Thanks for adding on guys - amazing how many of these coils go bad. Must be bad specs on the thing, I've had other cars for like 13 years and 200,000 miles, another for 180,000 and still going never put one on. I did have a '88 Dodge that burned one up (melted) and so I put another OEM one on, and it did t he same thing months later. Finally bought one made in Japan (!) and it worked until I sold the car with 160,000. Perhaps these things are getting hot? Hard to say, the exhaust is on the other side.

 

The small and crapped engine compartment in our cars is the equivalent of a rotisserie oven. I think components do suffer as a result of the high operating temperatures. And if you live in a hot climate, even worse.

 

This is listed as a valid part available from RockAuto. (!) Search for 0165006*

$113.79 at this time.

 

This is great as it appears to be the entire clutch and not just the coil. Too bad it wasn't available 3 years ago when I replaced the compressor.

///Rich

 

The question is, how long will it actually take to get the rock auto unit? They dont actually stock it, it ships from the manufacturer. We have no idea if the manufacturer actually has the item to sell.

 

It says "delayed three days," so I'd guess three days + transit time. I don't see the issue.

 

That's standard industry lingo for "we have no idea of the stock levels, or how long it will take the manufacture to ship it to us. This part is so obscure we are not the least bit interested in stocking it". There is no one at rockauto who has any clue how long it will actually take to get the part, so they put in the standard line. However, I will actually order the part, tonight. Then we will see what the story really is.

 

Just ran across this thread - great info.!

My AC died suddenly today after moderately heavy acceleration to about 5.5k RPM in 1st gear. Only blows hot air now. I'm fairly certain it's not a refrigerant charge issue b/c it was cold immediately before that acceleration and got warm immediately after it.

I'm thinking it's the A/C clutch as mentioned in this thread. Any thoughts? Does it make sense that a higher RPM speed would cause it to die suddenly if it was already on it's way out?

Will try to sort through the relay test mentioned in an earlier post and confirm with my multimeter.

FYI - the clutch from Rockauto ships from Great Britain, so shipping is a little steep - total would be $145.53 with a 5% discount code... still probably better than the clutch alone for $100 on ebay b/c you get a new pulley too...

 

As a reference for folks, here's a picture of the wiring on the A/C relay.

I tested for resistance between the terminals for 30 and 87, which I believe are for the clutch coil and didn't get a reading at any ohm level. Tested with engine on and off. So, this confirms that it is the clutch coil, right?

One thing that seemed strange though was that with the car running, I got about 14 Volts between the terminals for 85 and 86 both when the AC button was on and off. Shouldn't this only have voltage when the AC button is on? The light on the AC button does go on and off as expected.

 

Just remeasured the voltage with the AC button on and off. There is actually a small voltage change. With AC light off, voltage was about 14.05V, with the light on, 14.15V.

Still seems a bit odd to me that the change would be so small, but I'm certainly not an expert on relays...

For now, signs seem to be pointing towards the A/C coil clutch.

 

gknorr, I'm not electrical guru either, but a .1V change is noise level. .1V isn't going to be the difference between firing a relay and not firing it. Is that the voltage at the AC connector?

 

I'm not entirely clear with how you are measuring these voltages. Is the relay plugged in and you are probing the wires/terminals?

You have a couple of options to make sure things are working correctly wiring wise before giving the car a part.

The relay coil is controlled via the ground from the car's computer (ground switched circuit). There is always voltage at the relay when the car is running for the control side of the relay.

The output side of the relay 87 should get energized when the computer is trying to turn on the compressor. You should be able to test the harness + coil resistance by probing between 87 with the relay removed and ground to get an Ohm reading. Between 30 and 87 you're going to be measuring the resistance of the battery along with the coil and everything else. I wouldn't measure that if you want a clear indication.

Here's some BMW explanation of the system along with the wiring diagram.

WDS Page

Check out these videos/websites related to checking the solenoid / relay function:

 

Thanks anyheck - very helpful.

I've been testing with the relay unplugged, using the terminals in the fusebox.

