How to test the electro-viscous fan clutch

Mooseman

Original poster
Moderator
Dec 4, 2011
25,257
Ottawa, ON
So you suspect that your fan clutch is defective. One of two conditions is possible.

If your fan is roaring like a jet engine after it has warmed up, DO NOT TEST. It's already confirmed that the clutch is locked up and needs to be replaced. Normally, it will roar a little on startup and should quiet down as it warms up. If it is locked up, it's not bad to run it like this but power and fuel consumption will suffer and you may get a CEL with a code about the fan speed.

The other possible condition is that the fan is not working enough. The engine may or may not be getting overheated but a lot of times, A/C performance will suffer while idling but will be better while moving. You may or may not be getting a CEL with a fan code. It might not be overheating because the residual friction is spinning the fan just enough to pull a little bit of air. Here's how to test for this condition.

  1. Warm up the engine completely

  2. Turn on the A/C
  3. Try to stop the fan using a rag like in the video below
  4. Hold it for a little bit. You should feel some tugging from the fan trying to get it going again
  5. If you are getting that tugging, the clutch is working. If not, it's defective.
  6. If the temperature gauge is starting to climb and the fan is still not working, let go of the fan and turn on the heater to cool the engine
If it's winter or your A/C doesn't work (i.e. it's empty), use this alternative method:

  1. Turn off the heater and warm up the engine completely
  2. Stop the fan using a rag like in the video below
  3. Because the engine is at normal operating temperature, the PCM will not command the fan to run right away. Keep holding it while the engine gets hotter
  4. As it gets hotter, you should start feeling some tugging from the fan trying to get it going
  5. If you are getting that tugging, the clutch is working. If not, it's defective.
  6. If the temperature gauge is starting to climb and the fan is still not working, let go of the fan and turn on the heater to cool the engine



An extra test to do while holding the fan, try to push the fan edge fore and aft. If it moves more than about 1/8", the bearings inside the clutch are failing. It might also make you think that the clutch is working because of this wobble and failing bearings due to extra friction.

If your engine is not overheating, you can continue to run it like this until a replacement clutch can be ordered and installed. You might get a CEL but keep an eye on your engine temperature, especially if the weather is warm/hot and avoid using the A/C. Hayden is the preferred brand here while Dorman have been known to fail prematurely, despite their lifetime warranty.

Edit: Another test you could add to this would be for extra drag while not necessarily locked up. This could affect your MPG if it's dragging more than necessary. Unplug the clutch and let it run a while to warm it up. It should freewheel like or close to what's shown in the video. If it has significant drag when trying to stop it with a rag, it may be on its way out and heading to lockup.

As for brands, it has been found that the ACDelco/Behr/Mahle is the superior clutch running as designed.
EV Fan Clutch Recommendations for proper operation
 
Last edited:

littleblazer

Member
Jul 6, 2014
9,265
Hmmmmm..... that truck looks awfully familiar. :biggrin: Nice write up. :thumbsup:
 
  • Like
Reactions: mrrsm and Beacon

Mooseman

Original poster
Moderator
Dec 4, 2011
25,257
Ottawa, ON
Hope you don't mind me referencing your video? I keep using it as an example on how to test the fan so thought it would be prefect for this article.
 

littleblazer

Member
Jul 6, 2014
9,265
Not at all, I'm glad its useful.
 
  • Like
Reactions: mrrsm

Mooseman

Original poster
Moderator
Dec 4, 2011
25,257
Ottawa, ON
I just want to add another thing to check before declaring the clutch dead. Swap the fan relay with another similar one in the fuse box (headlight IIRC) and see if the problem persists. One last thing, check the wiring, especially at the clutch connector to ensure there are no broken wires.
 

swede

Member
Jun 22, 2014
1,576

Mooseman

Original poster
Moderator
Dec 4, 2011
25,257
Ottawa, ON
You could put +12V to the white wire, which will effectively put the fan to 100% engagement but my method will also show if it works at partial engagement. The PCM never puts it to 100%, 90% is the highest and that's at very high engine temps. It may work at 100% but maybe not at 20-30%.

Of course the absolute best way to test is with a Tech 2.
 

webdawg

Member
Jun 26, 2014
247
Okay, so in your other post to me about testing it you say that the one in this video is bad. The one in this video is bad? I am going to do a re test of mine, I think it is bad and I want to be sure.
 

