Hey guys,

So I just picked up a 2010x sedan for cheap with 172k kms. The rad fans were always on full blast no matter what.

Even with the key on, engine off on cold start, the fans will slowly kick up to full speed after about 10 seconds.

I replaced the fan module with one from a junkyard. No change.

I hooked up my scan tool to check coolant temps. They were accurate. Started at 20 celius, and worked it's way up to operating temp as the engine warmed up.

I then junked the fan module, went back to the junkyard and got another fan module from another g37. The fans now don't go to full blast, but are still always running at a medium speed when cold start even.

I then left the car idling for 45 minutes. Coolant gauge remained dead center and the fans were at medium speed for 85% of the time. After about 35 minutes, the fans picked up speed for a short time and slowed back down again.


Any ideas? Have I just been getting bad fan modules from the junkyard?

What else can I check? Also, Im thinking my next step is ordering a new module. I went to the junkyard for the convenice factor. I may just place an order for a new one, but I'd like to know the diag procedure for the other variables in the system before I go for it.

Any help is appreciated. Thanks!

 

Fan control modules rarely go bad. Let alone 2 or 3 in a row.

Typically, when the fans come on as soon as you press the START button, this indicates the car is in a "fail-safe" mode:


The question then becomes why is the car in fail-safe? You can try running the IPDM/er self test procedure though if the car is indeed in fail-safe the results might not be conclusive. But this will test the operation of the IPDM/er (post ECU), the PWM relay, the FCM (which you've replaced already) and the individual fan motors. If the test fails to initiate then, indeed, the car is in fail-safe.

Have you checked for any water leaks in/near/on the BCM (passenger-side footwell)? The BCM and ECU are in the same neighborhood and thus prone to the infamous water leaks these cars are destined for. With the battery disconnected I would unplug the IPDM/er, BCM and ECU to ensure the terminals and pins are clean and free from dirt or corrosion.

Though it seems your temperature test shows the coolant temp sensor is working, these are becoming more "unreliable" as these cars age. Something to consider as well.

 

Thanks for the reply!

So I did the ipdm self test. The fans were on throughout the entire test at a medium speed. When the test got past the wipers, lights, and AC clutch, the fans sped up to full speed and slowed back down to medium and the test repeated itself with the same symptoms.

So, I hooked up my scan tool and checked for errors codes in the ECM, bcm, ipdm etc. no codes at all in the car.

I checked data stream for the 'rad fan req' which shows a percentage (I assume at what percent capacity the fan is running). Even though the fans are running at mid speed, this test showed 0%

I then put data stream to coolant temps sensor and went for a 15 minute drive. Coolant temps got to a max of 91 Celsius which is about 195 Celsius. I'm sitting at idle as I type this out after my drive, it still shows 91 Celsius.

Any ideas? I'm still thinking I could just have bad fan control modules even from the junkyard as they look physically worse than mine. Also, the fact that the first two ran at max speed and now this one is running at a mix of mid and max.

Edit:Also the passenger foot well is completely dry and I plan to update the sunroof drain part as well if not already done.

 

I seriously doubt it is the FCM as the fans change speeds when commanded by the IPDM/er self-test. During the test, the fans will run at 50% duty for 5sec. then 100% duty for 5sec. Based on this alone I do not suspect the FCM, however, you must follow your gut.

You should be reading a numerical value here.

You could check the PWM relay though with the fans running you can likely rule that out entirely. Just tossing ideas...

Something is causing the fans to turn on in fail-safe when in KOEO state. The way I understand it, if the FCM fails the fans just stop working as the FCM is what supplies voltage to the fans. The FCM receives the PWM signal (speed) from the IPDM/er directly (as commanded by the ECU via CANBUS).

Maybe others will have more ideas...

 

Yeah it's a strange one for sure. The fan speed req shows 0% and as I said above the fans were running at medium speed throughout the entire duration of the test.

And the coolant temps sensor is showing accurate temps which is weird.

Im not sure if it's a wiring short, as on this fan module, it doesn't run at 100% capacity anymore. If I take my old fan module and install, the fan runs at 100% full speed which is confusing me.

Maybe a new fan module will fix it. Unless something else is wrong. Ill wait for others to chime in as it's not really affecting the car.

