More strange behavior today. Had a consistent 12v all the way to the compressor, just would not click it on.

Then, randomly all the test points dropped to 3v.

This was testing and jumping the freeze and low pressure switches. Connecting the compressor straight to 12v+ clicks it on just fine.

Sound like the internal relay now?

Quick question, what exactly does the climate control unit do when controlling the AC compressor?

Does it just turn it on and off based on the inside temperature? Or doe it get some sort of on/off cycling from the control unit itself?

Seem like the high pressure switch is what would cycle the compressor when the system based on the pressure in the system ?

Its simpler than you think - if the AC is on its just on - it isn't intended to cycle and rarely does - only if the freeze switch cuts out which is extremely rare - the pressure switch only cuts out of you have insufficient refrigerant.

Temp control is still via heater flap adjustment - so using AC in cool weather is very inefficient - you chill all the air coming in then heat it back up again.

Alan

The reason I ask is because I noticed in the OP comments that the AC switch basically has two outputs, one that runs the fans and one that runs the compressor from the relay.

I cheated just to check the rest of the system and jumpered these two leads together. So now when you push the AC switch my compressor comes on. All the other safety checks are still in line with the system, the power is just not running through the relay.

So what is the purpose of the relay in the control unit?

Seems like if you wanted a relay you could also just use the fan output from the AC switch to trigger a relay. (which would seem even simpler to wire in)



http://farm7.static.flickr.com/6075/...136789_b_d.jpg

The reason - (apart from protecting the switch which is probably not designed for that much current - a warning to flip it back) it that the AC compressor doesn't only come on due to the AC switch being on - it is also triggered (depending on year) by the DEF mode and also by recirculate mode - so regardless of the posistion of the AC switch.

Using just the switch would eliminate those modes - the relay is also dual pole and feeds the main HVAC control unit a compensation signal of some kind when the AC is on (not sure what for)

Alan

Well.....at this particular moment I haven't eliminated any function since they did not work in the first place:thumbsup:


Will repair the relay soon.

Voltage is not power. It is entirely possible to have a full 12 volts and not have enough power to pull the clutch in. For an illustration, suppose that the wire to the compressor clutch was almost broken, with only one tiny strand making contact. It would show a full 12 volts with no load, but would not pull the clutch in. It is not uncommon for the relay to make contact and show some voltage, but to not pass enough power to operate the clutch.

+1 for checking the power chain to the compressor clutch you really need a decent dummy load like a headlight bulb. If there's a problem with relay or corrosion in freeze switch you can see 12V then this will drop to a few volts or less with even a small load.

"It is not uncommon for the relay to make contact and show some voltage, but to not pass enough power to operate the clutch."

Actually, very common when the relay contacts oxidize over time and/or
from currents which exceed the relay's rating, as is the case for the
original (non-updated relay) CCU.

Read here under 'Generic': http://www.systemsc.com/problems.htm

Thanks Dwayne for this excellent documentation. I just completed the operation and Hooray I now have A/C for the first time. Keep up the great work!

I about spat beer out reading this when I saw the tidy job of Velcro'ing the new relay to the side of the head unit.

I'm such a hack!!!

:bowdown:

Has anybody had any experience in repairing the 1,2,3 in the window when you twist the knob? Mine is stuck between 2 & 3 irrespective of where the knob is. The blower works fine and changes speed.
Thanks
Bilal

Parts of this make sense others do not.

Yes you can use this method to figure out how much current the clutch takes. However to do this you have to break the circuit and insert the ammeter in series... somewhere... as you note you can do a total bypass direct measurement as a reference (however "normal" may be lower than this due to resitive actual wiring etc etc).

You then propose inserting the meter at multiple points in the curcuit - this is actually quite a lot more difficult to do practically. Also whatever current you measure in a circuit will be the same everywhere in that same circuit - it matters not where you measure it - if there is a poor connection the current will just be lower everywhere.

So a realistic reference measurement of your (or a typical) clutch alone would be useful to compare against the total actual circuit measurement - but that is about all.

Voltage measurements are actually typically easier to do and usually more useful - as long as the measurement is made when the curcuit is in operation (e.g. under load - in this case the load of the clutch).

When the measurements are confusing & not consistent - often the measurement method is suspect OR there are actually constantly changing conditions (but this is rare).

Make sure the reference ground is known to be good & solid - also try measuring directly across the load terminals to eliminate possible connection issues in the ground side of the load connections.

Alan

What you just described makes good sense - for the AC circuit primarly - because that happens to be much more accessible for individual connections than most..

However it isn't what you actually said before..

You said.. Now its obvious what you actually meant by that - but the literal reading is quite different..

I point these things out because lots of folks know very little about electrical debug and are looking for good easy techniques - and its important not to set them off down the wrong path. Especially in a generic sense - current probing is hard to even do in most 928 circuits - e.g. multipole connectors and all...

And genericaly for debug it usually isn't actually needed and to me would always be a secondary technique.

I am not an armchair quarterback either - I do spend a lot of time helping people debug challenging things... often their measurement tools & techniques are not so good and it helps to keep it simple... You seem to have a good idea what you are doing - but its really isn't easy to implant that to others...

Alan

I try not to post things that are wrong - But I see no inaccuracies in what I said above...

Does anybody have the part number of the relay for saving the existing relay in the head unit by installing a relay for the compressor in the engine bay as discussed earlier in the thread? My 91 has a brand new head unit installed by the PO for $1200 and don't want it to blow the relay.
Also what I had previously asked "Has anybody had any experience in repairing the 1,2,3 in the window when you twist the knob? Mine is stuck between 2 & 3 irrespective of where the knob is. The blower works fine and changes speed. "
Thanks
Bilal