Check your radiator/condenser fan operation NOW

[Central IL TDI Driver]

My fans went out earlier this summer. Parked the car to save a bit of $$ (had another ride as it is close to a timing belt). Sure enough, it is the fuse on the battery. Both fans kicked on after pulling it and cleaning the slot and fuse. It looked like it has gotten very hot as it is kind of brownish black up top around the fuse, but it worked with cleaning it. Plan on swapping the whole thing out as it isn't an expensive fix. Now to check the refrigerant level and make sure it is good to go.

 

[VincenzaV]

This thread allowed my AC to come back online (My DS fan was shot)...Pure Gold! Thank you OP!

 

[GWbiker]

2000 Jetta TDI

Don't know if this it the proper place for my post, but please bear with me. I'm looking at a 2000 Jetta TDI (1.9 motor) and am wondering what issues I should look for and if I would run into future problems with the car. Mileage is 150,000. Jetta looks to be in good condition. I have not driven it yet.

I presently own a '15 Gold TDI which I plan to keep and get "Fixed" by VW.

 

[Jlmickle]

I'm having some trouble with the a/c on an 01 mk4 TDI. I've spent a lot of time looking at these forums but this is my first post. I've run through the fan and manual a/c troubleshooting procedures but am a little uncertain if I missed something. For the fan procedure, when I turn the key to on with the cabin fan on and the a/c on my radiator fans do not turn on. I checked all fuses and fuse holders and they were good. When I jumped the thermoswitch per the procedure I was able to get both fans to turn on low and high. I thought it was a thermoswitch issue and I replaced it but that didn't fix the problem.

I moved on to the manual a/c troubleshooting procedure. My multimeter is on its way out and the only functions I can use are voltage and continuity. The golf I am working on is a bit of a project and I replaced the compressor and condenser with used ones before doing any troubleshooting because the condenser was visibly damaged and the compressor pulley bearing was basically gone. I had a shop vacuum test the system and pressurize it so the refrigerant pressure shouldn't be an issue. I ran through the control circuit test procedure as best as I could without being able to test duty cycle or resistance. Everything seemed normal except for the following: DC voltage on the G65 was 1.9V while plugged in which is a little on the low side. The compressor harness has 9.5 VDC when unplugged but drops to zero when plugged in. T14/2 pin showed about 3 volts with outside temp around 50 degrees F (can't recall the exact voltage). I checked the voltage on T14/3 at various engine RPM's and it stayed steady at .87 VDC. I checked for continuity from T14/10 to ground and it was open. I checked for continuity from the compressor harness to T14/10 and had continuity with the pin on the square side of the compressor harness. I checked both pins on the compressor harness for continuity to ground and both were open which I thought was odd. I put 12 volts on the compressor and it engaged immediately.

One thing worth noting is that over the past year I have been restoring this vehicle and I have found a ton of wiring issues so I would not be surprised if wires were crossed somewhere from a repair by the previous owner. I was told he hit a deer on the front left and I think a couple of wires were cut in the process. Any help you can give me would be greatly appreciated. I'm not sure what to take away from some of the tests I did. Still trying to figure out how to read the Bentley wiring diagrams...

 

[DanG144]

It sounds as if you did a good job troubleshooting with the equipment you have, and you did a good write up.

It sounds as if your ground system may be damaged.

Checking at the two wire compressor connection back toward the compressor, neither side should be grounded. Checking back toward the FCM, one wire should go to 14/10 and the other to ground.

I would locate the 4 pin connector and check ground continuity through it. Essentially work your way back from the round side of the compressor connector, until you get to ground. I think the T4 connector is over the starter (going from memory). Pin 4 of T4 should go to ground. This normally happens below the battery tray. Look for small collections of ground wires, in addition to the large main ground for the battery and engine.

If it gets too difficult, simply run a new ground wire to the round side of the compressor connection, and see if that fixes it.

 

[Jlmickle]

Thanks for the quick reply. I tore my wiring harness apart this weekend looking for the cause of the bad ground. There is a six wire crimp connection right above the FCM that connects a bunch of ground wires and that ended up being corroded. I redid the connection and now my ground issue is fixed... unfortunately, the compressor still isn't turning on.

Besides the ground problem, did anything else in my initial troubleshooting results seem off? I couldn't get anything definitive on the signals requiring duty cycle measurements so I'm wondering if one of those could be the issue..

