What's new

Diagnosing Aircon Problems

freddofrog

won 16.4k on Euro lottery :)
Messages
7,326
Reaction score
847
Location
Off abroad
Car
TBD
Note that all pictures below are from my car which is a 7th Gen 2.4 Tourer. The way the aircon system works will be the same in other models, but there may be differences in the layout of the fuse/relay block in the engine compartment.

The schematic below is from the Haynes for the petrol Honda Accord 2003 thru 2007, USA MARKET. I have found most of the sections on engine and transmission to be compatible with my car. In the schematic, I have labeled the relay used by the system to switch the compressor clutch on/off, as B. This relay is a good place to check the compressor.

Picture3.jpg


The picture below shows the location of the compressor clutch relay B in the fuse/relay block.

Picture2.jpg


With the engine off and without the key in the ignition, remove the relay. Removal is made easier by pulling out relays A & C (the two blue Omron relays), after which the compressor clutch relay B can be removed, then replace relays A & C.

The picture below shows the compressor clutch relay removed.

DSCF0742.jpg


The picture below shows a wire inserted into one of the slots. This slot is the feed to the compressor clutch.

DSCF0744.jpg


By connecting the wire to the positive battery terminal, the compressor clutch will be energized. This can be done without the engine running, and without the key in the ignition. The picture below shows the other end of the wire on the positive battery terminal. When this is done, an audible clunk should be heard, which indicates that the compressor clutch is working. If you have a DVM, the current is around 3 amps (so gauge of wire is not important). Do not leave the wire like this for more than a few seconds.

DSCF0745.jpg


If you cannot hear a clunk, then you have a problem with the compressor clutch (either the solenoid is not working, or the thermal protection circuit in the compressor has operated, or possibly the clutch itself is jammed). see here for exploded view of compressor

If you can hear a clunk, perform the next check. Use the wire to bridge across to the slot adjacent and parallel to the slot where you placed the wire earlier.

DSCF0746.jpg


If you cannot hear a clunk now, then check the compressor clutch fuse (D in the schematic and earlier picture). Remove the fuse using the accessory for pulling fuses, as shown in the picture below.

DSCF0747.jpg


If you did hear a clunk when connecting the wire to battery positive, and also when bridging to the adjacent slot, then you either have a faulty relay, or the system is not closing the relay.

The picture below shows my relay. This was faulty. I removed the plastic cover and sprayed the relay contacts with switch cleaner, and the relay has been working ever since. This saved me a few £10’s in the cost of a new Honda relay, and it has been working like this for a year now.

The two upper tabs in the picture operate the relay coil, and the two lower tabs are connected to the relay contacts.

DSCF0743.jpg


The picture below shows how to check a relay. I managed to do this with one hand (other hand is holding the camera). It’s easier with two hands, hold one of the coil tabs against the battery terminal then touch the other coil tab with a wire connected to the other battery terminal.

DSCF0748.jpg


Obviously, DO NOT let the wire touch both battery terminals at the same time, else you will get a big flash (and a small probability that a battery could explode). If you don’t have a steady hand, don’t attempt this.

DSCF0749.jpg


If the relay is working, you will be able to feel and hear the contacts closing.

An alternative is to run the engine until the engine cooling fan starts to operate. Then swap the compressor clutch relay with the cooling fan relay (they are the same). If the cooling fan still operates, then the relay was ok.

SUMMARY

If you can hear a clunk from the compressor, then it is working. If the compressor clutch fuse is ok, then check the relay.

If the compressor clutch is working, and the fuse is ok, and the relay is ok, then you have a fault elsewhere in the schematic diagram e.g. low gas pressure

HOW TO CHECK IF YOU HAVE ENOUGH R-134a GAS

Remove the compressor clutch relay. Start the engine, and put the aircon to its coldest setting. Make sure you can feel the blower running and air coming through the vents.

Now energize the compressor clutch by connecting the wire from the slot to the battery positive terminal. WARNING, DO NOT LEAVE IT LIKE THIS FOR MORE THAN 30 SECONDS, DEFINITELY NO MORE THAN A MINUTE AT THE MOST, AND DO NOT REPEAT AGAIN FOR ANOTHER MINUTE. IF YOU LEAVE IT LIKE THIS WITH THE ENGINE RUNNING YOU MAY OVER PRESSURISE THE SYSTEM. There is a safety pressure valve in the compressor, but do not rely on it operating !!!

