As the title says. I take the car for a short drive, temp outside is 10degrees C.

Get home, lock the car up and 5 minutes later the main coolings fans start to run at full speed - sounds like someone is outside with a leaf blower. On a couple of occasions they stop after 10-20 seconds. Other times i have to unlock the car, turn on the ignition which stops the fans.

No codes that look even vaguely related.

Any ideas where to start? I'm thinking to change the coolant temp sensor switch switch initially.

 

Thanks for the response, but in this case it's not normal. The car is not so hot that the fans need to run. It's only started doing this in the last month and i've owned GTC for many years.

Some times it can run for over 5 minutes, it wouldn't even need to run this long if the engine was idling so its definitely not happening by design.

 

Hi - when you say, no codes look even vaguely related....do you have codes, and can you publish them here?

The fan "run on" after ignition off is typically controlled by/through the engine management program (controlled in the ECUs). The ECUs determine fan runtime and speed. Part of the program is a variable time and speed the fans can run after shutdown. They are all driven by the single input ECT sensor. Here is a screenshot of a typical EMS FANS table that manage cooling fans:

I cut part of the screen off in this snip, but the fifth column is the control parameters for cooling fans with ignition off. Again, this is "typical" and not VAG specific. In the fifth column, the first variable specifies the run time, after ignition off the fans can run (in this case up to 300 seconds). The other variable(s) is ECT. In this case, the fans are not ordered on with ign off unless the ECT is 235*F or better and with a fan speed of 60% maximum.

The fact that yours is coming on only 5 mins AFTER ign off, suggests something is corrupt in this command file. The other strange factor, in your case, is that the ambient air temps are only 10*C. This is pretty cool to be holding very high ECTs. But, the strange part for me is the startup/run on a full five minutes AFTER shutdown. This shouldn't be, and no EMS program I know would work that way. Fan run on after ign off are set this way as no battery is going to last too long given that fans can draw 30/40 amps full speed, and why, along with a time element (300s), the max speed of only 60% as 100% would kill the battery in sort order.

 

I want to add, because realize I ignored it, Rico might be right on with the ECT sensor as it could be sending some wildly high temp reading well after shutdown causing the fans to initiate, but the program should NOT allow those fans to run at anytime past a set "ignition off" time. So, crapshoot on just replacing it. I am particularly interested in the addresses 05 and 09 given the condition that an unlock and ign recycle shuts off the fans and resets it all.

That is why I'd like to know if there are any codes sitting in the VCDS run, especially any codes in addresses - 01, 05, 07, 09, 11, 17, 19; but any codes anywhere.

 

So the car overheated today whilst stationary, no fans running at all. I’m thinking a temp sensor is defunct.

Which one provide input input to the ECM for the fan logic? Or should I just change both?

 

Are you capable of doing a comprehensive fault scan (VCDS)? A bad ECT will definitely leave a fault code. Also, did a check engine light come on?

Engine coolant temperatures control numerous functions in the engine management program. Such items as:

1. Idle startup RPM ramp
2. Switchover to closed loop
3. Changeover from CL to Power Enrichment
4. fan control
5. TCM shift verification
6. Engine spark timing management
7. Fueling management

To name only a few. If you have a defunct ECT sensor, in any of its range, or a complete failure, there will definitely be a fault code or multiples. Can you do a scan and post faults? A tool capable of a comprehensive scan such as VCDS is best, but since a bad ECT sensor is so basic to engine management, even a simple store bought emissions scanner (capable of only P-codes) should still reveal faults related to engine coolant temperatures.

This is a good example of throwing parts at a problem. You can replace the sensors, and not solve the problem, albeit you'll have some new sensors.

Fo example, in another thread with a somewhat related "fan control" problem, a failed FCM could be the issue here and replaced ECT sensors wouldn't fix your issue.

 

Yes i’ll do a full scan later. But no Engine Check Light, and everything else ran perfect. Temp gauge was just under 130, so I stopped engine, put ignition on and ran the cabin heaters on max heat for a few minutes to bring temp back down to 90. Then when driving home it maintained that temp.

