SharkSkin

Just want to double-check my assessment of the situation here. I went to an AC shop with an AC system that was in basically "unknown" condition. When I was checking out the car I thought it blew pretty cold. After getting it back from the shop after some major engine work, it no longer blew cold. The shop assured me that they never had to break open the system.

During various experimentation sessions, I had seen the AC clutch engage and disengage properly. Today, at the AC shop, the clutch would not engage though the low pressure switch was found to be operating properly. The tech bypassed the clutch relay for testing, and was able to confirm proper operation. He was unable to find any leaks in the pressurized system with his sniffer so he charged it up and added some flourescent dye. I'll rent the light & glasses later, to see if anything is leaking. He did mention that there was some oil on the safety valve at the rear of the compressor.

I went back home and established that the relay (IX, IIRC) for the clutch was not working. The coil contacts for the relay were getting power, and shorting the switched contacts engaged the clutch as expected. I did some running around to try to find a relay, and could not find the exact part right away so I put a generic relay in temporarily so I wouldn't cook. The AC was working fine, blowing 35-40°F which I can tell you made me VERY happy. After another 10 miles or so of driving, I had the correct relay in hand and I installed it. Again, the AC worked so I was very happy.

However, I began to notice a slight vibration that hadn't been there before, so naturally I fiddled with the AC control to see if there was any change with the AC off. It was still there, so I figured it was road surface noise. Well, I got into traffic and slowed to a crawl at which point I noticed that the vibration was RPM dependent. Nothing major, just a faint, low growl. I noticed that the AC was still blowing cold, even though I had the control all the way right on the last setting before defrost and it had been there for several minutes. At this point I stopped and did a little sleuthing with the hood open. The AC clutch remains engaged now regardless of the HVAC control position. I went ahead and drove it that way for the remaining 20 or so miles home. When I got home, I pulled the new relay, which disengaged the clutch. I checked the relay contacts, and the relay is not frozen "on".

Searching here and on Nichol's site gives me nothing specific to the early setup, but it seems that on a later car with an AC pushbutton these symptoms would call for replacement of that switch or the relay inside the head unit. Since the main AC control seems to have the AC switch built in, I'm concerned that it may need to be replaced... NOT cheap.

So, the question I have for the list is... is there anywhere else I should look, or is it most likely the control head? I probably won't be able to tear into it until after the weekend, but I'd like some pointers as to what else(if anything) I should look at.

fst951

Rebuild the clutch relay youself for about $15 if that is the problem.

Once question I have is what is the safety valve on the back of the compressor? What year is your car and did he show you what he was talking about?

Good luck

SharkSkin

The valve looked oily, but I'm not entirely sure that isn't just leftover residue from the cam tower leak, that I hadn't cleaned up. The valve is circled in red in this pic and is called out in the WSM as a "Safety Valve":



I didn't see the relay in the head unit on my wiring diagram, but it sounds like you're confirming that MY 1978 had it. I'm pretty handy and feel confident I could rebuild whatever's there. What I'm wondering is whether there is some other possible cause of the symptoms I described. The relay is described as a "speed relay" and ties into the ignition unit in a way that I don't understand. (I-17). So, should what is normally thought of as the "coil" on this relay actually be seeing a stepped voltage or AC? It looks like it is tied into the same circuit that feeds the fuel pump relay so it can decide whether the car is running or not.

Anyway, I'm just trying to make sure I've ruled everything else out before I go spelunking in the console. I noticed later, after I posted, that the relay buzzes at low RPMs, which could be from less-than optimal voltage or maybe that's it's speed-sensing act.

Any thoughts on this welcome.

SharkSkin

Bump

SharkSkin

Bump again... surely someone has BTDT?

SharkSkin

Board is slow tonight... I guess I'm not the only one trying to pull weekend plans together.

Happy 4th everyone!

SharkSkin

Well, having returned from my camping trip, I found some time to dig a little deeper into this mystery.

I have verified that the control head closes a contact between terminals J1 and J2 only when the control is in the "AC" or "Defrost" positions. This rules out the control head IMHO.

I have tried 3 relays. With the original, the compressor clutch never engages. With either of 2 replacements I've tried, the compressor clutch is always engaged with the engine running.

I attached a frequency counter across the relay pins 15 & 31b(as labeled in the wiring diagram) and I am getting a frequency about equal to that of the coil wire.

I'm going nuckin' futz with this.... anyone have any ideas here, with regards to the parameters I should expect? Voltage across the relay pins? Can anyone tell me what the adjusting pot inside the relay is supposed to do? Any ideas what else I should be looking at?

Thx

SharkSkin

BUMP... c'mon... it isn't that easy to stump all of you, is it?

WallyP

The A/C Speed Relay, used only in '78 & '79, is shown in the wiring diagrams on page 97-15, but is not listed in the relay list for the Central Power Panel. Relay IX is shown as the seat belt relay in the relay list.

The A/C Speed relay appears to be intended to cut the A/C compressor off at certain speeds. The original 928 Tech Info booklet doesn't mention the function of the Speed Relay, so I don't know whether is is intended to cut the compressor off at high RPM, or to keep it off until the engine reaches normal idle speed. My GUESS is that the speed relay functions much like the fuel pump relay, and prevents the A/C compressor from loading the engine during cranking. (The fuel pump relay and the A/C speed relay appear to have very similar wiring diagrams and have the same inputs/outputs.)

