[mtflight]

HISTORY:
A few weeks ago (over a month) I posted regarding my A/C not kicking on when it was supposed to, then staying off for long periods of time (in 100 degree weather). It all happened after I had an oil change at Wallyworld.

They do the 15 pt check or whatever they call it, and test your battery. Within 20 minutes of them doing all this, the problem started. I had a few stored codes, which I cleared (EPROM read access error, etc).

I found that if I stepped on it the AC would work (clutch would engage), and so I did a lot of unnecessary WOT$ when gas was higher.

The dealership wanted $700 to replace compressor.

I had read about some folks on here finding out the little box that was taped on the line to the AC compressor clutch power cord was the culprit. I could not find it. Then one day I felt it in those cords.... tiny little white box wrapped in black tape: the A/C clutch diode. Sure enough, I took it off the line (it had alligator type clamps cutting into each line) and the AC clutch was engaging like a champ. Determined to replace it, in case it actually had a function I went to O'Reilly autoparts, and even to the Cadillac dealership and they knew nothing of it. Finally got a Chevy parts counter to give me a part number.

PROBLEM:
I ordered it, and when I received it I realized there's no way to know which way it goes? I should've taken notes.

The two lines in the cord that plugs into the AC compressor are green and black. The two lines in the diode... are white.

There's an X on one corner of the box, and a Y on the other corner (opposing). Knowing the function of a diode is to restrict current to one direction only I guessed it was from X to Y. but each of the two lines has one of the letters. So in essence I could assign X to black and Y to green, in one direction or another, and viceversa (4 possibilities).

So, I tried X to Y (flow), assigning black to Y [green to X], and the compressor stopped engaging. My check engine light came on too. I thought it was from me tugging on wiring. PCM P1660 Fan Circuit fault, I found out. So I looked and sure enough both of my fans were not turning.

This is at 11 pm folks. I traced all wiring, looked for fuses to the fans, and only found the relays and MaxiFuses. Distraught I was.

So I redid the wires to the AC clutch without the diode, and it still didn't work. I was very frustrated late at night.

In the morning I'm trying to figure out what I'm going to do to get to work, and I thought to look in the service manual for the electric chart to the cooling fans. In the little box eplaining the maxifuse for the cooling fans, was another circuit breaker: the AC compressor fuse 10Amps. I went to look for this fuse, and sure enough it was blown. I replaced it and cleared the codes (fans were working as they should).

So I'm running the AC without a diode. I have the diode, but need to purchase at least 3 10 Amp fuses, in case I need to find out the hard way which way the diode connects. My likelyhood of success each shot is 1:3 (assuming the one that blew the fuse is out of the way).

Help is wanted and appreciated!

 

[Rock]

Since it's a diode, it ought to be a simple matter to determine its directionality (anode/cathode ends) by using either a simple continuity-checker, or a test-light. The diode will only conduct current in one direction, so put it in series with your test-light/meter. It should measure as an open in one direction (or no light) and a near-short in the other (light comes on).

Now that you know which end is the cathode and which is the anode, just look in your Electrical Wiring Diagram for the proper connectivity.

BTW: In case you didn't know... in the schematics, the cathode is the end with the line across the point of the arrow, and the anode is the other end. ----|<---- The left end is the cathode and the right is the anode. It will conduct current when Negative (ground) is on the cathode and Positive (12 volts, or anything greater than the cathode) is on the anode.

 

[mtflight]

Since it's a diode, it ought to be a simple matter to determine its directionality (anode/cathode ends) by using either a simple continuity-checker, or a test-light. The diode will only conduct current in one direction, so put it in series with your test-light/meter. It should measure as an open in one direction (or no light) and a near-short in the other (light comes on).

Now that you know which end is the cathode and which is the anode, just look in your Electrical Wiring Diagram for the proper connectivity.

BTW: In case you didn't know... in the schematics, the cathode is the end with the line across the point of the arrow, and the anode is the other end. ----|<---- The left end is the cathode and the right is the anode. It will conduct current when Negative (ground) is on the cathode and Positive (12 volts, or anything greater than the cathode) is on the anode.

Good stuff here, thanx! I need to purchase one of those test-lights. I don't have any of this gear, but I guess I may as well spring for it. I was going to purchase fuses, and try the other 3 ways it could "possibly" connect.

I like the logic and practicality of your approach. It is exactly what I was looking for--thanx!

 

[WarrenJ]

The anode of the diode goes to the clutch terminal that is grounded (black wire). The cathode goes to the clutch terminal that is connected to the A/C compressor clutch relay (dark green wire).

If you install the diode backwards you will:
1) possibly let the magic smoke out of the diode rendering it useless.
2) blow fuses/circuit breakers
3) both of the above.

