Viscous Fan/Clutch - PeachParts Mercedes-Benz Forum

LeaUK · #1 05-27-2004, 03:31 AM

Hi all.

Just wondered if anyone has some experience with re-filling MB clutches. I have just completed a refill (see this thread for pics, details and some instruction) for my R129, and all seems to operate as it should. But with one exception.

Should the clutch engage above 95deg AND whilst at 2000/2500RPM or does the RPM have to drop below a certain threshold to engage?

This is what happens to my rejuvenated clutch - even though 95Degs has been reached the fan fails to engage until the revs drop. Unfortunately I don't have a good clutch to compare!

Any thoughts chaps?

BTW - I have read every Viscous clutch post I could find - what great help this forum is

Cheers
Lea

JimF · #2 05-27-2004, 01:07 PM

on my web page (Menu #20), it is active at high(er) engine temps but dis-engages at 2000 rpm which, I think, is perfect.

A stock vfc should only engage at high (110degs) temperature and dis-engage at or around 2000rpm as I recall.

LeaUK · #3 05-28-2004, 06:09 AM

Hi Jim

I should have thought to look through your site - great pics, thanks.

However, I personally disagree with running the fan continuously as this will reduce the efficiency of the engine, increase fuel consumption and increase noise, when there is no real reason for the fan to be engaged.

R129s and I suspect most, if not all MBs with Viscous fans, are designed such that when the car is travelling in top gear the cooling system alone will maintain the optimum operating temp, ie around 85-90DegC - no fan is required. When at idle there is minimal air flow due to minimal movement and the cooling system relies on the viscous fan. The car should still remain at optimum temperature - I've tested both these conditions and my 129 supports this - Should you be in traffic then the fan clutch will engage due to engine RPM continuously changing (stop, start, stop, start) and the rad again has minimal air flow.

If your fan doesn't engage before the RED zone on the temp gauge I suspect it is faulty- sounds like lack of viscous fluid to me! The bi-metallic strip should deflect when the air flow temperature reaches around 90-95DegC which is around 85DegC on the water gauge and is significantly below the warning zone of my R129 (which is at 120DegC). As a result fluid is pumped into the second chamber causing the fan to significantly increase RPM - you should hear a whooshing sound!

The characteristics of the clutch are governed by the viscosity of the fluid within, so for correct operation below 95DegC the fan should rotate at around 1/2 engine RPM (as measured) ie there is just enough fluid in the second chamber to provide sufficient grip. At around 90DegC the bi-metallic strip will deflect and the clutch will engage - fan RPM should increase to match the engine.

Above 2000/3000RPM the clutch doesn't actually disengage but rather the viscosity of the fluid is such that is cannot maintain greater RPM. This is why the fluid plays such a major part. It governs the disengaged Engine RPM to Fan RPM ratio and the max RPM of the fan.

However my question (should anyone have finally got here) is:

Will the fan RPM increase (clutch engages) to match engine RPM if the engine RPM is already at around 2500?

The experiment I tried was to run the engine at around 2500RPM until I expected the fan to engage, but it didn't. When I reduced engine RPM below 1500 it engaged immediately. Now this could be by design as assuming the car is at greater than 1500 or so one could assume the car is in motion and therefore the rad has sufficient air flow even without the fan fully engaged. I think the only way to find out is to measure a known good clutch!!

JimF · #4 05-28-2004, 11:48 AM

but the answer to your question . . . . Will the fan RPM increase (clutch engages) to match engine RPM if the engine RPM is already at around 2500? . . . . is No!

The fan(s) only operate to 2000rpm area not matter what the engine/radiator temperature. I think you said that in the first paragraph.

But you are incorrect re the temperature at which the vfc engages: for the W140 and W129 cars, it's very high, as I said, about 110degC. One could ask, why bother since it's already a bit late!

Check my Menu #21 to see a test of some of the Sach's newer vfc fans. I was looking for a fan clutch to replace my W140's vfc but couldn't find one.

If you search here, you will find that Sachs does make a newer version vfc that operates at 90C or so. Check out this thread which is very informative and has pics. Here's the link:
http://www.peachparts.com/shopforum/showthread.php?threadid=43792

LeaUK · #5 05-28-2004, 02:49 PM

Jim

Thanks for your reply.

Quote:

but the answer to your question . . . . Will the fan RPM increase (clutch engages) to match engine RPM if the engine RPM is already at around 2500? . . . . is No!

According to many posts made in other forums here, the majority say it's around 3000RPM and from my own testing this seems true - although I've replaced the viscous fluid so the original performance may of changed - and that I appreciate.

