A/C compressor (clutch?) failure

In reply to TurnerX19 :

Well that makes sense. It doesn't make me happy, but it makes sense. If the small amount of R134 I added yesterday is gone when I check later this morning I'll look at the compressor shaft seal first. Of course, finding a replacement seal may be problematic at best....

In reply to 02Pilot :

I have yet to try and source compressor repair parts, but absolutely will try to do so the next time I need compressor service on my Volvo. The rebuilds and new chineseum now available tells me we can do better quality ourselves.

Once again I have managed to grasp defeat from the jaws of victory. After a number of fits and starts, I got the system - which had held onto the small amount of refrigerant I put in yesterday - to start taking more. Initial low side with system on @1500rpm was nil, but after a good bit of refrigerant I got it up into the 20s. Air started blowing cool. I began to think that I might have managed to make it work (I'm an ends-justify-the-means kind of a guy).

And then I saw that the compressor clutch was only sort of engaging, and the friction was producing the tiniest bit of smoke. Quickly shut everything down - I don't think it did any damage. However, it leads me to the conclusion that, while testing with the system empty showed the clutch working properly, adding the resistance of the compressor actually having something to compress reveals that the coil is not pulling hard enough to keep it locked up. So now I need to see about finding a coil.

If ohms was to be 5 and you only got 3.6 you are out of spec. I suspect that the electromagnet has failed/shorted and you are not getting enough clamping force.

Toyman01 (Moderately Supportive Dude) said:

You probably need to ohm the coil and compare it to spec. My guess is the coil fried and wasn't pulling the disk in hard enough to spin the compressor. 

The Googles says it should be 5 ohms. 

Ohms mean little for things like this, need to check current.

Also do a voltage drop on the wiring.  Check the voltage between the compressor clutch and battery positive.  I had one car smoke a brand new compressor because there was a 4 volt drop.  Oops. 

OK, I'll check that voltage. FWIW, I never found the spec of testing coil resistance; the ASE says 2-5 ohms is OK.

Also going to check the gap from the pulley to the clutch plate. There were 2 1mm and 2 0.2mm spacers in there - I'm wondering if there's enough wear to require removing one of the small ones.

Did you adjust the shims and clutch clearance when the new one was installed? There are typically shims on the compressor shaft.  I know the CR-V likes 0.35mm, and has three different thickness shims to get you there.

* Edit, looks like you posted about it while I was reading

Well berkeley me sideways. I checked the spacing, which was on the wide side of spec (either .3-.5mm or .4-.6mm, depending on who you believe), so I figured I'd pull the thing apart again and take out one of the small spacers. Once I pulled the plate off I saw that the felt ring had come apart catastrophically - I guess it didn't like being handled, as it was fine initially. Anyway, it made a mess, so off came the pulley again to clean everything up. Got that done and put everything back together...well, started to anyway, because as I went to put the snap ring back on to retain the pulley I saw that my pressing rig - which was used exactly the same way yesterday with no issues - looks to have buggered the seal on the new bearing. So now I'm back where I started.

And to add insult to injury, I seem to have lost the very shim I was planning to remove, so now it better damn well help or I'm proper berkeleyed. I'm so pissed that I'm tempted to slam it back together in the hopes that it burns itself and the rest of the car to a crisp.

I'm going to go rock gently back and forth in the corner now.

The saga continues. Got my replacement bearing, did the whole R&R again, and got it bolted up without issue (or so I thought). Started the engine, turned on the A/C, clutch engaged, air blowing cold. Ran it for a few minutes, saw the belt was still a bit loose, so I shut it down to tighten that up. That's when I heard a very slight hissing. From what I can see one of the small diameter tubes on the expansion valve (I think - I'm peering into the evaporator housing through a small hole) is cracked at the fitting. Or maybe the O-ring is leaking - I can't be sure. How did this happen? Best I can figure the repeated moving of the compressor and attendant stress from the low pressure line being maneuvered around took its toll, so I have once again broken something and made yet more work for myself. I'm wearing a huge happy hat right now.