With the compressor clutch energized, if there is enough gas in the system, you will feel cold air coming out of the vents. If you are confident that the compressor clutch relay is ok, then the pressure sensor may be faulty.

However, there could also be other faults that are stopping the system from operating the relay, such as faulty temperature sensors.

see here for locations of temperature sensors

see here for exploded view of heater and evaporator unit

[url="http://www.hondaoriginalparts.com/honda_car_parts_selection_pfk.php?block_01=17SEA01&block_02=B__5800&block_03=2768] the condenser can also be perished and inefficient [/url] (this is in front of the radiator and can be seen through the front grills

aircon pipework
 
Great write up Brian,

Thanks for taking the time to do this, it should help answer quite a few questions.

Thanks for including the warnings, they should help keep people be aware of potential issues.

rep point on it's way !
 
Good info there dude. Ac can be a nightmare.
 
Very good write up mate, rep point coming your way.
Just really enjoy reading detailed threads like this.
 
great write up fella a well deserved rep point i think.
 
no idea what a rep point is but good article..! :p

and yea, AC is a nightmare..

all the things that can go wrong is ........
 
Thanks for all the positive rep in my direction (Eric, rep = reputation), you're encouraging me to write some more: I might be taking my front bumper off again to replace my fog lights, when I get round to doing it I'll take pics and write it up ....but don't hold your breath waiting LOL

btw for those who don't know, aircon is similar to your household fridge, or household freezer (combined fridge/freezers are a bit more complicated).

If your fridge has some black tubing at the back, this is the condenser, and on your car the condenser is in front of the engine radiator. A condenser is like an engine radiator.

The compressor in your fridge works the same way as the compressor in your car. They are both switched on and off by a relay. On your fridge the relay switches mains power on and off to the compressor, though on the car the relay switches an electric clutch, because the compressor is driven directly from the engine.

On some older fridges and freezers you can hear the compressor starting and stopping. You can also hear the compressor on your car starting and stopping (when the aircon is working properly).

The relay in your car and the relay in your fridge are both operated by a control system, according to your temperature setting and the pressure of the R134a gas when he compressor has compressed it.

The compressor in your car has to endure a more hostile environment, and, there is no clutch on the compressor in the fridge. In theory I suppose it could be possible to have a compressor in the car that is electrically switched on and off the same way as in a fridge, a freezer, and domestic aircon units ....but then you would require a much bigger alternator in the car to electrically power the compressor.

Hence the idea of a clutch in the compressor so that the engine directly provides the power to drive the compressor. This clutch, and the hostile environment in the engine compartment (dirt and water and vibration and wide temperature variation) are the reasons why you are more likely to have compressor problems in a car than in a fridge, or freezer, or domestic airon.

And finally, the reason why the relay in your car is more likely to fail than in your fridge or freezer, is because mains voltage is much higher than car voltage, mains is ac not dc, so the current is much lower and crosses zero 50 times per second, so much less arcing on the relay contacts.
 
Great write up. Hope I never need it!
+1 from me too.
 
I have checked the pressure sensor on my car and it is a switch. It is closed-circuit in low pressure, and open-circuit in high-pressure.

When the aircon has reached target temperature, the switch is normally closed, because the system cycles the compressor on and off using the thermistors in the cabin (driver/passenger area).

However, if cabin temperature is initially high and target temperature is set very low, the compressor will remain on all the time. When the pressure has gone too high, the switch goes open and the compressor goes off. I have checked this on my car, when the compressor went off I pulled the connector of the pressure switch and measured the switch as open-circuit for about 10 seconds, then it went closed-circuit. Plugging the connector back on, the compressor came on again.

Thus if your pressure switch has gone permanently open-circuit, or your connector has gone rusty, then your aircon will not work.

IMO you are more likely to experience this on diesel models than on petrol models. On diesel models, Honda have put the pressure sensor (and the valve, which I will come to) down in front of the compressor, exposed to all the elements. On the petrol models, the pressure sensor is on the inner wing, and the valve is inside the heater box in the cabin.

