failure documented here CCU driver failure was due to a stuck servo overloading a IC driver chip. You don't have a stuck servo, but you never know what could happen. These cars must be getting closer to their MTBF threshold. That's what makes it so exciting.......
later, Kevin

Jeff Curtis

Good call Mojo!

I would've never considered that, a definite possibility and a ray of sunshine as opposed to changing out the CCU on a whim.

DavidI

Thanks all, some really useful ideas. My logic was too simplistic - I was assuming by moving the temp dial to red dot then the servo would move to the fully open position; I hadn't considered a faulty sensor. Does anyone know, would a faulty temp sensor telling the unit to close over-ride a red dot telling it to open?

Another snippet of information that didn't seem important to mention before but may be now (sorry!) - this all started with a faulty servo. I removed it and tested it and I could get the motor to turn when 12v was connected to the motor terminals, but it would NOT turn through the circuit board.

I bought 2 replacement servos, by the way.

If the servo circuit board did die, the motor would have turned when I applied voltage directly but not via the circuit board and it's lack of movement could have burnt out a chip in the CCU as Carlo found.

When I plugged one of the replacement servos in, it moved from a mid position to cold. Would you agree the problem is likely to be either Carlo's burnt chip or a dodgy temp sensor? Can the temp sensors be swapped? This would seem to be the best next step, if possible....

Thanks again for the help!

mojorizing

Hi David,
If you put a known working servo into your driver side location and it cycled to the cold position, I would say that it's not the same problem as Carlo's. In Carlo's situation the chip overheated to failure. The seimens tca2465 IC is a DUAL opamp and it's configured as for as I can tell in a bridge circuit, and therefore requires both amps on the same chip to be intact. I don't believe it could move a motor if one amp is fried.

I've had my CCU apart to check it out because my windscreen vent servo doesn't operate. It'll cycle open and close with an external +/- 12VDC, but it's not getting the voltage from the CCU. No burned IC's in the CCU so I live with it with the damper in the mid postion...I get a air to the windscreen and the center vents. My servo (#1 in Roly's website) compares it's position to the top slider adjustment on the CCU, while yours looks at the temp. sensor in the mixing box. Let me think about this for awhile. I could measure the ohms of my temp sensors at room temp at the pins of the CCU and get back to you for comparison.....

I gotta go to "OFF TOPIC" and bat a few penguins, then it's off to a dinner date!
later, Kevin

Randall G.

Hey David,

Kevin has a good idea, checking the temperature sensor's resistance (thanks for the kind words, btw, Kevin).

As Kevin suggested, you can take the resistance measurements at the connector in the back of the CCU. Right temperature sensor, between pins #G24 and G#18, left temperature sensor between pins #G23 and G#18. Resistance values from the shop manual:

0C (32F) = 30.6 - 34.7 kOhms
25C (77F) = 9.5 - 10.5 kOhms
50C (122F) = 3.4 - 3.8 kOhms

Though, in your case, I think you're really just looking to verify that the left & right readings are similar.

Best of luck!

DavidI

Hi Guys,
Thanks for the further input. Measuring between the pins on the connectors with the CCU removed I get:
G23-G18: 16.4
G24-G18: 17.4
This is with the multimeter set to 200k Ohms (If I was back at school my multimeter would have been confiscated for abuse as soon as Adrian's book came out - I am an electrical novice but having great fun measuring whatever I can get my hands on). The temp here in North London (again according to my multimeter - these things should be compulsory) is 8degC.

I have no idea if this is of any use at all but measuring between pins 18 and 23 and 18 and 24 on the back of the CCU (with all cables removed obviously) I get identical readings of 1.778 (multimeter set to 2 Ohms).

If I get a chance this afternoon I'll open the CCU up and see if I can spot anything that look like it could do with measuring

Sten

David,

This may save you some time.

