Ryan James

So a while back my a/c stopped working on me but after some other repairs and bleeding the coolant system it seemed to start working again. Unfortunately, as it got hotter outside my a/c stopped working again. I followed the steps in the service manual to figure out what I could but am lost. The compressor clutch won't engage when the a/c is on, and for some reason the condenser fan runs almost always, even when the a/c and everything is turned off.

After following the diagnostics listed in the manual I found that for some reason the pressure sensor switch is only sending 8 volts to the cpu. I further found that the switch is getting 12 volts and doesn't show any ohm resistance when I test only the switch so I have no idea where it's losing 4 volts. A hunch has me thinking the plug going into the pressure sensor is bad, but I don't know how I can test it. I went ahead and ran a pressure check I got 130 psi static pressure, and then I shorted the relay to the clutch to get the following readings: 60 psi on low and 260 psi on the high side with an ambient temperature of 98.

I know very little about the a/c systems so any tips are much appreciated. Also the car is a 97 Honda Civic DX

ezone

for a 97 your pressure switch is on tip of the drier, yes?

Use a paper clip to jumper the pressure switch connector and see if that allows the system to turn on the compressor.

Ryan James

Yes it is.
​​​​​​I went ahead and tried the paperclip but it didn't work. When I did it I also backorobed the connector after reconnecting it to the switch and one side gets battery voltage, the other side only gets 8.4 volts still when backprobing it.

ezone

With the pressure switch connector jumped, provide a ground (jumper wire) to that paper clip and see if the system will turn the compressor on.....



Also need more info about the condenser fan running?

Ryan James

Won't grounding out the paperclip just cause a short in the system? Sorry I just want to make sure I understand what I'm doing on that one before I go try it out.

Also as far as the condenser fan it seems to turn on as soon as the key is turned to the run position no matter what settings the a/c is set to, even witht it set to 0 with recirculation and A/C off. I noticed earlier today that at some point it will start turning off and back on at a normal rate, but I couldn't pinpoint what made it start to cycle.

ezone

The 12v you see (on the blue/white wire???) should originate at the condenser fan relay, but at the same time the ECM puts a bias voltage (5v or 12v are typical but the diagrams don't specify what this uses) on the request line (blue/red at the pressure switch) so it can sense when the thermostat in the evaporator core grounds the request line which also provides the ground to turn on the condenser fan relay.


What I want to know is if the system will run the compressor when the blu/red (at the pressure switch) is grounded, that will give me some direction to go.


There are several harness connections between the AC thermostat and the blue/white of the pressure switch that could be faulty in addition to the pressure switch and its terminals. Any evidence of corrosion in the wire connections there?


Do you have a decent wiring diagram of the AC system?

Have you checked the relays in the underhoood fusebox to see if someone has hacked it to keep the fan on, or hotwired the fan direct to some other circuit?

Ryan James

Ah, I think I follow what your saying there, and the colors sound right from what I remember. I will try it out and double check the connections in the next 30 minutes and let you know what results I get.

I did actually check all the relays, and I even tested them following the methods shown in the manual, as well as switching them with other relays that were working.

I've been using an actual Honda service manual that a friend sent over to me for the wiring diagram, and even the diagnosing I've followed so far.

Ryan James

Grounding​ the paperclip caused the compressor to engage properly.

Also for some reason the condenser fan seemed to work properly even before I made the paperclip connections. Earlier I did disconnect the battery for a bit, maybe that fixed it again? However the compressor only clicked on twice rapidly and then stopped again when everything had normal connections

ezone

Uhhhhhhh..... It is supposed to only run while the ac clutch is engaged.

While it's acting wrong, remove the relay and see if it stops running?
If it's acting right, don't bother doing this until it acts wrong..

Ryan James

Well then it wasn't running right, it clicked on when I expected the clutch to kick in, however it ran right when I connected the paperclip and grounded it.

Back when the problem initially started I actually removed the fan relay as a temporary fix and it did work in stopping the fan from running.

ezone

Ok so now you have to figure out

if the AC thermostst is actually grounding the request line, and
if so, where the poor connection or loss is.
Probably not related, this has minimal electronics involved: simple transistors and relays, not microprocessors.

Not sure I understand this line?

ezone

Ok if the AC system was turned on BUT the compressor was unable to operate, the fan would still be able to run as long as the AC thermostat is trying to tell the system to run.

What makes it confusing is you stated "no matter what settings the a/c is set to, even witht it set to 0 with recirculation and A/C off."
which means something else may be keeping the fan relay turned on (providing ground to the relay coil) when it should not be.


Good clue. That may be handy info later.

Ryan James

I'll start following this connections and see what I can find. Thanks so much for the help!

That last line was trying to explain what the compressor did before I made any changes to connections and everything was hooked up normally. When I turned the A/C on the compressor clutch engaged and disengaged twice in just a few seconds then remained off.

ezone

Looks like you're chasing a poor connection or open between the thermostat and the low pressure switch.