I have checked diagrams of 7th gen diesel and petrol models on the Lings website, and this seems to be a consistent difference. Why did you do this Honda ?????? [EDIT: 15:15 .... see my next post LOL]

Below is the location of the pressure switch on my car (2.4 petrol)

DSCF0750.jpg


Here it is in close-up

DSCF0751.jpg


Here is the location of the pressure switch on the 2.2 diesel (picture provided by Max10, thanks). Pressure sensor top, valve bottom.

IMG_7461.jpg


Here is the connector unplugged on my car

DSCF0753.jpg


Note that when unplugging, you have to press the plastic retainer clip, which you can see below (on the diesel, I think that the clip looks different)

DSCF0752.jpg


If you are satisfied that your compressor and relay are both working as described in the beginning of this thread, then unplug the connector from your pressure switch and insert a paper clip as in the picture below.

DSCF0754.jpg


Now start the engine and set the aircon temperature to about the same as the cabin e.g. if your cabin has been in the sun and outside temp is shown as 22 and you think cabin temp is say 26, then set aircon to 26. You should now see and hear your compressor cycling on and off.

On my car today, outside temp was 15 (yes June 5, 2012) and cabin temp was about the same, so I set aircon temp to 16. In order to test conditions in which the switch goes open circuit, I went out in the car with aircon temp at max, then when it was darned hot in the cabin I came back and set aircon to Lo, and waited a couple of mins for the compressor to go off, then quickly measured the pressure sensor and it was open-circuit for about 10 seconds. When I put the connector back, the compressor stayed on for a couple of mins, then went off for about 10 seconds, then came on again, etc, etc.

Thus be warned: WARNING, WHEN DOING THIS TEST, DO NOT SET YOUR AIRCON TARGET LOW WHEN THE CABIN IS HOT. AND ANYWAY, IF YOU USE A PAPER CLIP AS DESCRIBED AND THE COMPRESSOR COMES ON STRAIGHT AWAY, YOU HAVE SHOWN THAT YOUR PRESSURE SWITCH IS FAULTY.

From the tests at the beginning of this thread, if you are certain that your compressor works, and your relay works, and you are also certain that you have properly inserted a paperclip (or wire) into the connector, then if your compressor does not engage, then you probably have a wiring fault or some other fault in the heater controls.

THE VALVE. This is opened and closed by the system as part of temperature cycling. It is possible that the valve has failed, or the connector to it has failed. If this were so, then I would expect the compressor to initially engage and then go off and stay off (over pressure condition). Thus another WARNING, IF YOU HAVE SET THE AIRCON TARGET TO CABIN TEMPERATURE, IF THE COMPRESSOR STAYS ON (SYSTEM NOT CYLING ON AND OFF), THEN DO NOT LEAVE THE PAPERCLIP IN THE CONNECTOR (this is same condition as the warning at the beginning of the thread).

Finally, the diagrams on Lings:

PETROL PIPING pressure sensor is item 12
DIESEL PIPING pressure sensor is item 18

PETROL CONDENSER no sensor, no valve
DIESEl CONDENSER valve is item 11

PETROL HEATER UNIT valve is item 17
DIESEL HEATER UNIT valve is in the diagram but not itemised (Honda used the same diagram as the petrol but item numbers are different)



.
 
Top topic , wish it was put up a long time ago :rolleyes:

Mine was playing up intermittantly last year, but finally sorted it a month ago. Culprit was the pressure sensor/switch ( £54 from Honda)
 
Top topic , wish it was put up a long time ago :rolleyes:

Mine was playing up intermittantly last year, but finally sorted it a month ago. Culprit was the pressure sensor/switch ( £54 from Honda)
Thanks for that, because Matt (mattlee) and Max (Max10) both have the same problem with their diesels. The compressor and the relay are ok on their cars, and the conclusion is that it's the pressure sensor.

THERE IS A DIFFERENCE ON THE SYSTEM BETWEEN DIESEL AND PETROL

Both Matt and Max say that the plug on the pressure sensor has 3 pins (they have diesels), rather than 2 receptacles (my petrol). This means that you cannot do an over-ride test on the diesel sensor. This also suggests that the way the refrigeration unit works on the diesel is different from the petrol.

On the petrol, the valve is installed close to the evaporator in the heater unit, and the system cycles in more of a conventional manner.