To open up the CCU:-
1. Undo the two nuts that hold the fan on and remove it with the tube
2. Push (force!) the black clips that hold on the front panel (3 on the top, 3 on the bottom) and disconnect the plug - it's keyed so you can't put it back incorrectly.
3. Undo the Philips screws near the rear of the unit (two on the top, two on the bottom)
4. Push on the connectors at the rear and the circuit boards slide out. There are two boards, connected by solid wiring.

Inspect for dry joints/burnt components. The IC's that drive the servos are no longer manufactured; originally made by Siemens, part no. TCA2465. They can be obtained from surplus component suppliers, I've been quoted prices between £8 and £15.

Hope this helps.

Regards,

Barry

DavidI

Perfect timing Barry, thanks - am just about to open it up and have a look.

Cheers,

mojorizing

Hi David,
I got 20.9K and 20.3K ohms for my temp sensors at 44 deg. F - seems in line with the specs. This is measured from the plug at the CCU.

While the plug was off I measured into the CCU socket and I get 4.9K ohms between G18 and G24, G18 and G23. This is with the sample fan, K and G plugs, thin wiring harness from front panel all disconnected.... That's a notable difference from your reading.

I was wondering why porsche built this car with two mixing boxes ( with associated plumbing)? One mixing box and thermostat is the norm isn't it?

later, Kevin

DavidI

Hi Kevin

I get 4.8k ohms between G18 and G23 and G18 and G24 - not sure why I got a different reading earlier but the unit is on my coffee table now and I'm confident the figures are right!

The thermostat on the back of the CCU is the only one associated with the heating and ventilation isn't it? Until your question above (and seeing the unit in front of me) I assumed there were two, hence my theory about the left-hand thermostat being duff. Obviously that cannot be the case In which case I echo your thoughts - why the two mixing chambers? Seems unnecessarily complicated.

I'm going to try plugging everything back in. My only glimmer of hope is that the smaller of the plugs into the CCU was marginally lose. Given the fact that there is only one temp sensor feeding the two mixing servo's it looks like my problem is the CCU; the wiring all looks original so seems unlikely to be broken.

One other thought - there are no such things as 'left' and 'right' mixing servos are there?

Cheers,
David (who batted his penguin 321yds)

mojorizing

Hi David,
321 yards is great in golf, but not batting penguins...better switch to Titilist penguins.

Let me paraphrase what is outlined in Adrain's book. After leaving the heat exchangers, the air is drawn up either side by the forward blowers , along ductwork on either side of the car. In the luggage compartment, the left and right heating systems are identical but separate. The heated air is mixed in it's respective mixing chamber and servo , and measured by it's respective mixing chamber temp. sensor. The tepered air is then sent to the vent distribution system. The interior temp is measured by the CCU sample fan arrangement and set by the control ****.
What is happening then is this. Car interior temp. is a little cold sensed by the CCU sample fan (setpoint is the control ****) System actuates both servos (drivers side and passengers side) to mix in more heated air and less outside air till mixing sensors detect proper temp. The air comes out of the vents warmer, which warms interior till setpoint is met. There must be a PID function (propotional/integral/derivative) and feedback control for this system. If not and the system has only one sensor, hot air would blow from the vents when interior is below setpoint and cold air would blow when above temp. setpoint - essentially making the interior of the car the mixing box. The weather is bad for the next few days, and I'll look closer at my CCU for more clues. It can't be rocket science.... come to think of it, if you manage to fix this give NASA a call and tell them you were able to fix the porsche CCU and you could give a shot at fixing the Mars Rover.

DavidI

The Mars Rover is a bit out of my league; a Mars Bar* is probably about my limit electronically.

Question: You remove the left-hand mixer chamber servo, undo the 9mm nut and move the arm from 4 o'clock (cold) to 8 o'clock (hot). Does the CCU know that the arm has moved?



*Mars Bar: Delicious chocolate bar in UK (and Europe I think) with caramal, soft whipped nougat and thick chocolate.