Viscous Fan Clutch Replacement

[gerryvz]

As far as I know, the clutches have been out of production for several years, and are still marked as NLA from MB.

Sachs made them for many years, and the last years of production they were made by Horton, which is part of Sachs.

Where are these photos from? Are they recent?

Here is current availability from both MB Classic Parts in Germany and also RevParts catalog in the US.

 

[gsxr]

Is the new factory clutch made by Behr and looks like this?

That is a cheap aftermarket clutch, not OEM Sachs/Horton. I've never seen a Behr clutch for p/n 119-200-01-22. That may be a fake "Behr" label applied to a no-name clutch. And, aftermarket parts never use the Mercedes Star logo anywhere, which also makes me think that's a fake label (or, Photoshopped).

Behr is not OEM for this clutch. Looks like someone took a cheap no-name clutch, applied a fake "Behr Germany" decal, and then put it on an OEM Genuine box+label to make it look legitimate.


The genuine Sachs/Horton clutch looks completely different. Photos are at the link below.
Index of /images/M119/fan_clutch_119-200-01-22

 

[elietristan]

That is a cheap aftermarket clutch, not OEM Sachs/Horton. I've never seen a Behr clutch for p/n 119-200-01-22. That may be a fake "Behr" label applied to a no-name clutch. And, aftermarket parts never use the Mercedes Star logo anywhere, which also makes me think that's a fake label (or, Photoshopped).

Behr is not OEM for this clutch. Looks like someone took a cheap no-name clutch, applied a fake "Behr Germany" decal, and then put it on an OEM Genuine box+label to make it look legitimate.


The genuine Sachs/Horton clutch looks completely different. Photos are at the link below.
Index of /images/M119/fan_clutch_119-200-01-22



View attachment 198609

I inquired about a listing from a website that seems to be out of the UK somewhere. This guy was messaging me with this picture telling me it's an original MB fan clutch and that he has two units left. I was suspicious but I was having wishful thinking however I wasn't sure if Behr made these clutches for MB? So I reached out to Dave and he saved me from losing $250 on a fake cheapie Chinese aftermarket junk part. As always thank you Dave

 

[gerryvz]

It's good you were suspicious and asked. So many scammers out there these days.....

Behr is indeed the OEM for the fan clutches that are used on the M103 and M104, and even the M117 engines. But for the M119 as used on the E500E and 400E420, the only OEM ever was Sachs+Horton.

 

[Jlaa]

On balance, I feel like the label and the part number on the part are all suspiciously well-inked. I've never seen a part where the ink stampings are so ridiculously clear and unambiguous!

 

[gsxr]

On balance, I feel like the label and the part number on the part are all suspiciously well-inked. I've never seen a part where the ink stampings are so ridiculously clear and unambiguous!

I couldn't find any clutch with the part number inked on the face like this. I'm curious of the origin.

The Star logo on the label is, IMO, Hella sloppy. But I don't have a genuine Behr label to reference for comparison.

Regardless, this clutch is sitting on top of an OE box with 2010 date code and a genuine 2010 clutch would be made by Horton.

 

[gerryvz]

The part number looks photoshopped on, to me. The label is clearly not stock and the lettering and crispness of the printing on it, also shows it is fake. If you've ever seen a Behr label on a fan clutch, usually they are red, and have different lettering that is not as dark/bolded as what is shown on this fake yellow label.

I believe I have made and posted photos in the past of M104 and M117 fan clutches, in the respective forums.

 

[elietristan]

Okay guys. I know Dave and Gerry are going to love this one. I just received in the mail today an authentic Mercedes Benz dealer item only Fan clutch for my beloved 500E. This my friends I am holding a marvelous piece of history in my hands. Truly a very rare and precious item worth more than gold itself . This is a brand new NOS clutch!! Here are a few pics.

 

[Jlaa]

Re: Vicous Fan Clutch Replacement


Car #1 - Good Clutch:
With the engine at ~650rpm idle in park, pulley speed was ~810rpm, actual fan blade speed was ~790rpm (almost 98% lockup).
I didn't think to test it at higher engine RPM's, I will do that in the future for grins.


Car #2 - Bad Clutch:
With the engine at ~650rpm idle in park, pulley speed was ~810rpm, actual fan blade speed was ~275rpm (only ~35% lockup).
With the engine at 1500rpm in park, pulley speed was ~1900rpm, actual fan blade speed was ~500rpm (only ~25% lockup).
With the engine at 2500rpm in park, pulley speed was ~3200rpm, actual fan blade speed was ~500rpm (only ~15% lockup).