It appears you've already found the wrong way to hook up the diode. Test it first to verify it's still good and then reinstall it properly.

CAUTION: The purpose of this diode is to quench high voltage back EMF energy generated by the clutch inductor when the clutch is turned off. If you run without it, you run the risk of impressing high voltage on your car's very sensitive computer network. I wouldn't do it for an instant! I would urge you to pull the plug from the compressor until the diode is properly installed.

Regards,
Warren

 

[mtflight]

The anode of the diode goes to the clutch terminal that is grounded (black wire). The cathode goes to the clutch terminal that is connected to the A/C compressor clutch relay (dark green wire).
If you install the diode backwards you will:
1) possibly let the magic smoke out of the diode rendering it useless.
2) blow fuses/circuit breakers
3) both of the above.
It appears you've already found the wrong way to hook up the diode. Test it first to verify it's still good and then reinstall it properly.

Great, it was not cheap, and it took weeks to receive it.
Thanks for the advice and heads-up.
So, when I get off work here I need to know which is the anode and which is the cathode. I don't have any electrical testing equipment. What should I purchase (budget oriented... I don't think I'll be doing this repair again).
Thanks!
Edit: There are two other wrong ways to plug this thing in..... when I get this equipment (inexpensive, hopefully) where do I connect what to determine which is the anode or cathode?

Edit... picture attached

 

[WarrenJ]

. . . . So, when I get off work here I need to know which is the anode and which is the cathode. I don't have any electrical testing equipment . . . . What should I purchase (budget oriented... I don't think I'll be doing this repair again) . . . . There are two other wrong ways to plug this thing in..... when I get this equipment (inexpensive, hopefully) where do I connect what to determine which is the anode or cathode?

If there are any cracks in it the magic smoke is history; get a new diode.

Next, go to your nearest Radio Shack and ask for the "geek." He'll test it for you and identify which terminal is which (surprised there are no markings on the diode body . . . . or ARE there?).

He'll use an ohmmeter (VOM with a "diode test scale") to measure the diode in both directions. It will read a very high resistance in one direction and low in the other. When the leads are connected such that it reads low, the red (+) lead will be connected to the anode of the diode and the black (-) lead will be connected to the cathode.

Failing that, buy their least expensive VOM that has a diode test scale and do it yourself. It should set you back less than $20. Afterwards, carefully box it back up, wrap it and give it to a nephew or a geek as a Christmas gift. One gift down . . . .

Regards,
Warren

 

[mtflight]

(surprised there are no markings on the diode body . . . . or ARE there?).

Thanks Warren. Check out the picture I "made" on my post above. I'll go to Radio Shack to ask "Da Geek"

 

[WarrenJ]

First, why not edit your User CP to indicate the year of your vehicle under "Ownership:"? It makes it easier for others to reply to your questions. :halo:

Secondly, I'm surprised no instructions came with the new diode. Well, not THAT surprised I guess.

A diode has two terminals, not four. I'd guess that this unit was potted with four wires so as to make inline installation easier (emphasis on GUESS). If that's the case, the top two wires will read zero ohms between them since they are a "straight-thru" connection. Likewise the bottom two wires will read zero ohms between them as they are also connected to each other.

Again, if that's the case, the top two wires will both be the diode's cathode or they will both be the diode's anode. Sammy same ting for the bottom two wires.

The two diode cathode wires will connect to the two green wires in any direction and the two anode wires will connect to the two black wires in either direction.

You'll have to test this hypothesis by measuring for continuity between the two top wires. Then test for continuity between the bottom two wires.

Regards,
Warren

 

[Rock]

Well... I've not seen a diode that looks like that, but here's the schematics for a 97 ETC (attached). The Diode is in the lower left corner of the diagram. The anode (once you figure out which wire/wires that is) goes to a black ground wire and the cathode goes to the dark-green wire... as Warren indicated.

 

[WarrenJ]

Well... I've not seen a diode that looks like that, but here's the schematics for a 97 ETC (attached). The Diode is in the lower left corner of the diagram. The anode (once you figure out which wire/wires that is) goes to a black ground wire and the cathode goes to the dark-green wire... as Warren indicated.

I edited my post with an attached drawing. Hopefully, it will clarify my awkward words.

I'm looking at the same drawing, Rock. Presumably, it is electrically accurate without being wire-for-wire specific.

Four wire encapsulated diodes, while not common, are around. My guess is that it was potted that way for ease of inline installation. Still, its terminals should be identified better. :yup: How much extra effort could it have taken?