On the diesel, the pressure sensor must be two combined, an over-pressure switch and an under-pressure swtich. The over-pressue switch is there for the same reason as in the petrol, to cut out the compressor when the pressure is too high. However, the low-pressure switch is there to help the evaporator unit to cool the cabin when the condenser is itself in cold air. In this situation the compressor/condenser could frost up, and this system ensures that the compressor/condenser unit will start properly (aka defrost pump down). The thermistors will also be placed differently inside the air flow in the heater unit. I suspect that "defrost pump down" is in the diesel because the diesel may be the same spec as in Scandinavia. The fact that the system in the petrol is simpler, shows that the system in the diesel is not really required.


ONE MORE ITEM OF INFO ON THE DIESEL
The glow plug light started flashing on both Matt's and Max's cars after the compressor/relay test. Matt has an OBD reader and cleared it.


.
 
the code i got from my OBD2 reader was P0645 - ac clutch relay error.

i have no idea why the glow plug light started to flash but after clearing this code the glow plug light has returned back to normal. :huh:

anyway ill get my pressure sensor changed and let you know how it goes.

thanks for all your help Brian, much appreciated mate. :D
 
Another great addition to the thread Brian,

Thanks once again for taking the time to do this and help members out.
 
Brian that's a seriously well written and in depth how to. Well done mate. Rep point coming your way!!!!
 
Hi Guys,

just joined this Forum and must say - Very Helpful!!!

So I'm experiencing same symptoms, cluch works, relay tested and works (clicks on touch of 12 Volts)
I have managed to check with tester voltage, that comes to relay. And it is constant 8.5V at all times when the ignition ON. Doesn't go up or down when AC is on or off.

Does it looks like Pressure sensor? How should I check on my diesel 2.2 CDTI as it has 3 Pins?

I would be pleased for any professional advice.

Thank you.

P.s.: The glow plug light started flashing too :)
 
Hi Guys,

just joined this Forum and must say - Very Helpful!!!

So I'm experiencing same symptoms, cluch works, relay tested and works (clicks on touch of 12 Volts)
I have managed to check with tester voltage, that comes to relay. And it is constant 8.5V at all times when the ignition ON. Doesn't go up or down when AC is on or off.

Does it looks like Pressure sensor? How should I check on my diesel 2.2 CDTI as it has 3 Pins?

I would be pleased for any professional advice.

Thank you.

P.s.: The glow plug light started flashing too :)

Good work. My car is the petrol version, which seems to be different from the diesel version, and I haven't tried any checks on the pressure sensor on the diesel.

If you can get to the pressure sensor, pull the connector off the pressure sensor, and then see if you can check the continuity of the terminals in the sensor.

There are 3 connections, which implies there are two switches with a common connection. One of the switches will be high pressure, and one will be low pressure (the petrol aircon system is different, on the petrol system low pressure is not sensed).

However, since no-one has measured the continuity of a working sensor on a diesel, it is difficult to know what a faulty sensor on a diesel will be like.

I suspect that a faulty sensor on a diesel will have at least one pair of connections open-circuit, or possibly both pairs open-circuit. See what you can measure and let us know.

cheers, Brian

PS .... anyone who has a DVM and a new sensor, please let us know the continuity

.
 
Just a further note .... if the 3 connections are in a line, numbered 1,2,3 left to right, then the pairs could be

1+2 and 1+3 (1 is common)
or
1+2 and 2+3 (2 is common)
or
1+3 and 2+3 (3 is common)

One would expect 2 to be common, but not 100% guaranteed

e.g. if you found continuity between 1 and 2, and no continuity to 3 at all, then either 1 or 2 would be common to 3, but 3 is open circuit.

I would *expect* a good sensor to have both switches closed-circuit. So if 2 is common, continuity between 1 and 2, and between 2 and 3.

Do not try to remove the sensor, the gas in there is at high pressure without the compressor runnimg (the compressor compresses the gas to a liquid)
 
Oh yes,

looks like very logical. Unfortunately i was seeing this sensor at night time and didn't know what to with it :) The location to the sensor is not the best but I think that I can manage to check continuity at least on connector so will know if electronics and wiring are working properly. See what I can do with sensor.
By the way, should I measure connector with ignition ON, just to double check before I'll get there.