Car #3 - Good clutch, NOT engaged:
With the engine at ~650rpm idle in park, pulley speed was ~800rpm, actual fan blade speed was ~325rpm (~40% lockup).
With the engine at 1500rpm in park, pulley speed was ~1900rpm, actual fan blade speed was ~410rpm (~22% lockup).
With the engine at 2500rpm in park, pulley speed was ~3100rpm, actual fan blade speed was ~980rpm (~32% lockup).


Car #3 - Good clutch, engaged:
With the engine at ~650rpm idle in park, pulley speed was ~800rpm, actual fan blade speed was ~775rpm (97% lockup).
With the engine at 1500rpm in park, pulley speed was ~1900rpm, actual fan blade speed was ~1750rpm (92% lockup).
With the engine at 2500rpm in park, pulley speed was ~3200rpm, actual fan blade speed was ~2600rpm (81% lockup).

.... fast forward 14 years (!!!!)

I obtained this cheap laser RPM tester thing about 24 months ago and decided to use it to test my ACM fan clutch (was in a Borg Warner box) that I installed in 2016. I tested this at a cold start and have not yet tried warming up the car or anything. I had to use painters tape to reduce the reflectivity of the metal fan tips to get the laser to read correctly.

Does this tell me anything about the health of the fan clutch so far (given that this is a cold start and not up to temp - haven't tried that yet) ---- seems to be at nearly full lockup (794.0 - 854.9rpm)?

 

[gsxr]

Does this tell me anything about the health of the fan clutch so far (given that this is a cold start and not up to temp - haven't tried that yet) ---- seems to be at nearly full lockup (794.0 - 854.9rpm)?

Full lockup at cold start is normal-ish... however within 30-60 seconds, the clutch SHOULD disengage and RPM's should drop noticeably (to like 300rpm or so). That's the whole point of the clutch!

However, your problem had been a suspected lack of engagement at operating temp... so you'll need to see what happens when it's hot.

 

[Jlaa]

Full lockup at cold start is normal-ish... however within 30-60 seconds, the clutch SHOULD disengage and RPM's should drop noticeably (to like 300rpm or so). That's the whole point of the clutch!

However, your problem had been a suspected lack of engagement at operating temp... so you'll need to see what happens when it's hot.

Thanks! OK, I idled the car for many minutes until the car got to operating temp. I was able to confirm that revving the car to about 3500-4000rpm results in no-roar-sound.

Then I dropped the car back to idle and observed that for the next 30 - 60 seconds or so, the clutch seemed to be disengaged and revving to 2500 rpm or so yielded much-reduced-roar-sound.

Then, by the time I was able to exit the driver's seat and get the laser tool pointed to the right place again (after 60 seconds), I noted that the laser tool once again showed full lockup at idle. I went to the driver's seat and pressed the gas pedal and confirmed full-on-roar-sound while revving just off idle.

This sounds like OK functionality (if not exactly optimal for quiet operation ---- which we've always suspected is the case for these ACM / Febi / BW clutches) from a cooling perspective?

 

[gsxr]

It sounds like the basic function is working. However, as mentioned previously (might be buried in this thread), the clutch is analog in design. When at operating temp (say, 90C) the engagement percentage will vary based on a bunch of factors that affect the air temp coming through the radiator fins.

This results in the clutch engaging off idle, but disengaging at an RPM somewhere between, say, 1000-3000. The clutch on my one car I mentioned a few days ago was disengaging at ~2000rpm. At full lockup the OE clutch will remain engaged up to 3500rpm, at which point the centrifugal release will disengage the clutch regardless of temperature. The ACM-type skinny mini clutch may not have the RPM functionality, I don't know for sure.

You are wanting the clutch to remain engaged at 90-100C coolant temps and up to ~3000rpm or so, for improved airflow poking around town.

 

[Jlaa]

It sounds like the basic function is working. However, as mentioned previously (might be buried in this thread), the clutch is analog in design. When at operating temp (say, 90C) the engagement percentage will vary based on a bunch of factors that affect the air temp coming through the radiator fins.

This results in the clutch engaging off idle, but disengaging at an RPM somewhere between, say, 1000-3000. The clutch on my one car I mentioned a few days ago was disengaging at ~2000rpm. At full lockup the OE clutch will remain engaged up to 3500rpm, at which point the centrifugal release will disengage the clutch regardless of temperature. The ACM-type skinny mini clutch may not have the RPM functionality, I don't know for sure.

You are wanting the clutch to remain engaged at 90-100C coolant temps and up to ~3000rpm or so, for improved airflow poking around town.