Lacking that extra effort, its terminal identification should have been prominently displayed in the manual. An explanatory enclosure in the parts container should be the least expected. :mad2:

Regards,
Warren

 

[mtflight]

Guys, extremely helpful. It's like a lesson here. Unfortunately, the diode was shorted in every possible way (when I tested at Radio Shack, as per Warren's suggestion).

So, the puff of magic smoke came out, along with the 10A fuse for the AC compressor.

Good news... I pulled the little box apart, and saw the little diode. $0.79 for two of them.

Attached is a diagram of the way it is inside the box. To see if I understood properly. I will probably do this tomorrow. Please let me know if I interpreted the diagram properly, and the explanation.

Many thanks again! Wow.

 

[mtflight]

:yup: How much extra effort could it have taken?

Lacking that extra effort, its terminal identification should have been prominently displayed in the manual. An explanatory enclosure in the parts container should be the least expected. :mad2:

Regards,
Warren

I think GM does not label the diode, so that they can sell you another $0.25 plastic box with gel in it, two 10" wires going through it with alligator clamps and a $0.35 diode for a smooth $30 retail. Slick, aren't they? :cool2:

BTW Warren, my ETC is a 98 with 81K on the clock (had 69K in Feb '05), crimson pearl.

 

[WarrenJ]

. . . . Slick, aren't they? :cool2:

Umhh, maybe yeah. Hmmmn . . . .

Regards,
Warren

 

[WarrenJ]

BTW Warren, my ETC is a 98 with 81K on the clock (had 69K in Feb '05), crimson pearl.

That's good to hear. Please consider putting it in the "User CP Ownership =:" part of your Control Panel.

Regards,
Warren

 

[mtflight]

Hey Warren or Rock...

was the last illustration I made, showing the internals of the diode box, correct in terms of placement (compared with the also present diagram from the service manual)?

I'm not experienced with schematics of this type and I posted that just to be sure I was on the right page.

Is there any sense to the "X" and the "Y" labels on the diode container box? There are also three letters in the mold casting... "P.E.D."

Here are the specs for the diode:

Peak inverse voltage: 400V
Forward Voltage drop (Vf) at If: 1.6V
FOrward current (If): 1A
Max. surge current (8.3ms): 30A
Reverse current at PIV: 30mA

Thanks again, fellows.. you saved another diode!

 

[WarrenJ]

Was the last illustration I made, showing the internals of the diode box, correct in terms of placement (compared with the also present diagram from the service manual)?

I'm not experienced with schematics of this type and I posted that just to be sure I was on the right page.

Is there any sense to the "X" and the "Y" labels on the diode container box? There are also three letters in the mold casting... "P.E.D."

Here are the specs for the diode:

Peak inverse voltage: 400V
Forward Voltage drop (Vf) at If: 1.6V
FOrward current (If): 1A
Max. surge current (8.3ms): 30A
Reverse current at PIV: 30mA

Thanks again, fellows.. you saved another diode!

I believe your placement is correct. The "X" and "Y" labels have no signifigance in any electrical parlance I'm familiar with. Your drawing of the diode placement indicates to me that the left side of the diode (cathode) connects to the two wires on the top (X) rail of the diode package. It would be nice if these two wires were green. Likewise the two anode wires connect to the two bottom (Y) wires on the package. They should be black . . . . IF I've guessed correctly about the potting.

Interestingly, that agrees with the drawing I posted previously. Guess I guessed good.

The diode ratings are not exceptional; 1N4007 diodes should be adequate. Good thing RatShack doesn't know they go for $35.

Regards,
Warren

 

[Rock]

Yep, your picture shows it correctly connected. Looks like Warren called it right... "X", "Y", GEEZE that's ludicrous! For 35 bucks, they shoulda put two connectors on it so there wouldn't be any question... or built the damn thing right into the compressor's clutch solenoid.

 

[mtflight]

I just wanted to sincerely thank both of you. I replaced the diode in the little box, and followed the diagram we agreed upon... connected the wires, insulated them with tape, plugged it in... screwed the splash cover back on. Got in the car, turned it on, and it all works--new diode and all!

While relatively basic, I had no idea about what I was doing. You guys gave me the courage to open the little box and see what was inside. Radio Shack had the part and well... I'm pretty happy.

Thanks!

 

[mtflight]

1N4007 diodes should be adequate. Good thing RatShack doesn't know they go for $35.

No kidding... maybe they messed up the decimal point by two places! ya right.

Thanks Warren.

 

[WarrenJ]

No kidding... maybe they messed up the decimal point by two places! ya right.

Thanks Warren.

Glad it worked out. Whew! That was a lot of text back and forth!

Regards,
Warren

P.S. Thanks Rock.