Mooseman

Original poster
Moderator
Dec 4, 2011
25,257
Ottawa, ON
Yes, the one in this video is bad. All it does is freewheel and has no resistance.
 

Mooseman

Original poster
Moderator
Dec 4, 2011
25,257
Ottawa, ON
I found this video by Scanner Danner where he's trying to diagnose a fan code. Shows the signals to and from the the fan and how the fan operates when commanded by the PCM and the scanner. Very interesting info and could be useful to anyone that has a scope and/or high end scanner or Tech 2.

A bit long but very informational.

 

Mooseman

Original poster
Moderator
Dec 4, 2011
25,257
Ottawa, ON
Very good info on the function of the EV Clutch however, the quick function test requires a Tech 2. The info I posted regarding a locked clutch (beyond a few minutes) is still accurate.
 
  • Like
Reactions: mrrsm

Thunderwing

Member
Jun 18, 2017
4
Sardis, BC
Very good info on the function of the EV Clutch however, the quick function test requires a Tech 2. The info I posted regarding a locked clutch (beyond a few minutes) is still accurate.
I think you missed the main point. (of the Tech Link, which is that a scan tool is not required if you read on)

If the goal is to determine the root cause of the issue(faulty fan clutch or the control circuit):
-unplug the fan clutch connector
-run the engine at 2000rpm for 2 minutes

The Fan Should not be roaring/fully engaged

The fan clutch should be disengaged with no voltage applied.
This applies to the complaint of 'Fan Running All of the Time'.

you said 'DO NOT TEST'.

You could have an issue with the control circuit.
It's always good to properly test and confirm a failed component before guessing and throwing parts at a complaint.

The key information in the Gm Tech Link Bulletin is that the Fan is controlled 'ON' by an electrical signal, therefore unplugging it, the fan should be 'OFF'.
 
Last edited:

littleblazer

Member
Jul 6, 2014
9,265
I think you missed the main point.

If the goal is to determine the root cause of the issue(faulty fan clutch or the control circuit):
-unplug the fan clutch connector
-run the engine at 2000rpm for 2 minutes

The Fan Should not be roaring/fully engaged

The fan clutch should be disengaged with no voltage applied.
This applies to the complaint of 'Fan Running All of the Time'.

you said 'DO NOT TEST'.

You could have an issue with the control circuit.
It's always good to properly test and confirm a failed component before guessing and throwing parts at a complaint.

The key information in the Gm Tech Link Bulletin is that the Fan is controlled 'ON' by an electrical signal, therefore unplugging it, the fan should be 'OFF'.
Short of a relay sticking, I see no possible cause that it would be electrical. The pwm feed is positive so if that breaks it defaults to unlocked. If the ground breaks it's the same. If you end up in full lock, the thing is somehow getting a constant positive, the pcm should see that and set a light because its command and actual are different.

And checking the relay is addressed in post 7...
 

Mooseman

Original poster
Moderator
Dec 4, 2011
25,257
Ottawa, ON
When I said DO NOT TEST, that's with the method described with the rag. As you could imagine, trying to stop a locked fan with a rag would not be a pleasant experience.

And pulling the relay would have the same effect as unplugging it. And I also mention checking the wiring which is a common problem.
 

Thunderwing

Member
Jun 18, 2017
4
Sardis, BC
Short of a relay sticking, I see no possible cause that it would be electrical. The pwm feed is positive so if that breaks it defaults to unlocked. If the ground breaks it's the same. If you end up in full lock, the thing is somehow getting a constant positive, the pcm should see that and set a light because its command and actual are different.

And checking the relay is addressed in post 7...

Electrical faults don't have to be wiring, they can be control module input/output related including:
a faulty sensor(input) giving the control module information that would make it command the fan on(output)
-wiring to one of the sensors(inputs) that as a result, affect the output of the control module
-a faulty control module

Ok, here is an example of how it could be electrical...

Complaint: 'Fan Running All of the Time'
Confirmed complaint of the fan running all of the time
Disconnected the Fan Clutch Connector at the Fan Shroud
Ran Engine at 2000rpm for 2 minutes=fan disengaged (ok)

The fault is not with the Fan Clutch as it disengaged like it is supposed to with its' connector disconnected.

The fault is within the control circuit.