So i don’t think it’s a case of the car not knowing how hot it is, it’s just the fans are coming on at the wrong times, hence why I wondered if there is a separate sensor for fan management that isn’t used for engine management.

 

130C is definitely overheated and you had a good solution to temporarily getting the engine coolant temps back down. This is hypothetical, perhaps the engine coolant sensor that feeds the ECU is failed at it's high end (so maybe stays at a reading of 90 or 100) while the gauge sensor is reading accurately. I do not know if there is a crosscheck here, but if this is the case, maybe the EMS thinks all is good. That would speak to a failed ECT sensor, but the problem with no code is all guesses.

 

Ok so plugged in VCDS and checked codes in a few modules. Also checked measuring blocks on Engine 01 to see what was going on with temps. Had Engine Coolant & Rad Outlet both over 105 but the fans didn’t run. Also tried output tests to run the fans but despite VCDS saying they were activated, nothing came to life.

Then I found some other measuring blocks relating to temps and whilst majority showed 105/106 there were two other that showed 87 and -48. Where is this data coming from?

 

So, you have a basket full of faults. As I've perused them, you have several "electrical intermittents" along with some specific component failures, a couple of which point to the fans. Then there is one fault I consider a priority. It is the U-code. U-codes are faults in the CANBUS network, either in intermittents or complete failures. If the CANBUS has faults, it could be generating the failures/faults in component specific failures.

When I'm faced with this, the U-Code(s) must be fixed first. Then clear codes, run the car, see what returns. A single U-Code fix sometimes solves everything else.

Some of the codes in your list have a common check/remedy. That is, check the connection integrity to the ECUs. As much as this comes up out of the clear blue sky on this forum, this is the one good example where it is warranted to remove the cowl, see what's going on at the ECUs, the fuse box, the connectors surrounding the ECUs. Visual will tell you a lot. Look for common water damage to the cabling and components including the fuse box.

Once you get past this, especially if you've found something there, correct it, then clear codes, see if the coolant issues automatically correct (correct because the intermittent communications on the CANBUS have been fixed).

After you reestablish integrity on the communications (CANBUS and intermittent electricals), then tackle the fans issue. Codes are directing you to failed fan(s) or/and FCMs. BUT, I would not start here, only end here after the U-code fixes. The worst thing you can do is proceed in all directions at the same time.

Best,

 

The other faults are because of a low battery caused by the earlier issue where fans were running for some time after the car had been locked - that’s why my initial post was about and the car hadn’t over heated at that point. I’ve cleared those and they haven’t returned as the battery is fully charged.

Is is there a way to test the FCMs? I’ve just checked the price of replacements 😥

 

There is a thread on this already. Instead of typing it all over again, this might help you...along with a more economical parts replacement source.

https://www.6speedonline.com/forums/...ans-issue.html

 

BTW, just as a point of mention, in another thread, I was explaining how the fan control module worked, receiving puls signals from the ECU. A look at your first two graphics, when these measuring blocks were recorded, the ECU was commanding a 94.1% duty cycle from both fans and in both cases. If the fans themselves do not run or the FCMs are bad, the ECU doesn't know that, but it continues to signal the fans with a specific duty cycle commensurate with ECTs. In these cases, 105/107C. As the fans continued to fail to respond, the motor eventually overheated. Also, no manufacturer I know of ever commands 100% duty cycle, because of the very high amps these fans use.

Here is a table of a typical EMS the ECU draws from to determine fan speeds within operating range:



When AC is on, and since this taxes engine temps in a different way, the ECU defaults to a fan speed based on AC KPa. This is "off topic" from your issues, but hopefully just a little more background info on what's going on with these fans.

 

Also Mr. BWings has duly warned, past stickies the importance of being grounded when tinkering around the ecu's and electronics. I've always used a non conductive floor mat, wear rubber soled shoes and rubber gloves. However the idea of also having a grounded wrist strap sounds like a sage caveat.

THX

 

Thanks BWings - it's all pointing to faulty FCMs but i'm going to spend some time going from A>B>C to confirm the diagnosis. I'll also supply 12v direct to the fans to ensure the motors are serviceable, but i suspect they will be.

Could i scope the input to the FCM from the ECU to see if indeed the signal is getting to the FCM from the ECU?