If the cond fan relay is being triggered at the correct time, then the stat must be able to provide the necessary ground, and the wiring must be intact at least from the AC thermostat to the point where that part of the circuit branches (looks like within the fusebox) ......

So I think the problem probably will be found between the fusebox and the pressure switch.

Connector C214 is below the fusebox
C209 is out front, looks like on the fan shroud, that is a great place for wiring damage and corrosion LOL

Do your wire diagrams include photos of connector and component locations? Connector terminal diagrams? Those are a big help when chasing wiring.

Ryan James

You sir are quite brilliant I must say! So I was following the wire until I got to c214 where I found the red wire in c214 is actually melted, so it looks like a new harness in my future. What does the red wire go to? I'm assuming that somewhere in this wire I've got even more damage that I'm not seeing since this wire is melted and charred on both sides of the harness

ezone

Solid red wire in C214 is for the compressor, goes from the comp relay in the fusebox to the compressor clutch

If it burned or melted, see if it originated within the connector. Loose or poorly fitting terminals can generate a lot of heat while electrically loaded.


Now that I look at a picture closer, the C214 sits below a small relay box (ABS controls?) not the main fusebox?


A harness replacement may be a nightmare LOL
I do a lot of wiring and harness repairs, rodent damage is prevalent in my area.

Ryan James

The connector doesn't seem loose, in fact it seems pretty well stuck in place. There must be some bad connection right there like you said though, because I followed the wire to the fuse box and towards the clutch and all melting/charring go away within a short distance of c214.

On mine c214 is actually not covered, but that may be because I don't have ABS if that's what you were seeing.

Im going to price harnesses for now and hope this is all it will need. Besides the obvious clutch wire potentially shorting out... If I were to temporarily patch up the melted wire and cover/protect it, how much damage could I do if I left the paperclip and ground rigged up for a day or so of driving?

If that is a viable temporary fix I'd love to do it as this is my daily driver and it's getting hot here in Phoenix haha

ezone

Do not do that:
Paperclip, yes only if the pressure switch is stuck open and must be bypassed to run, but do NOT leave the thing grounded because the compressor would not be able to cycle because you would have eliminated the request/temp control/cycling input from the thermostat.
.


I'd patch and jumper damaged wires as needed, maybe even completely bypass anything damaged within C214 to make the system work again.

Assuming the problem with the blue/white wire (high resistance or open circuit) also lies in the same area.

Ryan James

Ah, I was afraid you'd say it would keep it from cycling. Oh well tomorrow I'll patch a few of the wires and see what I can get running. Thanks so much for the help! I'm definitely giving some +rep

old97civic

Interesting to follow along... A/C is one of those things I never needed to worry about on my car, because it doesn't have it

Ryan James

So I'm guessing it's a bigger problem. Today I cut the blue/white and the blue/red at the pressure sensor as well as the red wire near the a/c clutch and connected them directly to some spare wire I had. I then took that wire all the way up to c214 and then I went another 6-7 inches past the connector and cut only these 3 wires, and then soldered my new wire in. I made sure to connect the new spare wire to the side of wire that goes to the relay box and firewall, and it still isn't working. Everything actually still tests the same (clutch works when I use the previous grounded paperclip method) after I made these connections so it doesn't seem like I made anything worse.

I was going to check for an open wire in the connections, but I am getting confused as to where the wire harness goes once it is inside the firewall. If I check the wires at the a/c thermostat to the wires at the a/c thermostat there should be no resistance between like wires, right? Or what would be another way of checking these individual wires for hidden problems? Could the A/C thermostat need replacing?

Ryan James

So I took some shots at testing. The blue/white wire from the pressure sensor has no resistance when checked at the blue/white at the a/c thermostat. Then I looked at the diagram I have and it looks like the blue/red at the a/c thermostat should be grounded when I turn the a/c on, right? If I read the diagram correctly then that should be where my problem lies as the blue/red isn't grounding.

Edit: at the a/c thermostat the voltages are as follows. Blue/white and the black/yellow are getting over 12 at all times. The blue/red is getting just under 5 only when the a/c is switched on so I am guessing I read the diagram wrong

ezone

No resistance (zero ohms) is very different from open circuit (infinite, OL or over limit, dashes, or blanks). Which is it?
Is your ohmmeter autoranging, or do you have to manually select 200/2k/20k etc.?

It's very hard for me to follow a written description LOL

Remove the relay for the compressor clutch so the melted red wire can't become powered which might affect other circuits shorted to this wire. I would separate and isolate or maybe even cut apart any that melted together.
What other wires were damaged?

The AC thermostat will provide a ground the blue/white wire when it believes the evap core needs to be colder....but the thermostat has to be enabled (turned on) by the control panel before it can do this.