Just tested the resistance between 87 and ground (used the ground connector on the battery) and got a massive resistance - about 55k ohms.

From dirkdaddy's youtube video, it looks like it should be 5 or 6 ohms. If this is the case, it is the clutch coil.

With the test between the terminals for pins 85 and 86, I was attempting to verify that the ECU would be switching the relay open and closed, but after a little more thinking about it, I think the relay would have to be plugged in to see a voltage difference there with A/C button on versus off.

Without the relay plugged in, 85 would always show 14 volts b/w the terminal and ground an 86 would always show 0 volts b/w the terminal and ground.

 

Based on what you've said here my vote goes to the clutch coil.

What you were seeing when measuring between 85, 86 with no relay and seeing the voltage change by .1V with the A/C button on and off is that the 86, computer, side is floating when not engaged, and then grounded when on, but you get (luckily) miniscule current flow through a voltmeter as it is very high resistance. The 86 side would be "pulled up" somewhat when not grounded, thus the lower voltage differential with the A/C button off.

Best regards.

 

I think the small voltage change was actually due to the DME changing the idle for A/C on. The crank speed increased slightly when pressing the A/C button too. Whatever the case, I think it's safe to say the switch and relay appear to be working fine.

Just ordered the clutch coil from Rockauto. We'll see whether it's actually available soon. Maybe they'll decide to stock it once they figure out they're the only one selling it in the states.

For some reason the price dropped by a dollar to $112.89. Maybe because mutt2jeff ordered it Sunday....

 

If you check 86, with the relay in you should see it at ~14V with the A/C off and 0V with the A/C on measured with respect to ground.

Please update us when your coil ships/arrives.

 

Just got a notice that the coil has been shipped. Should be getting it by the end of the week.

 

Pulley/clutch coil have arrived. One day in transit from Great Britain. Nice that it was fast, but wish there had been a cheaper/slower shipping option - shipping was $37!

Hope to do the repair tomorrow - will take lots of pictures and plan to type up a detailed how-to.

Does anyone have details on how exactly to use the load strap to pull the engine rearward to make more room for the front bolts on the A/C compressor? Once releasing the lower engine mount bolt, any recommendations on where to attach the load strap?

 

I put the bolt back in the rear engine mount and wrapped the strap around the crossmember further back.
Taking the bolt out that was in the center I found a real pain.
I welded a bolt to it and then used a socket. I think it had lock thread on it.
The other picture is of the 2nd circlip which you need to remove.
The first was easy with circlip pliers.
The 2nd one was more difficult.
I think the clutch solenoid and the ac pump was slightly corroded which made everything tight.

 

I tested my old and new solenoid.
Also a picture of the first circlip you need to remove.
Works nice now.

 

Thanks bd8134! Very helpful pictures. I'll be giving this a go tomorrow. Hope I don't have to weld a bolt on the pulley bolt like you did - I don't have a welder! Might just have to get an electric impact wrench from Harbor Freight if it gives me trouble...

 

I just did this and put the vertical mount bolt back and ran the strap behind the front mount bracket. Not much room to spare, but it pulls it back as far as it needs to go.

 

Put the new pulley and clutch coil in today and A/C is nice and cold again! Took lots of pictures and hope to do a write up tomorrow or Monday...

 

Great to hear that it was the solution. I look forward to the writeup.

Regards.

 

Didn't have time for the write-up today, but hopefully tomorrow...

A few quick suggestions, just in case anyone is going to tackle this before I get to the write-up.

1. Check to see if your crank position sensor is weeping/leaking first. The o-ring is $2, and I think you'll have everything apart to get to it if it is leaking. There was a bunch of road grime/oil build on the front of my engine block, which I'm pretty sure was due to the seal on the crank position sensor. Wish I had known this before having everything apart. There's a good how-to over on motoring alliance (NAM won't let me link directly to it...)

2. An impact wrench on that pulley bolt is absolutely necessary. I tried getting it out without one first, and ended running out and buying an electric one from Harbor Freight - glad I had a 20% off coupon handy!