Seems like my G65 could be broken, I could need to redo throttle basic settings, my pressure could be low, or the FCM could be shot. The DC readings I took for the G65 were a little on the low side but still close enough that I'm not sure that is the problem. The pressure shouldn't be low since I just had it leak tested and filled at a shop a few weeks back. Is there another way to know if throttle basic settings need to be changed other than checking duty cycle? Other than that I am almost ready to cross my fingers and order a new FCM.

Thanks again for the help.

 

[DanG144]

Nothing seemed too far off. While checking G65 health with a DC voltmeter is problematic, I also felt the reading was close enough.

The next most likely thing is that the voltage on the power lines supplying fan voltage is dropping too far when the fans try to start. If the voltage sensed by the FCM on the low speed fan power line drops below a certain voltage the FCM will not allow condenser operation.

Since your ground splice was corroded, is it possible your power splices are also corroded? Or the connectors are high resistance? Or the battery boxtop fuse panel is corroded inside? When you pull your 30 amp fuse for the low power fans, what does it look like on the blades? I had to provide an alternate power source for my own fans when the fuse panel had high resistance contacts internally. I used a 40 amp external mounted circuit breaker and fed it off the main battery bus supply on the fuse panel.

There have been only a couple of cases where the ECM was faulty, or the full throttle setting needed redone, in all the years of this thread.

Corroded wires and connections in the snow belt cars is pandemic.

If your belly pan is missing, then the "corroded wires" can also be inside your FCM.

 

[Jlmickle]

It's definitely possible that the power wire connections are corroded. They looked good from the exterior but then again so did the ground splice. I'll probably take the battery box out again this weekend and take a look at the power splices.

I think the vehicle spent most of its life in West Virginia and there are definitely some corrosion issues under the hood probably from years of running without the belly pan. I did replace the fcm about a year ago with a good looking one from the junkyard since when I bought the car the fans were on constantly.

I also replaced the battery top fuse box a year back and the fuse connections still look good. Only problem is the main power wire overheats but I believe that is a separate issue that is covered elsewhere in these forums.

I should have my hands on a Fluke 87 next week sometime and then I'll check the duty cycle measurements. I'll also be able to get better resistance readings for the fan circuits since my multimeter finally died this week.

 

[richmondvatdi]

No AC and no fans (low OR high)

Struggling to get to the bottom of this problem. Here is where I am.
Per the A4 Coolant Fan Testing write-up, I can jumper the fans and get the passenger side fan to run in low speed and high speed. (The driver's side fan is dead and needs repair or replace but that is not the cause of my no-AC). Fuses and their sockets at the top of the battery look as clean as can be.

Ran through the first part of the Air Conditioner Control Circuit Troubleshooting write-up and can report:
*12 volts at all times on the 4-pin connector T4a/1 and T4a/3 are OK.
*Resistance across the compressor is normal (around 4 ohms).
*Putting 12 volts to the compressor produces an audible click of the clutch engaging.
*Fan circuit resistances are low. 3.4 and 3.0 ohms respectively instead of 9.6 and 8.9 for T4a/2 and T4a/4.
*Voltage across the 2-pin connector going to the compressor is low (6.5 volts)

On the 14-pin connector:
*T14/8, T14/9, and T14/4 all have 12+ volts when they should.
*T14/6 ground is good (0.3 ohms)
*T14/14 to T14/5 continuity is good at 1.08 Kohms.

I don't have a duty-cycle meter to test from 11 on (unless I go out and buy one), but can I learn anything from the fact that the fans aren't running when the key is on and the AC button is pressed (with the blower on) at the same time I am not getting sufficient voltage to the compressor? Does this indicate a bad fan control module?

 

[DanG144]

I would check the compressor pressures with a set of gauges, verify that the pressure is at saturation pressure for the ambient temperature. The pressure must be over about 70 psi to reset a low pressure trip at 15 psi.

The fact that your FCM is able to do one of its jobs - run a high speed fan - argues for its good health, but is not conclusive.

The fact that your FCM can shift one fan to fast speed when requested, but is not doing so all the time, indicates that the signal for AC off and fast speed fans from the ECU on T14/13 is not present.

The ECU could still be grounding T14/3 to prevent AC operation. This is not a frequent issue, but has been found.

The G65 could be reporting pressure too low, even if your pressure check shows it is actually high. If the G65 pressure sensor is oily on top near the electrical connections, replace it. DO NOT try to jumper the pressure sensor. Even though VW calls it a switch it is not.