I will let you know soon about my case.

Thanks,
Arturas
 
Question about High and Low pressure sensor.

As I understood there are 3 wires to this sensor 1, 2, 3.
Lets say 2 is main one and creates circuits between 1 (lets say high pressure) and 3 (lets say low pressure).
I came to the question what is happening when the air con is operating - should there be a circuit between 2 an 1 or 2 and 3. Or should there no circuits at all?

I'm not that deep in all of this, but my thought is to close the circuit on the connector for example 1 and 2 (if there is a circuit however) and then see if the system starts working (with AC ON). Something similar that has been done with Petrol version.

Maybe someone will have an idea before I'll put my fingers there :D

Thank you!
 
To measure continuity of the switches in the sensor, you don't need any power in the system i.e. you don't even need the key in the ignition.

In answer to your other questions, without knowing the exact layout in the diesel car of the piping, the valve, the sensor, and the location of the thermistors in the airflow in the cabin, it is not possible to predict the way the system would work.

But if we ***ume that 2 is common, and 2-1 is high pressure and 2-3 is low pressure, then I would *expect* a known good sensor on the bench to have both switches closed i.e. 2-1 closed-circuit and 2-3 closed-circuit. I would also "expect* this to be the state of the sensor in the car when the aircon and engine have not been run for several hours (key not required in the ignition).

The system in the petrol car is simpler and works by using thermistors and a valve in the cabin to control the temperature and compressor around small variations. If the system is trying very hard to cool the cabin, then the pressure sensor in the petrol car is an over-pressure sensor, the system will stop the compressor if over-pressue is detected.

The system in the diesel car does not work this way. As well as having a low-pressure switch, there is a valve next to the condenser (the petrol car does not have a low-pressure switch, nor a valve next to the condenser). Also, diagrams for the petrol car itemise a valve next to the evaporator in the cabin (this is normal) but diagrams for the diesel do not itemise a valve next to the evaporator. This latter omission could be a mistake, or it might not be a mistake. So it is not safe to predict the way the system works in the diesel car. Without taking a diesel car to pieces, IMO it is unsafe to try to over-ride the switches in the same way that I did in the petrol system.

However it is safe to measure the continuity of the switches in the diesel car when the aircon and engine have not been run for several hours, without needing the key in the ignition.


EDIT: DISCREPANCIES IN THE DIAGRAMS IN THE CABIN HEATER SYSTEM
Here are the discrepancies that I have noticed, which indicate that the aircon system for the diesel is not fully itemised in the diagrams

petrol car The valve is shown in the diagram and is itemised as 17 in the diagram and is in the list. Two thermistors are shown in the diagram, one is itemised as 21 and is in the list, the other is itemised as 36 .... but it is not in the list

diesel car The valve is shown in the diagram, but it is not itemised, and is not listed. Only one thermistor is shown in the diagram, itemised as 22 in the diagram and is in the list.

.
 
Max (Max10) has told me that he had the sensor on his diesel changed today, and now his aircon works.

He took the car to one place to get the gas removed, then another place to swap the sensor, then back to the first place to get it regassed (all pre planned). Sensor was removed without having to remove bumper. I think he said the place that removed and regassed charged £45, and the place that swapped the sensor charged £20. Obviously new sensor was from a Honda dealer.

He emailed me some pics of the sensor.

This is the new sensor

new_diesel.jpg


This is the old sensor

old_diesel.jpg


And a close up of the 3 pins in the sensor

pins_in_old_2.jpg
 
OK,

there was a confusing TEST, however i have some results, but my brain can not make any conclusions or thoughts about my aircon system problems.

So here I go...

1) I have tried to measure sensor it self for a closed circuits. I couldn't find any closes between 1,2,3 wires in any combination. (I wasn't sure about this ****ysis as I had just a bit of space so did everything with one hand) However I tried my best.
2) After that I got an idea to attach a phone wires, that are marked a) orange zebra B) orange c) blue zebra and inserted wires in a male connector and attached connector to the sensor in the sequence way of 1, 2, 3
3) Then I have measured circuits with 20 K Ohms function on my tester. Results were 1+2 = ~450 / 3+2=450
It all looked promissing, but aircon wasnt working anyway.
So what I did next:
4) I have merged 2 and 3 wires. On the AUTO AC ON when merged the fans were engaged for a half of a second. Doing the same with AUTO AC OFF, nothing happened. When switching it ON made the same result. 1+2 didn't make any sense at all.
5) merging 1+3 switched AUTO AC OFF automatically once it was ON.
6) merging 1+2+3 made a specific sound on engine (like a bit of harder job on engine) but without luck to engage the Compressor.