Several Reasons that the fan could be commanded on including:
-A/C Command
-Coolant Temp Command
-Gearbox Oil Temp Command
-PCM Faulty
-Short-to-power
-Are there any active DTCs?

The Main point is that you want to quickly do these 2 things:
-Confirm The Complaint(and obtain as many details about the operating conditions during the time of the fault)
-Determine if the fault is mechanical or electrical(and then follow the appropriate troubleshooting steps)

In Review: The complaint is an issue with the fan clutch running all of the time(overcooling the engine cooling package). By starting diagnostics at the place in the fan clutch system where the mechanical and electrical components separate(the fan clutch connector), you successfully start the troubleshooting process in the right direction.
 

djthumper

Administrator
Nov 20, 2011
14,950
North Las Vegas
You do realize that you are posting things that don't work for the everyday person trying to figure out his problem. We don't all go to a dealer to get work done right away. You are also talking to people that have been doing this for a while on this platform. You should probably slow your role a little and do more reading.

I deleted all of your cross posting and starting of a new thread on this very thing. I understand you are the new kick on the forum and you think you have something to prove. So far you have proved that you just want to get into the conversations and not listen. You didn't even start with an intro thread to introduce yourself.
 

Thunderwing

Member
Jun 18, 2017
4
Sardis, BC
You do realize that you are posting things that don't work for the everyday person trying to figure out his problem. We don't all go to a dealer to get work done right away. You are also talking to people that have been doing this for a while on this platform. You should probably slow your role a little and do more reading.

I deleted all of your cross posting and starting of a new thread on this very thing. I understand you are the new kick on the forum and you think you have something to prove. So far you have proved that you just want to get into the conversations and not listen. You didn't even start with an intro thread to introduce yourself.

I don't know what you're talking about -cross-posting and starting a new thread-
I was just trying to add some info that was missing to diag a fan running all of the time.
I didn't know I was supposed to introduce myself. I don't want to prove anything, just trying to be helpful and post that Tech link bulletin on this that I came across.
 

littleblazer

Member
Jul 6, 2014
9,265
Electrical faults don't have to be wiring, they can be control module input/output related including:
a faulty sensor(input) giving the control module information that would make it command the fan on(output)
-wiring to one of the sensors(inputs) that as a result, affect the output of the control module
-a faulty control module

Ok, here is an example of how it could be electrical...

Complaint: 'Fan Running All of the Time'
Confirmed complaint of the fan running all of the time
Disconnected the Fan Clutch Connector at the Fan Shroud
Ran Engine at 2000rpm for 2 minutes=fan disengaged (ok)

The fault is not with the Fan Clutch as it disengaged like it is supposed to with its' connector disconnected.

The fault is within the control circuit.

Several Reasons that the fan could be commanded on including:
-A/C Command
-Coolant Temp Command
-Gearbox Oil Temp Command
-PCM Faulty
-Short-to-power
-Are there any active DTCs?

The Main point is that you want to quickly do these 2 things:
-Confirm The Complaint(and obtain as many details about the operating conditions during the time of the fault)
-Determine if the fault is mechanical or electrical(and then follow the appropriate troubleshooting steps)

In Review: The complaint is an issue with the fan clutch running all of the time(overcooling the engine cooling package). By starting diagnostics at the place in the fan clutch system where the mechanical and electrical components separate(the fan clutch connector), you successfully start the troubleshooting process in the right direction.
I get what you're saying, but the way the circuit is laid out, you would only get a short to power via a stuck relay. Any other command is pwm unless you are running insanely high temps. And at that point you should be more worried about destroyed components. And in the case of a faulty pcm, they brick themselves, ask me how I know...

I get the whole test it theory, but when there was a design revision preventing a lockup failure mode, you'd tend to think that is going to be (and generally is) the problem. Every electrical issue I've encountered with it is usually no feed and not the other way around. Stranger things have happened though.
 