So for testing, all you need to care about is that it does ground the blue/white wire at the appropriate times.
That means you may see some sort of voltage while backprobing that wire up until the time it grounds the circuit, then it should show zero volts,
At this time the condenser fan turns on (you said it was doing this correctly anyway) so I assume the thermostat does its job.....but it has an open circuit somewhere between the blue white wire at the condenser fan relay and the pressure switch on the drier.

===========================================

How about this......Have you patched up or bypassed the red wire? and if you jumper the compressor relay socket does it engage?
Plug the relay in. We already verified the ECM can turn this on when it sees the blu/red wire at the pressure switch is grounded.

Go into the dash and unplug the 3 wire connector for the thermostat, backprobe with a paper clip and jumper wire to ground the blue/white wire there.
I say backprobe it so you don't spread out the female terminals.

Does the AC run?
Does the condenser fan turn on ?
If yes to both, we work on the control side.
If no compressor operation but the condenser fan does turn on, you need to check if the same blue/white at the pressure switch measures zero volts. If you have it manually grounded at the thermostat but its not grounded at the switch, it's open somewhere between the two.

Ryan James

The blue/white from the pressure sensor, ro the thermostat had no resistance and is closed like it should be, now that I have patched it.

I have to select settings on my voltmeter, I went to ohms section and set it to ground option (it's the lowest setting)
Only the red, and one of the blue wires were damaged. I actually replaced all 3 wires just to be on the safe side.

Since patching the wires the condenser fan is actually no longer powering on when I switch the A/C and it hasn't had any of those random times where it stayed on either anymore.



Yes, this still makes the clutch engage

Doing this made the a/c function normally again. Both the clutch engaged and the condenser fan turned on.

Sorry it took me so long to respond, I have an odd work schedule.

ezone

Does the entire system work properly again now that the wiring problems have been corrected?

Ryan James

No, now neither the condenser fan not the clutch engage when I turn the a/c on normally

ezone

But when you grounded the blu/wht wire at the AC thermostat, the compressor clutch and condenser fan both kicked on, right?

So that means the engine compartment stuff will (should) work once the control panel and AC thermostat tell it to.
(Stop if this is an incorrect assumption)

Assuming it does...
Next need to figure out if the thermostat is faulty or if the control panel is faulty.

Short version is the control panel grounds the blue/red wire to the thermostat when you turn on the AC (fan + button).
This enables (turns on) the thermostat.

Thermostat then watches the evaporator temperature and requests the compressor when the core temp is above 37*F.
It grounds the blu/wht wire to request AC. (If everything in the engine compartment is ok, this signal makes its way to the PCM and it turns on the compressor, and you said this did happen with a jumper wire.)

Thermostat is transistorized so it has to have a power source as well, so that needs checked.

Aaaand go!



Thermostat 3 wire plug disconnected, key on, check for 12v on the blk/yel wire
(+ from dash fusebox #17). OK?
Assuming ok.

Connect thermostat connector.
Start engine, turn HVAC control panel on AC/cold/fan on

Backprobe the blue/red wire, determine if it is grounded. If not, use a jumper wire to ground it.
AC system should kick on and run
Does it?


If there was no ground supplied to the thermostat on the blue/red wire, go to control panel and double check there at the same wire. If the control panel can't ground it when you turn on the AC, it's bad.

If the blu/red was grounded but the thermostat can't kick the system on, stat is bad.
You can ground the blu/wht again just to make sure the AC system does run from the stat plug.







See, the stat failing is kinda unusual, except if the melted wires you found have shorted 12v power into the transistor that was trying to ground the request line. I could see that damaging the thermostat electronics.
Kinda tough to just say replace any one thing without knowing if it's really at fault.

Ryan James

I ok, so it's definitely not getting any ground through the blue/red when I turn in the a/c. It'll be a few days before I can go through and take the a/c panel out and check the wire there, I'll let you know what I find then and post results. I think this problem started when someone stole the stereo, the wires seem tight but they may have broke some.

In the mean time would backprobing a ground a ground wire to the blue/red be the same as turning the a/c on? If I follow the diagram right that would allow me to use the a/c temporarily, and keep the a/c pressure sensor and thermostat wired properly, right?

ezone

Oh. Well yea check that area then.

Absolutely, you would be providing the ground that the control panel can't supply to enable the AC thermostat. The rest of the system would then be able to work as intended.

If it works for a day or two, you could secure the wiring better, run a toggle switch for it and it could be permanent LOL

Ryan James

So it took me a bit longer to get back to the a/c than I expected but I am pretty sure I found the problem after following the rest of the diagnosis laid out in the service manual.

I checked several wires at the heater switch to find they are all good, blue/red gets voltage, ground wires grounded, and all connections seemed snug and tight. So I have it narrowed down to the heater switch, I know it can actually be tested but I'm going to just replace it and the whole control panel anyways as it has cosmetic damage from my the break in. In the mean time I just have the thermostat grounded out for a manual A/C switch

Thanks for all the help in getting this straightened out, idk if I would have ever found the melted clutch wires without some assistance.