Maker Pro
Maker Pro

AC Compressor blowing fuse

JerryG

Dec 17, 2013
20
Joined
Dec 17, 2013
Messages
20
I was wondering if I could get some help with an automotive ac circuit. I've tried the automotive forums but haven't been successful. I separated the ac compressor from the rest of the circuit and it is the culprit for the blown fuse (This test was done without a diode in the circuit). It is a new compressor with a new clutch but I wonder if it is faulty. Initially the fuse doesn't blow until the compressor has had some time to warm up and is pulling less amperage. Then when the clutch is cycled off the fuse goes. There is an ohm test that I have tried to do on the coil for a resistance of 2 to 5 ohms and I got low readings. Does this sound right? My meter is cheap so I'm not sure if the results are trustworthy. Could there be different readings for cold vs hot? Also I would really like to know how the diode protects the circuit from a voltage spike if it is wired parallel and not in series. Any help solving this problem would be appreciated.
Thanks
Jerry
 

Kiwi

Jan 28, 2013
471
Joined
Jan 28, 2013
Messages
471
Jerry,
The diode is wired in parallel with the compressor clutch to absorb the high voltage spike when the clutch is turned off.
Search Google for "flyback diode" for information on their use.

Your meter gave a "low reading". What resistance did your meter show?

Can you measure the resistance of the old clutch?

Have you got a know resistance that you can check the meter accuracy with?

Clutch coil resistance will increase slightly as it warms up.

Why was the original clutch and compressor changed?
 

JerryG

Dec 17, 2013
20
Joined
Dec 17, 2013
Messages
20
I will try to answer all your questions.

The original compressor was changed because of the same fuse problem. I thought that I narrowed it down to the clutch. The compressor had to be removed to change it and since there is not much difference in price anymore between a new coil and a new cheap compressor I thought I would just change the whole thing. Also I don't have the tools specifically to pull the clutch so it seemed like the better choice.

I checked my meter to a resister that should check 523 ohms. I got 524 on the 2000 setting. I get 0 probe to probe. On the 200 setting I get .3 probe to probe.

The original compressor reads 3.5 on the 200 setting cold. I had the same problem before where the fuse didn't blow for a couple minutes until things were warmed up.

The new compressor reads 3.5 ohms cold as well. Trip around the block, engine hot, fuse blown, resistance 4.0 ohms.

I read the reason for the diode in the circuit. What I don't understand is if there are two complete circuits and the ac clutch will run with the diode removed from the fuse box, how does it protect the circuit. Can't the surge travel around the diode thru the other path?
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
The clutch has a coil inside it which passes the applied current. This coil with its iron core will be highly inductive, this means that there will be a considerable amount of stored energy. Switching this inductor off will need a very high voltage to get rid of the energy, sparks will fly and the switch may be wrecked. Putting a diode across the inductor allows current to continue to flow until the resistance replaces the stored energy as heat in the windings. It makes it much easier on the switch. The diode should only pass current during switch off so should not cause a fuse to blow unless it is faulty.

Your best plan is to measure the current taken. This can be done by soldering a couple of wires to a blown fuse and measure the current going between them. Keep a very watchful eye on the current or add another fuse in series as protection, you do not want a fire. A headlight bulb can provide a current source to look for shorts in the wiring. Have the rats been chewing the insulation or has it been damaged by vibration?

Your resistance readings look wrong to me. I have no experience here but would expect 5 to 10Ω or so. Edit : 3.5Ω looks good. What is the fuse rating?
 

Kiwi

Jan 28, 2013
471
Joined
Jan 28, 2013
Messages
471
OK, it is reasonably safe to assume that the clutch is NOT your problem. Chances of fitting a new unit with same problem as original are pretty slim.

Diode is in the fuse box, not at the clutch?

Increase in resistance from 3.5Ω cold to 4.0Ω hot is probably fine. On a 12volt vehicle this would give a current flow of about 3.5 to 4 Amps.

You need to measure the current going through the fuse and also the current going through the clutch.
The fuse is probably supplying more than just the clutch, eg blower fan motor.

What size fuse?
Is it the correct size?

What is the car?

Have you checked a wiring diagram to see what else is on the fuse?
 

JerryG

Dec 17, 2013
20
Joined
Dec 17, 2013
Messages
20
The car is a 99 Lincoln Continental. The diode is in the fuse box along with the relay. The reverse lights are also on the circuit. They have been disconnected. There are some lights and maybe a cigarette lighter that is on the circuit. It has also been disconnected. The low pressure cut off switch is on the circuit. I have plugged and unplugged this while the ac was running in park and couldn't cause the fuse to blow.

I have monitored and driven the car with the multimeter plugged in place of the fuse inside the vehicle and watched the amperage while driving. The line amperage without anything on is .03 amps. When I turn on the ac the amperage starts at about 4.5 and then drops down to the low 3s over a couple of minutes. Before I disconnected the reverse lights the amperage was 8-9 while backing up. The fuse is a 15 amp and I have been using 10 since I have run out of 15s. When the fuse does blow it goes instantly. I don't see any increase on the meter.

When I split the relay and powered the clutch from another source the amperage with the ac on was .2 amps.
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
You seem to be doing the right thing.
The symptoms you see are what you would get with an intermittent short, they are the devil to find as the insulation is often worn away out of sight.
I would be inclined to replace the wire to the clutch. People often do not consider a cable to be a component which can fail.
 