The good news is that the expansion valve seems to be available and cheap. So is the receiver-drier, so I'll do that too. What I can't find is a definitive list of what O-ring sizes I need, at least for those bits. Is one of those cheap kits of R134a O-rings likely to have what I need? In other words, do these tend to be standard sizes, or is every manufacturer using their own specific stuff? It's a Seiko-Seiki compressor, if it matters.

It is not uncommon for the expansion valve to hiss for a few minutes after shutdown. It will do that until the high side and low side equalize. Unless you actually see gas and oil escaping, I'd run the system for a couple of days and make sure it is actually leaking.

In reply to Toyman01 (Moderately Supportive Dude) :

I see it leaking. It's definitely busted, I'm just not sure exactly in what way. I need to figure out the O-ring situation, then get everything ordered. Fortunately, the expansion valve and evaporator are very easy to access on the C900.

In reply to 02Pilot :

There is a great deal of commonality in the O-rings used in A/C systems. I buy them in boxes of 10 from the FLAPS. Also, frequently new expansion valves come with O-rings, and occasionally I have seen receiver dryers with them too. Make sure you get the green ones, not blue or black, and give them a little PAG oil on install. Probably should add 1 ounce of oil to the system while it is apart to replace what is in the old receiver.

In reply to TurnerX19 :

Yep, definitely going for green. Didn't think about adding oil, but that's easy enough. I'm going to keep digging to see if I can pin down the sizes, but if I fail to do so (I've been failing a lot lately), I'll order a selection and hope for the best.

Time for an update. Parts finally arrived and I finally had time to tackle replacing the expansion valve and receiver/dryer. The expansion valve went pretty well. I was nervous about bending the copper piping to mimic the old one, but I got it in and seemingly didn't wreck anything. That should have been a warning.

The receiver/dryer is buried between the inner and outer fender in a very inaccessible spot. One fitting is partially exposed, but the other one is right in the line of the inner fender, so there's no good access to it. Guess which one was rusted in solid....

There was no way I could get it to move with hand tools, even with several soakings in CRC Freeze-off. I got out the appropriately sized impact crowsfoot and started trying to rattle it off. Eventually this worked. Unfortunately, in the process the aluminum line was ruptured (I was afraid that might happen, but I didn't have a choice but to use a lot of force - it wasn't coming off any other way), so of course in trying to fix something I've created yet another problem for myself. The line doesn't appear to be available anywhere that I can find, so now I have to either find someone who has a good used one or find a shop to make one up for me. I'd prefer to get one made, but those sorts of shops don't really exist around here any more. I have to do some searching.

I really do not need any more reminders from the universe that no good deed goes unpunished.

Anyone got a C900 parts car?

Related question: the replacement receiver does not have a low pressure switch out of the box; it does have two hex-head plugs in that location (one on each side). Am I correct in assuming that I should pop one of those out and install the low pressure switch, or is it non-critical to have one in the system?

In reply to 02Pilot :

You are correct as to the installation technique for the low pressure switch. It is a good thing to have, but not an absolute requirement. It helps preserve the compressor when the pressure gets too low to push lubricant into the upper cylinders in the compressor. FYI there are as many as 6 cylinders in there.

In reply to TurnerX19 :

Thanks for the confirmation.

I did actually find a local shop that can make up a line, but the money they want for it is exorbitant. I also found some services online that can do the same thing, though without some sort of recommendation I have no idea how good or bad any of these might be. I've got a line on a used replacement hose - not ideal, but the cheapest, easiest solution (I hope), so with any luck I'll be in business soon.

OK, finally got all the bits collected and the time to put them together. Got the line in, hooked up the manifold and vacuum pump, and crossed my fingers. Pulled to 30inHg very quickly and held there for 30 minutes. Turned off the pump and left it for an hour - still at ~30inHg when I came back. If anything's leaking, it's fairly small. Charged with 3 340g cans, which was almost exactly the called for 1kg charge (bearing in mind this was originally an R12 system, so I'm not sure just how precise that figure is with R134a). Took a while, but with the full charge in the pressures looked to be within spec on both sides, and the air was blowing cool. Shut down, disconnected, and listened for leaks - nothing that I could make out.