  • Like
Reactions: G09MVOYC

mitchr316

Member
Apr 28, 2019
4
Yakima, WA
Mooseman, post was a long time ago and not sure this is even active! I just joined today and been digging into whether I should replace EV on 2004 Buick Rainier 5.3. AC little warm
At idle and concerned temp climbed to 220+ a few time la pulling hills last summer when not towing. Bought a new 3k RV and picking up Saturday and want to run cool. Replaced pump and stat last year cuz dripping, assumed weep hole. Stopped dropping but a little seepage still somewhere. AC problem and heating up last year was with coolant topped off. Most of tech stuff in thread over head, but tried your fan test with rag. Seemed like stopped pretty easy other ac on and at 210 but then more forceful after tried again. What should feel like if really working good? Thank you.
 

littleblazer

Member
Jul 6, 2014
9,265
Mooseman, post was a long time ago and not sure this is even active! I just joined today and been digging into whether I should replace EV on 2004 Buick Rainier 5.3. AC little warm
At idle and concerned temp climbed to 220+ a few time la pulling hills last summer when not towing. Bought a new 3k RV and picking up Saturday and want to run cool. Replaced pump and stat last year cuz dripping, assumed weep hole. Stopped dropping but a little seepage still somewhere. AC problem and heating up last year was with coolant topped off. Most of tech stuff in thread over head, but tried your fan test with rag. Seemed like stopped pretty easy other ac on and at 210 but then more forceful after tried again. What should feel like if really working good? Thank you.
Iirc if it is working properly it should try to pull out of your hand. It should spin back up instantly after you let go too.
 

mitchr316

Member
Apr 28, 2019
4
Yakima, WA
Thank u LB. Will go feel it again. I think it’s doing what u r saying, just hard to tell how hard it should be trying to pull out of hand. Appreciate the help.
 

mitchr316

Member
Apr 28, 2019
4
Yakima, WA
Once stopped I can hold with light pressure of single finger or thumbpad, and after revving at 2k RPM same result to stop and hold. Stopped with rag pushed in for a few minutes but couldn’t see any temp rise, with air on, but it’s 55F here.
 

mrrsm

Lifetime VIP Donor
Supporting Donor
Member
Oct 22, 2015
7,639
Tampa Bay Area
Time for a Radiator-Heater Core Flush? ...and... A Transmission Fluid & Cooler Lines Flush, too? Perhaps the Heat Soak cannot get carried away because things may have gotten a bit "sluggish" inside of the radiating units?

Transmission Flush:

Radiator- Heater Core Flush:
 

Mooseman

Original poster
Moderator
Dec 4, 2011
25,257
Ottawa, ON
Try it again but pull the fan relay. If it feels the same, then I would assume the clutch is done. You could also put 12v+ to the white wire at the clutch (or at the relay if you can figure out which terminal it is). This should lock the clutch and pull air like a dump truck. If it doesn't change, it's toast.
 
  • Like
Reactions: littleblazer

4girlsintow

Member
Jan 28, 2019
5
CT
I have a 2003 EXT 4WD Trailblazer 235,000 miles. May daughters use it now. Engine still runs great! The engine temp reads a little low all the time, about 2-3 notches below 210. Now somethimes it only gets to halfway to 210, even in the summer. No CEL. I've been trying to read as many of these waterpump/fan post as possible. My fan seems to run fast all the time but not loud like the jet engine i've read about. Even if I pull the relay or unplug the fan it still runs fast. Too fast to try and slow it with a rag. Has the waterpump bearing seized around the fan shaft? Any input would be appreciated. Original fan and water pump. Thank you
 

Blckshdw

Moderator
Nov 20, 2011
10,665
Tampa Bay Area, FL
The engine temp reads a little low all the time, about 2-3 notches below 210. Now somethimes it only gets to halfway to 210, even in the summer. No CEL.

This is an indication that your thermostat is stuck open, and needs to be replaced.

Has the waterpump bearing seized around the fan shaft?

The fan is screwed onto a shaft on the water pump, if the fan clutch isn't engaged, you should be able to stop the fan. Or at least slow it down to the point you can feel the pulse of the clutch engaging when it needs to. Does the sound of the fan increase/decrease with engine RPM?
 
  • Like
Reactions: mrrsm and Mooseman

4girlsintow

Member
Jan 28, 2019
5
CT
Thanks for the quick response. I just checked, and yes the speed increase and decrease with rpm. Also when I first press the gas pedal it seems like the engine wants to stall which is probably something totally different?

I don’t want to go near the fan and try and stop it. It would definitely do damage.
 

Blckshdw

Moderator
Nov 20, 2011
10,665
Tampa Bay Area, FL
I don’t want to go near the fan and try and stop it. It would definitely do damage.

You wouldn't need to completely stop the fan, just apply enough resistance so you can feel if it's pulling constantly (clutch is fully engaged, as it sounds is the case), if there's a pulse due to the PWM signal commanding the fan to spin at some value less than 100%, or the clutch isn't engaged at all, and you'd be able to stop the fan with ease.
 

