Cruise Control (Tempostat) schematic

[icb]

Hi
Has anyone reverse-engineered the late 944 tempostat controller and developed a schematic that they'd be willing to share?
and...
Is anyone else having or has had the issue where you set the speed and the car instantly picks up 5-10 MPH? This behavior only started after I replaced the last two original 2.2 µF/60V electrolytic caps (one of them had been leaking). I had replaced the other electrolytics previously and the speed control was perfect after that repair. Then it stopped working again, that's when I did these two, and that's when the speed increase started happening.

[944er]

I'd love a schematic of it as well, if someone has it. This is a typical problem with the units, and one of the cases where Clark's gets it wrong. In Clark's, there is a table which would indicate this is a problem with the servo unit. But, since the car stabilizes to the new higher speed and can maintain that speed perfectly, the servo is actually working correctly. What has happened is the computer unit has stored the wrong speed. This, like many other problems, is likely caused by cracked solder joints on the circuit board. When you changed the capacitors, you likely flexed the board slightly, inducing a new cracked solder joint.... somewhere. I had mine apart four times, resoldering every single connection the last time, before getting it back to working condition. Clark's has information on identifying only the bad ones, but they are subtle sometimes, and easily missed. It takes about an hour and a bit to touch every connection and add add a bit of solder if needed. Tedious, but doable.

[icb]

If I create a schematic I'll certainly share it here. It won't happen for a while if at all; it's quite a lot of work and I have a million projects going already.

I don't suspect the solder joints in this case. I checked the board closely (PCB microscope) during the last two repairs and the joints are perfect. I'm not 100% ruling it out, although (insert controversial statement here:) I think the whole bad solder joint thing has grown into a bit of a myth and scapegoat in the 944 universe. I'm not saying defective joints are not a possibility here and there, but in 40 odd years of electronics work, I've run into maybe 6 legitimate bad solder joints (and broken or cracked traces, not including violent physical damage) on a PCB, and they were mainly responsible for intermittent faults.
Hitting all the components with 750 degrees of heat will drive out moisture and may help some components whose values have drifted over the past 40 years get back closer to spec on at least a short term basis, and I think in many of these cases that is the reason for the "fix" when somebody reheats all the joints on the board and the unit begins to work again.

[944er]

Well, when all else fails, go ahead and take an hour and a bit to touch up all the solder joints. Whatever wave solder process the Porsche supplier used back in the day left little hollow tents of solder that easily fractured, so be sure to fill in with an actually good solder joint where where it didn't previously exist.

[JamesM]

I can confirm the claim of the cracked joints when it comes to dmes, and reflowing them by hand has been successful for me I can post my process if anyone here needs it, even on one that was "repaired" recently by one of the ecu repair companys. The cruise control on the other hand although I haven't tried others have said they have some even replacing all the electrolytics and it still not working. I would before you throw away the cracked soler joint theory, strip off the outer pcb coating off your units and look again, there could be a chance that outer film is covering up the cracks?

[icb]

I haven't said I'm throwing it away. Solder joint flaws and cracks are a reality. But not in every case, and just re-flowing every single joint on a pcb is no way for an electronics tech to troubleshoot an issue, if you ask me. it's a type of parts cannon. More importantly it clashes with my OCD.
In the grand scheme of things, there's really nothing harmful about carefully touching up the solder joints. Just don't cook the pcb or the components, and don't break or lift any traces.
Solder away as much as you like, I don't mind!

Reverse engineering the schematic diagram will be quite time consuming, but also rewarding and fun. Once done il'll be able to figure out exactly what the issue is, whether it's based on a component or a connection, and it will take time, but when that time comes, it'll be a high point in my OCD filled day!

[JamesM]

It would be a great project, to do, I have a friend who is reverse engineering his idle control computer in his e30, he has it to a point where it will run off of it but still needs to tweak and fix his pid loop he is running it off of. It hopefully shouldn't be too hard for you. Reflowing the electrical components, I can agree with you is only a temporary fix at best.

Regarding your comment about it being a parts cannon, me reflowing them eliminates the board and joints as a possible fault and allows me to know if it still doesn't work that it is a component that failed. However, every time I have reflowed it has solved a nonfunctioning board and I have yet to need to delve deeper into the component level with the boards I have done. Having repaired an already professionally repaired board I believe that they reflow first because the one I fixed had no components replaced, even the original electrolytic was still there.


As stated, before everything I have heard leads me to believe that the cruise control unit fails from faulty components rather than bad joints.

[Tom]

I haven't said I'm throwing it away. Solder joint flaws and cracks are a reality. But not in every case, and just re-flowing every single joint on a pcb is no way for an electronics tech to troubleshoot an issue, if you ask me. it's a type of parts cannon. More importantly it clashes with my OCD.
In the grand scheme of things, there's really nothing harmful about carefully touching up the solder joints. Just don't cook the pcb or the components, and don't break or lift any traces.
Solder away as much as you like, I don't mind!

Reverse engineering the schematic diagram will be quite time consuming, but also rewarding and fun. Once done il'll be able to figure out exactly what the issue is, whether it's based on a component or a connection, and it will take time, but when that time comes, it'll be a high point in my OCD filled day!

I would LOVE to see a reverse-engineered schematic for this! Carpokes can even sponsor the effort somehow if interested. (Same goes for the climate control unit.)

[JamesM]

I am going this week to a small injection molding company to see if they could get molds done for parts to fix climate control switches, I may even see what the possibility is of redesigning the switches to use a latching mx keyboard switch. I believe there are already posted schematics on rennlist or somewhere. I know they simplified the pcb design later on in production to use hot swap switches, but I haven't looked close enough to know if the circuit is the same, just using more compact components or if they simplified it. Also, the latching mechanism on new oem switches is slightly more robust that the original design, I can take pictures when I get home if interested.

[Tom]

I am going this week to a small injection molding company to see if they could get molds done for parts to fix climate control switches, I may even see what the possibility is of redesigning the switches to use a latching mx keyboard switch. I believe there are already posted schematics on rennlist or somewhere. I know they simplified the pcb design later on in production to use hot swap switches, but I haven't looked close enough to know if the circuit is the same, just using more compact components or if they simplified it. Also, the latching mechanism on new oem switches is slightly more robust that the original design, I can take pictures when I get home if interested.

I have seen at least 3 different climate control pcb's -- the latest of which is significantly different. They did change it to make the snowflake switch easier to change (resulting in it becoming a separate part number), though it went well beyond that. For what it's worth, I have changed them on early units. (In fact, if I recall, that was the very first thing I did on my car 25 years ago, after deleting the toggle switch in the glovebox that was hard-wired to the a/c clutch!).