So I took it for a test drive. Not so good. Coolish, but far from properly cold. Came back home and discovered that I still have the same old problem - the compressor clutch is not locking down hard enough, and when the engine revs to any degree, it slips. At idle it's OK, but much above that it and it's a crap shoot as to just how much motion is transferred.

So now the question is how to proceed. I can't find a source for a new replacement clutch or coil - I'm sure that this compressor was used on multiple makes (it's a Seiko-Seiki SS170PSS), but I'm not having any luck finding an interchange or even a listing of these parts by size so I might be able to find something that would fit. I either have to find a good used one and hope it lasts, or...?

Follow-up question. I read something on a Volvo forum about shimming the springs on the clutch face to increase clamping force. This got me thinking that perhaps these springs were part of the equation. I cleaned underneath them and ensured they allowed even movement and pressure. This helped some, but the problem remains. I may try shimming under them a bit to see if it makes a difference, but that raises an additional issue - clearance between the clutch disc and the pulley when the clutch is not engaged.

The question that I've taken all this time to get to is regarding measurement of that gap. The spec is .3-.5mm (IIRC - that's the ballpark); I'm at .3mm cold, and of course it shrinks as the metal heats up. But when I slip the shim in to measure, of course the clutch deflects outward, as the springs allow a good bit of flex. I can easily fit a considerably larger feeler gauge in there if I accept the springs deflecting more. Should I be measuring the static gap a closely as possible (i.e., minimal spring deflection), or the gap with the springs at maximum deflection (i.e., the biggest feeler gauge I can cram in there)? Obviously, the latter method means that I'd have to pull a shim if it's out of spec by that measurement, and also that there will very likely be some contact between the clutch and the pulley even when the electromagnet is not engaged. Not sure if that will be a problem.

I realize this could all be easily solved by a new clutch, but since I can't find one, I'm grasping at straws. It's damn hot and want my A/C to work already.

Remove shims until it touches, then put the last one back in. 

Do you have a full 12 volts when its operating?  No voltage drop anywhere?

The smallest shim is 0.2mm, which is going to bring it very close.

IIRC, when I measured the voltage way back when it was within 0.5v of operating voltage, but I'll check it again. On the same Volvo forum I saw some discussion of adding a relay, but I have to look at the wiring diagrams to see exactly how that is set up on the C900 - it may be a moot point.

I'm also wondering about just how flat the clutch face is. I don't have the ability to do much with it regardless, but it's showing its seen 30 years of use.

The last thought in the post above got me thinking. I checked the surface of the clutch disc, which was pretty flat, if uneven. I noticed, however, that in spite of being reasonably flat, it was shiny around the outer edge, but the inner contact surfaces had some spots of rust and were less shiny. So I looked at the pulley, and found to my horror that it was noticeably more worn in the center than at the outer edge, meaning essentially that when the clutch engaged only the outer edges were making contact, and this was not enough to hold things reliably together.

With little to lose, and no proper tools to resurface these two, I improvised. First I put a bolt through the center of clutch disc, tightened it on with a nut, and then spun the whole thing with my cordless ratchet on the bolt head while running coarse emery cloth on the mating surface. After a few minutes this looked considerably better than it had. The bigger problem was the pulley side, but then I realized I had the perfect apparatus for spinning it - the engine. Started up the car with the clutch disc off and used the same emery cloth to resurface the pulley and try to take some material off the outer edge. After quite a few passes I got it a lot smoother and much closer to flat that it had been.

Running in the driveway and a brief test drive suggests improvement. Engagement seems better, and it doesn't seem to be slipping. I need to test it more, especially when it gets hot, to see how it does longer term, but right now I at least feel like I didn't make it worse.

As seems usual, testing reveals that it's still slipping when it gets hot. The gap hot is tiny - less than 0.2mm, so I can't remove any shims. I'm running out of options, but I'm going to look at the both sides again and see if they need further leveling. I also discovered that the clutch has little nubs between the static and spring portions that limit the travel of the sprung outer ring - I'm thinking about grinding those down, just in case they're limiting clamping force. I'm also going to triple check the voltage to the coil to make sure I'm getting as much pull as possible - might run a wire direct from the battery just to see if there's any difference.

The good news is, that for the few minutes it was running properly, the air was icy cold.