Intermittent AC operation

[RDoc]

Intermittent AC operation (SOLVED)

Solution: http://www.nsxprime.com/forum/showt...AC-operation?p=2037673&viewfull=1#post2037673

On my 1991 NSX the AC doesn't always come on. It seems pretty random, sometimes when I start the car it's fine, sometimes it doesn't come on.
The CCU was rebuilt and passes all the self tests and the display looks and behaves normally.
The climate control self test seems to work
The fuses are OK
The AC was just charged professionally
When it runs, cool air comes out all the vents
Starting and stopping the AC seems to have no effect
When it runs, the air temp is about 45F with ambient 75F. It cools the car OK
When it works, it doesn't matter if the car is in motion or not.
The coolant was changed to 134A a long time ago
There doesn't seem to be a sight glass - because of the coolant change?

Does anyone have any ideas?

 

[dablackgoku1234]

Fan control unit: http://www.nsxprime.com/forum/showt...mittent-Condenser-Fan-Mystery-I-ve-searched-(

 

[RDoc]

Thanks for the reply, that sounds worth investigating. I'm away for a couple of weeks but will check it when I get back.

 

[Old Guy]

The fan control unit is a good place to start because they have a 'bit of history' with intermittent failure and flakey operation. They are typically repairable if the unit is the problem. However, since you had the refrigerant replaced and you seem to be uncertain about the effectiveness of the AC system, perhaps have your refrigerant pressures checked. If you have low refrigerant pressure or a marginal pressure switch, you may be just on the edge of permissible operation causing an intermittent permissive from the pressure switch. The other item that could be a problem is the fan clutch relay. However, relay failure is less likely to be intermittent. They tend to stay dead; but, checking the relay is really easy to do so simple to cross off the list.

 

[Tio Juan]

My A/C has been flaky for the last 2 years and finally quit completely - no condenser fans or clutch engagement. The manual pointed to possible CCU so I sent it to Briank for a check up but he found no problems (Thanks, Brian!). Following the Fan Control Module thread above I re-soldered the 12-pin connector's leads to the board and all is good now. I have a high-end soldering/rework station so it was pretty simple.

 

[RDoc]

I removed and opened the fan control unit from behind the seats and found the electrolytic capacitor clearly buldged. Amazon was the only place I could get a replacement quickly so I got 4 and replaced the original on the board. This seems to have completely fixed the problem. I can hear the motor RPM change at idle when I switch the AC on an off and it blows cold air.

I also checked all the solder joints for cracking and cold solder joints but they all looked fine so I didn't mess with them. For anyone doing this, note the stripe down the side of the capacitor and the white printed band on the left connector on the PC board. That shows the polarity of the capacitor. Be sure to put the new one in with the same polarity. If you don't the results can be spectacular (bad!)

Overall a simple and satisfying fix.

 

[Old Guy]

Is that photo the original capacitor? If so, I don't think that qualifies as a failure budge. Electrolytic cap failures usually involve ejection of the rubber plug on the bottom of the capacitor where the wire leads enter leading to leakage of the electrolyte - the common problem for ageing capacitors on the NSX. Catastrophic failure of the capacitor caused by an internal short ruptures the blast disc on the top of the capacitor (typically Y shaped grooves on the top of the can).

The fractures in the soldering joint can be exceedingly difficult to see with the naked eye. I have been trying to diagnose a blower fan running at odd times (CCU turned right off and it still runs) and have done the plug / unplug the fan control unit to isolate the problem. The problem fails to re materialize when I reconnect the unit so I think I must be disturbing a solder joint which makes the problem go away. I plan to pull the FCU and reflow all the joints to see if I can make the problem stay away.

 

[RDoc]

Is that photo the original capacitor? If so, I don't think that qualifies as a failure budge. Electrolytic cap failures usually involve ejection of the rubber plug on the bottom of the capacitor where the wire leads enter leading to leakage of the electrolyte - the common problem for ageing capacitors on the NSX.

The bulge and dripping on the side is certainly not a good sign to say the least! The seal at the bottom seems to have held up and the capacitor wasn't completely dead, the AC still worked intermittently. Replacing it fixed the problem in any event.

I photographed the board with a magnifying lens and looked over the solder joints. They were all smooth and shiny so if there is a solder problem, it's pretty subtle.

 

[Old Guy]

That drip on the right side of the 100 uF capacitor looks more like a huge glob of conformal coating. If you look at the other components they all look like they are coated in some form of varnish, almost like the whole board was dipped in a vat of the coating rather than just being sprayed over the exposed mask on the back. However, I won't argue with success. If replacing the capacitor solved the problem that is all that really matters..

 

[RDoc]

When my wife and I were doing software, we had a saying: "There's no such thing as a minor bug". Until you understand what's really going on, you have no idea what the implications are.
Personally, I think that's true of things like capacitors. If it's clearly not correct, in this case it's not cylindrical, you don't know how bad it is, however, you do know that it's not right. Attempting to decide if it's a bit bad. kind of bad, bad, really bad, or dead seems like a very questionable approach.

Based on my experience I'd strongly suggest that if someone opens up a board and sees bulged capacitors, they should replace them without a second thought.