I have to make a 800km trip on Friday , figured I'd try and get it fixed before going.

Also, every thread I have read so far here, on Reddit, on 370 forums, g35 forums and q50 forums have fixed their key on, engine off, fan running issues with a new fcm.

 

An update:

I went out for some ice-cream with my wife for about a 35 minute ride. When I got home and parked, I didn't hear the fans running. Popped the hood and fans were off.

Cycled the AC on, fans turned on as normal. Turned AC off. Fans back off again.

I will update with what I find over the next few days on what it's doing.

 

Hey guys,

Started car again cold this morning. Fans were at full blast.

I'm wondering if I have a wiring problem or something. Are there instructions anywhere on tracing the wires or checking for shorts?

Also, is there any way to check if my temp sensor is bad before replacing it?

 

If you haven't already, go to NICOClub, navigate to the year/ type of your car, and download the Factory Shop Manual.

For this issue you will need to download sections EC (Engine Control System) and probably sections PCS (Power Control System) and PG (Power & Ground) though I would suggest you take the time and download each section.

The FSM has all the info you need to work on these cars, including how test the ECT sensor using hot water and a multimeter.

 

Thanks I'll look into this

 

Check the connections to the FCM. I changed mine last year and it turns out the connection was dirty so I blew it out with compressed air and wrapped the connections with some electrical tape to secure it. Works like a charm.

 

Hey guys,

So I inspected a few things.

The connections look good. I tried jiggling them, pushing them in, pulling them out a bit, no change. Rad fans still stuck on high when key is in the on position.

Took out the plug for the coolant temps sensor, no corrosion or anything I can see. Reinstalled with some dielectric grease.

I removed the ipdm looking for a 'relay' for the fan. No such thing on the ipdm. Is the fan control module the only thing in the system and there is no mechanical relay? Correct?

I removed the battery as I saw some wetness down there. Full of leaves and debris. Shop vac'd it all out and cleared out the little hole under the harness which I assume is a drain.

I reinstalled everything and started the car again. Rad fans at full blast.

I connected my scan tool and checked rad fan req percentage. It was still at 0% even with the rad fans at full blast.

I then turned on my AC, rad fan req changed to 70%. Turned AC off, rad fan req went back to 0%.

I think I'm going to order a fan control module off Amazon and see if it makes a difference. If not, we know it's something else.

I don't suspect the temp sensor either as I don't have a code and its showing accurate temps on my scan tool, and gauge.

Any suggestions?

Thanks

 

As indicated before, the fact that you are reading "0% Duty" indicates the ECU is in a "Fail Safe" condition. When you turn the AC on you are reading 70% duty but does fan speed actually change? My point being, the ECU might be calling for 70% duty but is the ECU telling the IPDM/er to send the same 70% duty signal to the FCM? Or is it the 0% duty. This is where a potential CANBUS error comes into play. Testing the PWM circuit might help narrow things down.

The fans use a external, mechanical relay for the PWM (Pulse Width Modulation) circuit. Aside from that relay, everything else is "solid state" and incorporated within the IPDM/er and is not user-serviceable. You can try swapping the PWM relay with the VVEL relay (they are identical) and see if the problem persists.

I found this thread over at NicoClub which has a better explanation of, and some diagnostic guides for, the PWM circuit.

Everything online points to, and everyone screams, "it is the FCM!" It could be. BUT, I know you tried several "junkyard" donors and the fact that all of them still yielded this issue tells me your issue is upstream (IPDM/er, ECU, ECT, etc.). I can see one being bad but not two or more random FCM's. Sorry, never was a fan of coincidences.

 

Thanks!

Yeah it's a weird issue for sure.

Any idea of the location of this 'relay'.

Also, are there any instructions on testing these other components.

I.e., a guide which will explain the values I'm searching for and what pins etc to inspect/check to rule things out.

 

My bad, I though you knew where the relay was. There is a small relay box along the right fender edge next to the IPDM/er. There should be 2 relays: VVEL and PWM.


The factory manual has all of the reference values and protocols to test and diagnose. There are no other "quick guides" that I am aware of.

The post in the NicoClub thread shows what the voltage values should be at the PWM signal wire on the FCM harness (0-5V+).

 

Thanks. Ill try and do some testing when I get some time and post back my results.

Thanks!