Thanks. Now I understand the difficulty in testing proper operation. I guess I need to:

(1) Maintain ambient air temperature hot enough
(2) Get engine hot
(3) Presume everything is hot enough that clutch is at full lockup
(4) Accelerate engine steady state to, say, 1500 rpm, 1750rpm, 2000 rpm, 2250 rpm, various points up to 3000 rpm (simulating poking around town). Doubt that wifey is willing to help much with the science experiment.
(5) Measure RPMs.

 

[Jlaa]

It sounds like the basic function is working. However, as mentioned previously (might be buried in this thread), the clutch is analog in design. When at operating temp (say, 90C) the engagement percentage will vary based on a bunch of factors that affect the air temp coming through the radiator fins.

This results in the clutch engaging off idle, but disengaging at an RPM somewhere between, say, 1000-3000. The clutch on my one car I mentioned a few days ago was disengaging at ~2000rpm. At full lockup the OE clutch will remain engaged up to 3500rpm, at which point the centrifugal release will disengage the clutch regardless of temperature. The ACM-type skinny mini clutch may not have the RPM functionality, I don't know for sure.

You are wanting the clutch to remain engaged at 90-100C coolant temps and up to ~3000rpm or so, for improved airflow poking around town.

What is the failure mode of these clutches? I ask because I am assuming failure modes could be:

1. The viscous fluid part not working right (or leaked out) so the thing is dead OR
2. The bimetallic strip is not working right so full engagement is not happening at the right temperature.

Is there any other failure mode? What I am wondering is ---- since I verified that basic operation is OK at idle (full lockup), can I assume my clutch isn't deciding to have, say, 15% lockup at, say, 1500rpm?

 

[gsxr]

What is the failure mode of these clutches? I ask because I am assuming failure modes could be:

1. The viscous fluid part not working right (or leaked out) so the thing is dead OR
2. The bimetallic strip is not working right so full engagement is not happening at the right temperature.

Is there any other failure mode? What I am wondering is ---- since I verified that basic operation is OK at idle (full lockup), can I assume my clutch isn't deciding to have, say, 15% lockup at, say, 1500rpm?

Those are the 2 primary failure modes. A third, which is pretty rare, is a failed center bearing.

Note that with #1 the engagement may gradually decrease / get worse as fluid is lost. Everything about these clutches is analog, not binary. The changes can happen gradually over long periods of time.

Unfortunately, the lockup percentage at idle doesn't tell you anything about what's happening at higher RPM. This is why the "roar test" is the only valid test. Problem is this really needs to happen on a warm day with the engine hot. It's really difficult to test in winter ambient temps.

 

[Jlaa]

Those are the 2 primary failure modes. A third, which is pretty rare, is a failed center bearing.

Note that with #1 the engagement may gradually decrease / get worse as fluid is lost. Everything about these clutches is analog, not binary. The changes can happen gradually over long periods of time.

Unfortunately, the lockup percentage at idle doesn't tell you anything about what's happening at higher RPM. This is why the "roar test" is the only valid test. Problem is this really needs to happen on a warm day with the engine hot. It's really difficult to test in winter ambient temps.

For the first failure mode, where the viscous fluid part is not working right, is it possible for this failure mode to happen, in an analog way (meaning degraded function --- not clearly fail / not fail), WITHOUT have any traces of leaked out fluid anywhere?

Thanks!

 

[gsxr]

For the first failure mode, where the viscous fluid part is not working right, is it possible for this failure mode to happen, in an analog way (meaning degraded function --- not clearly fail / not fail), WITHOUT have any traces of leaked out fluid anywhere?

Theoretically possible, but unlikely. There's usually evidence of fluid leakage. If all of the fluids are still inside, the next issue is #2.

This also assumes the radiator & condenser fins directly in front of the clutch are completely clean and not blocking airflow... and, that the radiator doesn't have a "cool spot" ahead of the clutch.

All of which are reasons to ditch the stupid thing and go PWM!

 

[Jlaa]

Theoretically possible, but unlikely. There's usually evidence of fluid leakage. If all of the fluids are still inside, the next issue is #2.

This also assumes the radiator & condenser fins directly in front of the clutch are completely clean and not blocking airflow... and, that the radiator doesn't have a "cool spot" ahead of the clutch.

All of which are reasons to ditch the stupid thing and go PWM!

Yes! I have all the parts for the PWM project! It will take a few years to get started though … I need to build a new garage with a lift and a large house addition first. I’m in the “arguing w the city” phase.

 

[nocfn]

No arguments in Texas or Idaho.