One potential issue, seen fairly often, is that the fans drag the voltage down too low. Low voltage supplied on T4 will interlock the system off. You have proven that it is ok with the fans not running (T4a unplugged), but you don't know what it is with the fans powered on, potentially dragging the voltage down. I would unplug the fans, turn on the engine and AC, and see if the compressor engaged. If it does, do not run it long or it will overheat and overpressure your AC system. if the compressor engages during this test, look for what is dragging your voltage down - usually a high resistance connection.

 

[richmondvatdi]

I would check the compressor pressures with a set of gauges, verify that the pressure is at saturation pressure for the ambient temperature. The pressure must be over about 70 psi to reset a low suction pressure trip at 15 psi.
The fact that your FCM is able to do one of its jobs - run a high speed fan - argues for its good health, but is not conclusive.
The fact that your FCM can shift one fan to fast speed when requested, but is not doing so all the time, indicates that the signal for AC off and fast speed fans from the ECU on T14/13 is not present.
The ECU could still be grounding T14/3 to prevent AC operation. This is not a frequent issue, but has been found.
The G65 could be reporting pressure too low, even if your pressure check shows it is actually high. If the G65 pressure sensor is oily on top near the electrical connections, replace it. DO NOT try to jumper the pressure sensor. Even though VW calls it a switch it is not.
One potential issue, seen fairly often, is that the fans drag the voltage down too low. Low voltage supplied on T4 will interlock the system off. You have proven that it is ok with the fans not running (T4a unplugged), but you don't know what it is with the fans powered on, potentially dragging the voltage down. I would unplug the fans, turn on the engine and AC, and see if the compressor engaged. If it does, do not run it long or it will overheat and overpressure your AC system. if the compressor engages during this test, look for what is dragging your voltage down - usually a high resistance connection.

I am borrowing a set of ac pressure gauges and will check. If pressure is good, can I test the pressure sensor for continuity? I just ordered a new one because it did seem shiny/oily when I was looking at it last night, but would be curious to know if it is the root cause. I will also try unplugging the fans and running that test to see if the fans are causing a big enough voltage drop. I'm thinking that is unlikely since when I tested resistance across the fans last night it actually seemed too low instead of too high (3.4 and 3.0 ohms).

 

[DanG144]

I do not know of any continuity test (or other electronic test) for the pressure sensor, except to compare it's duty cycle output signal to actual pressure.

Usually if it is leaking oil it shows a duty cycle output much lower than it should have for a given pressure.

My fan ohm readings have never really made sense to me. That is the danger of using only one set of data...

 

[richmondvatdi]

Still trying to meet up with the guy with the pressure gauges so have not been able to measure pressure. I did disconnect both fans to see if that might allow the compressor to kick in. Answer is no. Still no compressor, even without the fans.

Ordered a bunch of parts from idparts that will arrive tomorrow including a pressure sensor (which is oily and will definitely be replaced) and a drivers side fan (which is dead and will be replaced). I also got an ambient air sensor (cheap) and a FCM (not so cheap) just in case.

 

[DanG144]

The ambient air temperature sensor resistance you got was good, so I would do that absolutely last. The FCM next to last. G65 pressure sensor first. The fan just needs to be done for sure before city operation.

At this point I would check the pressures after replacing the pressure sensor, though you should not lose but a tiny puff of refrigerant as you screw off your old one, as long as the schrader valve works.

Once you get the control system working so the clutch engages, hope you still have a good Refrigerant Control Valve, which is the internal regulator that adjusts the displacement of the compressor. As they age, they often slowly fail such that the compressor can never get to full load. The symptoms are low or low-normal discharge pressure and high suction pressure. There are other threads on how to test and replace the RCV, so let's not turn this one from its purpose - troubleshooting electrical control issues.

 

[scottd]

At 153K on my 2005 Jetta, Both fans turned on.

 

[richmondvatdi]

The ambient air temperature sensor resistance you got was good, so I would do that absolutely last. The FCM next to last. G65 pressure sensor first. The fan just needs to be done for sure before city operation.

At this point I would check the pressures after replacing the pressure sensor, though you should not lose but a tiny puff of refrigerant as you screw off your old one, as long as the schrader valve works.

Once you get the control system working so the clutch engages, hope you still have a good Refrigerant Control Valve, which is the internal regulator that adjusts the displacement of the compressor. As they age, they often slowly fail such that the compressor can never get to full load. The symptoms are low or low-normal discharge pressure and high suction pressure. There are other threads on how to test and replace the RCV, so let's not turn this one from its purpose - troubleshooting electrical control issues.