TJBaker57

Member
Aug 16, 2015
2,897
Colorado
I just repaired a electro viscous clutch today on my 02 TrailBlazer. It was a junkyard unit off an 03 Envoy. (The previous clutch had loose bearings.) My issue was the fan would only engage intermittently using the Tech 2 and commanding full engagement. Found loose internal connections at the harness connector on the clutch and tightened them up a bit. Could not find any bad wires, just looseness in the actual pins inside the connector. gave the two in question a little twist with a fine nosed pliers. that did the trick.
IMG_20190829_105908_HDR.jpg

Anyway... My understanding of the operation of this thing was helped immensely by a recent post by @MRRSM in which a document describes how this thing works. (Post #8 in https://gmtnation.com/forums/threads/chasing-gremlins.20197/ )
Specifically the opening and closing of a spring loaded valve inside allowing fluid to move inside the unit. That really helps in understanding how the fan speeds up rapidly upon command but takes a while to slow back down. In testing mine with the rag method I felt no pulsing and I could stop the fan entirely. I will note that rather than put a rag at the location shown in the video I went to the opposite side where the fan is moving downward and I had less chance of getting pulled into the space between the fan and the shroud. Eventually I hooked a ohm meter to the 2 wires (black & white) leading to the fan clutch and reached in through the fan blades and wiggled the wiring. The resistance went all over from 10 ohms to open circuit.

I have had that thing out and back in so many times in the last month that this time around I had the clutch out and in hand in 5 minutes 20 seconds including having to go back in the shop to fetch a hammer!!
 

4girlsintow

Member
Jan 28, 2019
5
CT
OK. I went out and started up the truck and the fan was spinning FAST. I took a rag and dragged it against the outer part of the fan and was able to slow it with some firm pressure and eventually stopped it. Right when it stop it tried to "tick" forward as I was holding it. I released it and it took about 30 sec to a minute to get back up to speed and then I heard it engage and spin faster. I feel like I reset something because I've never seen it go slow then engage.
 

Blckshdw

Moderator
Nov 20, 2011
10,665
Tampa Bay Area, FL
Honestly, that sounds like normal behavior to me.
 

bigoilman

Member
Sep 7, 2019
3
Ocean Springs, MS
Resurrecting this old thread. Anyone seen a case where the EV Fan Clutch DISENGAGES completely? I replaced it earlier this year while chasing an overheating condition (along with the Water Pump and Thermostat). I only found it while trying to force it to overheat. It disengaged once at 206 degrees, another time at 217 degrees but sometimes it stays engaged. When I say it disengages, I mean it coasts to a stop! Squeaks while doing it too. Never seems to lock up completely (or even the 90%) I tried applying 12v to the clutch directly (white wire) but there was no speed change. I'm open to ideas.

Further update: I applied 12v to the clutch and grounded it manually as well (just in case) with no success. I realized that it only disengages when it's connected to the PCM. when I have it unplugged, it never disengages like that. Makes me think it's receiving a command to lock up but instead does the exact opposite.
 
Last edited:

Blckshdw

Moderator
Nov 20, 2011
10,665
Tampa Bay Area, FL
Resurrecting this old thread. Anyone seen a case where the EV Fan Clutch DISENGAGES completely? I replaced it earlier this year while chasing an overheating condition (along with the Water Pump and Thermostat). I only found it while trying to force it to overheat. It disengaged once at 206 degrees, another time at 217 degrees but sometimes it stays engaged. When I say it disengages, I mean it coasts to a stop! Squeaks while doing it too. Never seems to lock up completely (or even the 90%) I tried applying 12v to the clutch directly (white wire) but there was no speed change. I'm open to ideas

I'm kinda dealing with that issue right now, although mine isn't behaving quite the same as yours. My transmission is running "normal temps" after running 20-30 degrees cooler due to my aux cooler in front of the condenser for years. I still get the normal, residual spin of the fan, but it's easily stopped with a rag in a few seconds, and there's no tug of the clutch engaging to be felt. Jumped the relay socket, and no change there either.

Gonna be trying the thermal clutch this week.
 

Forum Statistics

Threads
23,271
Posts
637,476
Members
18,472
Latest member
MissCrutcher

Members Online

No members online now.