JerryG

Dec 17, 2013
20
Joined
Dec 17, 2013
Messages
20
I plan on replacing that plug and two wires next . It looks like it needs it.
 

Tha fios agaibh

Aug 11, 2014
2,252
Joined
Aug 11, 2014
Messages
2,252
The car is a 99 Lincoln Continental. The diode is in the fuse box along with the relay.

That might be across the relay coil.

I would want the diode right at the clutch where the connector plugs in.

Another thing to look for, is wiring that might have been melted by touching any hot metal, or perhaps rubbing on a pulley or belt.
 

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
Definitely sounds like an intermittent short in the wiring from the fuse to either the clutch or to some other load.
 

JerryG

Dec 17, 2013
20
Joined
Dec 17, 2013
Messages
20
Changed the plug. Problem still exists. I do not think that the problem is intermittent. It consistently blows the fuse under the same conditions and not randomly. When I split the relay the fuse to the clutch blew when the pcm cut the power to it. It has to be the coil or something with the diode.

How would you use a bulb to look for shorts? Do you power it with the clutch plug then look for dimming or is there a test where you look for any light?

Will look closer at the diode connections when I get some time.
 

Martaine2005

May 12, 2015
4,932
Joined
May 12, 2015
Messages
4,932
Hi Jerry,
Some people do use bulbs to find short circuits.
The bulb goes in series with the load.
Or for a car can replace the fuse. Just make sure the wire connected from fuse box to the bulb is heavy gauge.
I have never done this but know many car electricians that do this.

Martin
 

Tha fios agaibh

Aug 11, 2014
2,252
Joined
Aug 11, 2014
Messages
2,252
Some good suggestions, but I still think your getting spikes from the highly inductive load when it is switched off. Your plugging and unplugging the clutch does not simulate the actual (reactive) circuit.
The best place to have a flyback diode to suppress the inductive spike is right at the coil, not 10ft away.
To prove the wiring is not the problem, You could unplug the clutch, and put a resistor or lightbulb in its place. Drive it around for a few days and see if the problems remains.
 

JerryG

Dec 17, 2013
20
Joined
Dec 17, 2013
Messages
20
I will try putting a diode right by the clutch. A couple of questions. I have several extra plug-in style diodes that go in the fuse box. Should I use one of these or is there a better choice to handle the surge? I would need help selecting a diode that would be better choice. Second question is should I wire it in series or parallel?
 

Martaine2005

May 12, 2015
4,932
Joined
May 12, 2015
Messages
4,932
Hi Jerry, according to what I have just read, a 1N4007 or 1N4004 would (or seem) to work well as a flyback diode.
Leave your other wiring as it is.
series or parallel?
Just to complicate things...Antiparallel.:)

Martin
 

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
Wire it in parallel with the coil. The diode must be reverse-biased, i.e. its cathode (usually marked with a coloured band) should be more positive than its anode.
Edit: Martaine beat me to it. Reverse-biased = antiparallel.
 

Martaine2005

May 12, 2015
4,932
Joined
May 12, 2015
Messages
4,932
This is what I was reading..

The process of choosing the proper flyback diode for the suppression circuit is quite simple. First, determine the current the clutch or brake demands. For example, the Warner ERS-42 electrically released brake needs 0.239 A at 90 V to release its total maximum holding torque of 7 lb-ft. The specifications of the flyback diode should, at a minimum, double those values: a continuous forward current (If) of at least 0.478 A with a peak inverse voltage (PIV) rating of 180 V. A common diode used in this application is the 1N400x Series. The 1N4004 has an If of 1 A, with a PIV of 400 V, or four times the demand of the brake.



Low-voltage brakes usually need more current to operate. The release current for the 24-V version of the ERS-42 is 0.973 A. So it needs a diode that can handle at least 1.946-A If. In that case, switch to the 1N540x Series — specifically the 1N5404 with a PIV of 400 V and an If of 3 A. Note that the 1N5401 would also work in this application, though its PIV rating is only 100 V. That is still over 4× the 24-V applied to the brake. The cost of the 1N5401 and 1N5404 diodes are typically the same and the 1N5404 can serve in place of the 1N4004 if necessary, even if its physical size is slightly larger. A higher PIV rating does not affect circuit operation in this application.

Martin
 

Tha fios agaibh

Aug 11, 2014
2,252
Joined
Aug 11, 2014
Messages
2,252
Good info Martin.
I have heard of the doubling rule of thumb.
But I wonder if it's always adequate, given the fact that inductive spikes can reach about 10 times the voltage. Example; a 24v relay coil can release spikes of >300 volts.
Wouldn't it prudent to use a PIV closer to 300v, rather than say; 50v?

John
 

Martaine2005

May 12, 2015
4,932
Joined
May 12, 2015
Messages
4,932
Hi John, to be honest, I have no idea.
I was just reading about a similar story and found the result |I posted.
I thought it might have been useful, but obviously not.
My keen interest to help does sometimes obscure the obvious!!

Martin
 

Tha fios agaibh

Aug 11, 2014
2,252
Joined
Aug 11, 2014
Messages
2,252
No problem. I'm sure one of our esteemed posters will chime in.

I too share your zeal to help, but I'm limited by the resources between my ears.
 
Top