I think I am chasing more than one issue right now, as well as dealing with some frustrating, self-inflicted wounds. I wanted to change the pressure sensor last night after work but did not have a 17mm crows foot wrench and NAPA was closed. So I tried grabbing it with a large pair of vice grips. No luck. What I did succeed in doing was to create a small leak in the top of the sensor. I could hear a small hiss and see a tiny bit of gas escaping from what was likely a small crack in the sensor plastic near the pins..

Today I got the wrench and replaced the sensor, but by then I am sure all the refrigerant was gone. I got a vacuum pump and manifold gauge and ran the vacuum pump for a couple hours. I closed the valves, switched off the vacuum pump, and waited. Within 10 minutes or so, vacuum had dropped from 30" to 24" and continued to drop. I pulled a vacuum again and once again lost vacuum.

Now I am trying to chase a leak (I think). I bought some dye and started to add R-134a in an effort to get the compressor to kick in, but it won't and pressure is up to the high end of the band for the ambient temperature, at least according to the gauge on my AC Pro. It only took a fraction of a 340g can (maybe 200g) and the system takes 750g when empty. If I can't get the compressor to kick and and circulate the dye, I won't be able to find the leak.

What next?

Edit/Update: I realized that the ranges on the AC Pro dial are for the low pressure side of a running system. The equilibrium pressure of the system with the compressor off will, of course, be higher. So I went back and slowly added the rest of the can of refrigerant. The pressure is now just a bit above 80 psig but still no compressor start. I read on another post that G65 will trip at a low pressure of 14.5 psig and reset at 68 psig, so with a brand new G65 and 80 psig of refrigerant, what could be causing this?

 

[richmondvatdi]

UPDATE: Fixed

So this morning, I concluded that based on the steps I had taken so far, there were very few possible issues that would keep both the low speed fans and the compressor from working, so I bit the bullet.

I removed the old fan control module and plugged the new one back in. Started the car, and the instant I pressed the AC button on the dash, the compressor clicked in and the low speed fans began turning. Success!

I am disappointed that it was the $159 part that fixed it instead of the $49 part, but thrilled that I was able to do it myself with your help. Thanks Dan.

As far as the possible leak, that may not be an issue. I think that the leak may be in the manifold pressure gauges I borrowed. They are a cheap Harbor Freight set and I noticed this morning that one of the fittings was loose (not even finger tight). I checked pressures a few minutes ago and will check them again in a few days. I suspect they will be OK.

Thanks again. It will be nice to be able to turn on the AC when the temperature tops 90 this coming Friday.

 

[DanG144]

Glad you got it fixed.

 

[ejallison1]

What New Radiator Fan to Buy

2001 NB

Need to know which brand of radiator fan gives longest service. Dorman/VDO/TYC/?

Old ones will not run on low speed when connected to 12v directly as per post #1. High speed works fine. Opened up and all looks good-brush holders not burned-exercised the brushes(4 on each fan)-still no joy, resistor for slow speed fried? I think they are original.

IDParts have both fans made by TYC for $120. Dorman from review of this thread are crap. Any input greatly appreciated-don't want to do this again.

Thanks

 

[Genesis]

Yes, it's the resistor. That's what blows up on them.

I went through two Dorman ones in 3 weeks but the one in there now has been there (driver side) for close to four years and is still fine. The pax side is OE and never did fail. Autozone has 'em around here (in the full shroud; dismount the spider and just replace that it's a hell of a lot easier!)

 

[ejallison1]

Decided to go with IDParts. Hopefully they have done the research and their TYC brand will hold up. Thanks.

 

[DanG144]

Yours probably have an open in the low speed resistor circuit.

i wish I knew the best brand to buy for longevity, like others I know it is not Dorman.

If you get some OEM fans with metal brush holders and bearings instead of bushings, they are by far the best fans I have seen - once you have mounted an external low speed resistor, or a Pulse Width Modulation controller on them to give you a reliable slow speed. On these fans it was almost impossible to keep good connections to the low speed resistor, which is why you could use an external resistor and cure the problem.

I would love to know more about brushless fans and their control circuits - that seems very promising.