Then I have disconnected the connector and tried to measure with 20 K Ohms function.

7) measuring the 1+3 made the same 450 figure, but all other combinations were negative - zero. So I tried to merge the wires again, and it didn't made any different results. Just same fan ON for a half of a second merging 2+3 - but only on AUTO AC ON mode. On manually AC ON nothing happened.
8) merging wires 1+2+3 made the same sound for load a bit of work on engine, but AirCon pump wasn't engaged.

Also I managed to check the voltage, that goes to relay that engages AIRCON pump, and I had bit different results. When I was merging 2+3 wires the Voltage that comes to relay went from 8.5 V to 13.5V for the same half of a second when the fan starts and stops, but dropped to 10.5 V afterwords and stayed the same.

So looks that something works, but is it a pressure sensor I don't understand :DD

So please tell me if you think that it is the one that I need to change then I'm ready to do it as this part at least is not much expensive comparing with what you can be charged at some garages where they say that they "did work a lot"to sort the problem.

Actually i went for this ****ysis of the problem by my self, as from my 3 previous "engineers" no one could solve this problem and it is ridiculous and I'm tired of all this problem ... :D Probably all of this forum members tired from the same thing.

So I will appreciate if there is a chance to make a diagnosis on my problem.

Have good evening, hope hear from you soon.

Regards,
Arturas
 
Hi Arturas

A rather adventurous and industrious time you have had :)

However, you should never measure resistance looking into wiring that is connected to a system, whether the system is powered or not. Two reasons why
(1) It does not tell one anything
(2) It can potentially damage the system.

When you use a DVM to measure resistance, you are placing a small current across the two points where you are measuring. This current is unlikely to damage anything, but there is the remote possibility that it might damage the electronics that would be trying to “look back” when the system is later working.

Two pictures that I have made to explain ….

Picture4.jpg


Picture5.jpg


You can measure a disconnected sensor, provided the sensor is not an active component.

Most sensors are not active, they are passive. Moreover, the word “sensor” is semantically ambiguous in English. The word “sensor” usually implies a device that measures a continuum, such as temperature. But the sensors that we are talking about here are merely switches.

So the pressure “sensor” on the diesel car will have two switches. I do not know the pressures at which they switch, but for example the low pressure switch may open below 1 bar and close above 1 bar, while the high pressure switch may open above 2 bar and close below 2 bar. If this was the case, I would expect a non-faulty example of this pressure sensor to show both switches closed on the bench, and possibly when it is in the gas system when the system has been dormant and cool for a several hours.

My guide here was initially only intended to help a DIY person to determine whether the compressor was faulty, and also whether the relay that controls the compressor was faulty. I extended the guide for people to also determine whether the pressure switch on the petrol car (my car is petrol) is faulty or not.

Thus for a petrol car, it is possible to test the compressor, the compressor relay, and the pressure switch. If one found all these to be ok on a petrol car, then the problem is likely to be in the wiring, or, in the heater control system behind the dash.

But at this moment in time, for the diesel car, it is possible to test the compressor and the compressor relay, but there is no conclusive test for the pressure “sensor” until someone has measured a known good one on the bench and in the gas system.

As I said “it is safe to measure the continuity of the switches in the diesel car when the aircon and engine have not been run for several hours, without needing the key in the ignition” …..but I would not advise anyone to try to simulate the switching of the switches in the system on the diesel car, nor measure resistance into the wiring on any modern car.

What I would say is this. If you measured all 3 contacts in the “sensor” as open circuit, then in my opinion (and it is only opinion), the sensor is faulty. This is backed up by experience of several people in the forum with diesel cars who have had the sensor changed and their aircon now works.

Hope this helps,
Cheers, Brian :)
 
Dear Brian,

thank you for your help. I will try to arrange a new switch to be changed and let you know if this helps.

Regards,
Arturas
 
Top