If that is the case, the relay contacts should close at all normal running speeds, and open when the engine is not running. Other than that, the A/C Speed Relay does not appear to control the A/C compressor clutch at all.

This means that somewhere, you are getting a 12 vdc feed that is bypassing the A/C console switches. Go to the A/C antifreeze switch (under the black plastic cover at the base of the windshield on later cars) and see if there is 12 vdc on the terminals there with the relay pulled. If so, the feed is between the relay socket and the clutch, and will probably be an added wire. If there is no power on the terminals with the relay pulled and the ignition switch on, the bypass wires is somewhere between the A/C switch in the control unit and the relay socket.

The problem is almost certainly not your relay - the two replacement relays appear to be working properly.

Sounds as if the PO installed a jumper to have A/C even with the faulty A/C Speed Relay - you need to find that jumper.

SharkSkin

Thanks Wally, I had the same guess about the function of the speed relay but was unsure how to test that it's getting the right signal. I'm still unsure about that but I'm glad you came up with the same guess.

I was of a mind to go in and check all of the connection points as you mention. Up to this point my tests have consisted of testing various points on the fuse/relay panel and looking to see if the compressor was spinning. I verified that the AC switch opened and closed the circuit between J1 & J2 but what I didn't verify was whether there is another 12V feed coming in somewhere. I'll do some quick checks tonight after work and dive in with both feet this weekend if needed.

Thanks for the input!

SharkSkin

OK, progress of sorts. I've established that no harness-hacking has occurred under the hood, though it seems that a connector from the harness to the air valve and clutch has been replaced. Also, luckily I didn't buy both replacement relays... I bought one from the dismantler and was able to try the second one gratis when I found that the relay was always activated.

I have verified operation of the freeze valve and wiring; the FV switch closes when I blast the AC for a while(a very good sign IMHO that it can get that cold without the front fan engaging). Wiring checks out, no mysterious 12V there.

I have checked and the "AC-on" signal from the control head is arriving at J2 as the expected 12V above ground on that pin, when the selector is in either "AC" or "Defrost" position.

Pin 30 on the relay OTOH is always at 12V with the key on. So, I have finally narrowed the problem to this area, internal to the fuse panel(circled in red below):



So, finding this it hit me... the butt connector that I found on the back of the fuse panel must be the problem. When I reconditioned the panel I didn't know what this was for, I simply improved the connection from a butt splice to a soldered splice. I went back to look at the pics I took at that time, and it's clear as day on the 2272X1704 version of the following pic from my Bzzzt page... Unless there is some illusion going on in the pic due to the mass of black wires, Pin 30 on the relay is hard wired to the yellow connector in the foreground, which goes straight to the main busbar from the battery!



So, I feel pretty confident that the problem is identified, if not solved yet. The way I read the diagrams(and my meter), what I need to do is pull the panel and re-establish the connection between pin J2 on the panel and pin 30 on the speed relay, removing the connection to the main busbar. This seems to be one of the more subtle PO wiring hacks I've heard of, but I'm sure that between Wally and Jim, you guys must have seen it all... a few times over.

I'll update here and on my site after I've had a chance to dig in this weekend. Thanks Wally, for steering me in the right direction and confirming the areas where I was on track.

WallyP

Ya know, it's almost understandable...

Terminal 30 is SUPPOSED to be a direct battery connection per the DIN standard. In this case, though, "30" is used to show "power in", even though the power is switched rather than constant.

Just a case of wiring something per the standard, rather than as Porsche designed it, apparently.

Bill51sdr

Just a thought here, but I do believe that there are some systems in existence that do not cycle the compressor on and off. The evaporator pressure or temperature is regulated by the expansion valve (the name of this type of valve escapes me for the moment), not by cycling the compressor. It seems to me that this is the type of system Porsche may use in these cars, so don't be surprised if your compressor does not cycle after you have made your repair. Could be you have a faulty expansion valve or pressure switch somewhere in the system causing the compressor to work against higher than normal pressures, causing the vibration you have felt.

SharkSkin

Wally,

I've decided to add a relay to the system, activated by the signal from the control head on J2. The relay's main contact will supply power straight off the busbar to pin 30 on the speed relay. It should have been designed this way in the first place.

Hmmm... come to think of it, the speed relay is a discontinued part that's getting hard to come by. Maybe I should route the low current signal from J2(control head) through the speed relay's contacts, and allow the speed relay to control whether that signal is allowed to activate a more common power relay.

I can use a rugged, cheap 40A relay to take the brunt of the wear and tear of switching the highly inductive clutch and air valve. I've heard of doing this sort of thing to prevent the control head relay from frapping out, but taking it a step further to prolong the life of the speed relay might just be a new twist. Hmmm...

Bill,

I don't think this system is that fancy. It sounds like you are thinking of a thermostatically controlled expansion valve of some sort. As I understand Wally's AC writeup, the system just pumps freon until the freeze switch(or low pressure switch) disengages the compressor. When the freon temp switch decides that the freon is hot enough, the condenser fan turns on. Real basic all-or nothing design where the compressor and fan are intended to cycle as needed. All of these controls check out fine.

Just about every AC compressor on any car adds some vibration when switched on. I'm not sure it's a problem, it's just something I'm not used to.

SharkSkin

Problem resolved... at least the electrical part of my AC fun. Here is the modified circuit:



More details on my site in the new "AC Chillin" section.