 

[Genesis]

Brushless DC motors are a permanent magnet on the rotor and coils on the periphery. In order to run at all they require a modulated signal that switches current to the coils (and possibly modulates it) in sequence. Practical examples also have an optical encoder on the shaft so the controller knows where the rotor is.

The nice thing about them is that since they're brushless other than shaft bearings there's nothing to wear. They're inherently ignition protected (doesn't matter in this application, but can in others.) They can run MUCH more efficiently since the drive current to the coil can be "shaped" to match the flux of the permanent magnet on the rotor; you can get into the high 80s efficiency wise pretty easily, and 90+ is achievable in some circumstances. This means they run a LOT cooler as well.

So instead of dissipating the extra power for "low" the motor can instead run much cooler and slower, pulling far less current in the first place. It would be expected to last a LOT longer too. But.... they're more expensive because you need the controller circuit (and the electronics are not free.)

 

[ejallison1]

AC System for A3 and B4 Similar to A4

Wanted to thank the major contributors of this thread, especially Dan on explaining the AC systems on the A4's. Have an A3 and B4 which will need AC work as well once finished with the bug. Researching on A3 forum found they have the same Sanden compressor. Are they pretty similar systems? The A3's and B4's have an actual switch vs. the sensor. They both have a FCM and the T4 plug-not sure about the the T14plug. Both have two speed fan's. Are they identical enough can use the Post #1 diagnostic steps to get it back running. Realize the fuses etc. may not be identical.

 

[NCedwardm]

testing duty cycle on G65

First of all thanks to Dan and all contributors for this excellent thread and the documents on testing the fans and whole AC system! I am testing the white wire that goes to the G65 sensor. I am just using a DVM to get some idea since I don't have another way to test duty cycle. BTW, is there a way to use an olde timey dwell meter for that or is this much lower voltage/current than a points ignition system?

So I used a safety pin to pierce the insulation on white wire and am testing that on positive of my meter and negative I'm connecting to engine ground. is this correct? Should this be done with the engine running and AC switched on?

Thanks!

 

[DanG144]

Wanted to thank the major contributors of this thread, especially Dan on explaining the AC systems on the A4's. Have an A3 and B4 which will need AC work as well once finished with the bug. Researching on A3 forum found they have the same Sanden compressor. Are they pretty similar systems? The A3's and B4's have an actual switch vs. the sensor. They both have a FCM and the T4 plug-not sure about the the T14plug. Both have two speed fan's. Are they identical enough can use the Post #1 diagnostic steps to get it back running. Realize the fuses etc. may not be identical.

I know very little about the A3 and B4 systems, sorry. Maybe someone else can chime in; I don't even have a manual to reference.

 

[DanG144]

First of all thanks to Dan and all contributors for this excellent thread and the documents on testing the fans and whole AC system! I am testing the white wire that goes to the G65 sensor. I am just using a DVM to get some idea since I don't have another way to test duty cycle. BTW, is there a way to use an olde timey dwell meter for that or is this much lower voltage/current than a points ignition system?

So I used a safety pin to pierce the insulation on white wire and am testing that on positive of my meter and negative I'm connecting to engine ground. is this correct? Should this be done with the engine running and AC switched on?

Thanks!

You look for signal on the white wire with the AC on. I don't think the engine has to be on.

A dwell meter is an interesting thought, there may be some useful knowledge gained by trying one - but you have to have a known good system to compare to.

 

[SilverGhost]

G65 I believe is sending a square wave signal back to FCM. You can read actual pressure on some cars from this sensor. Not sure if later models even just sent a BUSS signal (networked sensor). Diagnosis procedure setup is pretty involved, to the point we keep one we know works and just switch them to test.

Later A3 have a variable compressor(mechanically) and clutch like the A4, but earlier A3 cycled the clutch like Asian cars do. Not sure about B4. Both A3 and B4, that I have seen) only have one radiator fan motor and connect in the shroud with a belt. A3 and part way through A4 model runs they used AC pressure switch - 4 wire high and low threshold. Later in A4 model run they switched to 3 wire pressure sensor.

Jason

 

[mattg]

My system won't hold vacuum. What's the best way to find the leak? I think it is coming from near the dryer.

 

[SilverGhost]

Honestly - if its not gross leak then add dye and charge the system. Let it run for a while then use the black light and glasses to find the leak. Sometimes you get lucky and can find a wet spot or black light it and find a glow, because of prior service added dye.

But if the system is already empty (you said won't hold vacuum), then I would check the o-rings and probably just replace them.

Jason