So, why doesn't the clutch attempt to engage at higher than 1500RPM ? I suspect it's because the fluid is thrown such by the centrifugal force - it cannot enter the correct chamber to increase fan RPM - what's your thoughts?

Quote:

But you are incorrect re the temperature at which the vfc engages: for the W140 and W129 cars, it's very high, as I said, about 110degC. One could ask, why bother since it's already a bit late!

I would be very surprised if Benz designed a system that was so poor, but I have no experience of other than my R129 clutch, so I take your point. My 129 engages at water temp of 85DegC, which (without measurement) I suspect is around 95C air temp accounting for the mass of the engine and coolant channel. No changes to the bimetallic strip have been made!
Don't forget that the strip operates on air temp not water temp, and there will always be a substantial difference between the two. Also, if your clutch fluid has begun leaking out then the performance will be significantly affected causing the effects you have witnessed. A reduction in clutch fluid is VERY common for any viscous clutch for any car!

My plan is to measure a known working clutch as I see this the only way to confirm correct operation - unless I talk to the designers of course!!

Lea

JimF · #6 05-28-2004, 03:10 PM

Sure maybe a brand new vfc could be as high as 3000rpm. But 2000 - 2500 rpm is about optimum, wouldn't you agree?

Maybe you missed the point of the 'adjustable' screw height in Menu #20; it's simulates a slight bending of the bms which is amplified when the engine and radiator get hot.

The bms bends a little; check the pictures in Menu #21 of the bms in 212F water. Note that they bend enough so if you have pre-bent it, it will tend to actuate the clutch earlier. With the screw height adjustment to 'help' the bending process along, you find that just a little bending will fully engage the clutch.

By adjusting the screw height, you can somewhat control the onset of the clutch. Granted that it's not as good as having the correct bms (say 90C verison described in the other thread) but it does work nicely. Since there's none available, this is about the best we can do.

LeaUK · #7 05-28-2004, 06:42 PM

Hi

Thanks for your comments and after reading your #Menu21 I see you've carried out some very valuable experiments yourself too - great stuff.

Yes, I did misinterpret your proposed mod as I skimmed through it and noticed that most of the text points to the fan being engaged constantly

Quote:

This will give you full tilt fan engagement right up to the high speed disconnect around 2,500 rpm. You won't believe the amount of fan noise but the engine runs cool as a cucumber.

and

Quote:

So this modification changes the clutch to lock up at all times.

but you are quite right, the mod will in fact allow more control and adjustment than I originally considered - however TBH, this isn't my question as I'm not doubting your mods or experiments.

I want to know if the clutch should engage if the engine RPM is already greater than 1500. That's my uncertainty.

Quote:

But 2000 - 2500 rpm is about optimum, wouldn't you agree?

Yes, most definitely. 3000RPM does seem a little high - but I've chosen 12,500CST fluid and maybe MB original is slightly lower. However I have canvased several mechanics opinion on how mine feels compared to original clutches and most agree that it has the correct amount of resistance - in fact one said it feels slightly less resistive than he would expect!!

My personal opinion is that Benz/Sachs designed the VF to operate at around 95DegC (being a sensible temperature) ie the strip bends (n)mm releasing x(cc) of silicone fluid thus gripping the fan by (y)N - but as the VF ages the fluid either changes viscosity or the volume of fluid reduces thus the strip has to bend further to engage (ie higher temp is needed to achieve the same cooling). This maybe wrong but this makes more sense to me, rather than poor design.

BTW - have you any thoughts what the black plastic casing's purpose is? Is it there for protection only or maybe some kind of airflow regulation?

Lea

JimF · #8 05-28-2004, 07:53 PM

the error; I'll fix that asap.

Yes, if the cutout was around 2500rpm initially, it has changed to 2000rpm today. That could very well be since it's about 3 years or more.

So this modification changes the clutch to lock up at all times.. Yes, it was touted to do that but with careful adjustment of the screw height, it really doesn't. Guess I forgot that I wrote those words. More correctly stated, it works very nicely "part-time"!

My personal opinion is that Benz/Sachs designed the VF to operate at around 95DegC (being a sensible temperature). . .
No, they didn't!!!! It was designed for the 110C temp range. Yes, strange isn't it? But you saw, in the referenced thread, that a new version is available that has lowered the cut-in temp but not for the 129 or 140s.

The black plastic actually 'shields' the heat and you can improve the heat transfer from the radiator by removing it.

Just because it's "benz", don't think it's perfect!!!

myarmar · #9 05-29-2004, 12:25 AM

This is from W140 Introduction into Service Manual, 1992 Model:
119 engine
"Visco-fan clutch
The visco fan clutch is maintenance-free and turns CCW. The visco-fan clutch is temperature and rpm controlled.
Identification:
Yellow label with number 119 200 00 22.

Upon reaching an air temperature of approx. 92 to 100C at the bimetallic spring, the fan smoothly engages.

Up to an engine speed of approx. 3600 to 3800 rpm, the fan speed is proportionate to engine speed, whereby slippage in the visco-fan clutch will not allow fan speed to exceed approx. 3000 rpm.

The safety disengagement of the fan occures audibly with increasing engine speed via centrifugal force at approx. 3800 rpm, at which point fan speed drops to approx. 600 rpm.

Reengagement occurs audibly after engine speed drops below 3300 rpm."

Attached is a crossection of the visco clutch for 120 engine:

JimF · #10 05-29-2004, 02:49 AM

'specs'! Note that the cut-in temp is 100C (212F) which translates to at least 110C as read on the IC's engine temperature gauge. It's no where near 92C since I've tested them and they start to bend at 212F (100C).

Learoy, hope we have changed your opinion that MB couldn't have designed anything that 'bad'? Ya, they did!

LeaUK · #11 05-29-2004, 04:31 AM

myarmar

Thanks for your confirmation, I thought I was going mad for a minute but the specs confirm my thoughts. However, I'm particularly interested in the 3800 cut out protection. I haven't actually took her above 4000 or so with my head in the bonnet and therefore this is definitely a test I need to perform. How the clutch achieves this I'm not sure!

[Edit] - I see a spring - very neat. That's how it works. Thanks for the piccy much appreciated.

Quote:

Re engagement occurs audibly after engine speed drops below 3300 rpm

damn, still doesn't answer my question; or does it?

I'm just going to have to get testing!!

Jim

Quote:

110C as read on the IC's engine temperature gauge

Uhm - I've had a change of thought! I originally surmised that the cooling system temperature must lag the air temp. but of course this cannot be true as the stimulus of the air temp is caused by the water temp. itself! However, there maybe a delay in gauge readings as I'm not sure if these are traditional or electronic - even so the display is obviously averaged.

So assuming a lag, the temp. could be even higher than 110C

110C is rather close (assuming 10C between water and air, 92C would yield say 100C; a much more sensible choice) and just doesn't seem right to me, especially as mine cuts in at around 85C on the IC - maybe I have a different strip or maybe a very lazy IC gauge??? And another thing - my electric fan cuts in at IC 110C. Why design a system that has two fans operating at the same point??!??

I'm almost temped to remove the strip again and carry out some further testing. Or I could place a thermocouple in line with the air and measure the air temperature accurately - if I do I'll make a further post for info.

JimF · #12 05-29-2004, 01:00 PM

than 110C, maybe closer to 120C; it depends on the coupling coefficient between the VFC and radiator and secondarily. the engine heat radiation.

If your aux fans don't come on until 110C, I'd recommend you change your B10/8 sensor or, if not, purchase a Cool Harness. This will force the ax fans to come on around 95 to 100C depending on the model.

By MB spec, the aux fans are supposed to come on at 100C with AC off or 14bar with AC on. Check the table in Menu#18 about halfway down the page.

3800rpm: I think that about halfway to the speed of light!

If I remember correctly, 3100 is 105mph, so 3800 is very, very fast! MB's choice is certainly up for question.

myarmar · #13 05-30-2004, 01:46 AM

If you have W126 CD there is a very good information on the visco clutch operation for 116 and 117 engines. The design is identical to 104 - 119 - 120.
The file is on the second CD. Search for file 20-320.pdf

LeaUK · #14 05-30-2004, 06:58 AM

Jim

Again, great info on your site re the B10/8 sensor - nice work.

Best to get the DVM out I think and take some real measurements before I replace this. Trouble is, now my viscous is working, my car never reaches above 85C

Time to remove the sensor and dip it in some hot water!

Yesterday I took her for a spin - ambient temp was 20C. Urban driving for 45mins with A/C on, temp reached 85C max (IC) and was constant throughout. Following, I spent 20mins at around 40-50MPH constant - IC gauge reports 82C. Then onto some hard driving using manual shifts and keeping the revs high for around 15mins - gauge never exceeded 85C. This sounds almost ideal and how I would have designed the vehicle's cooling performance.

I believe the most efficient operating temp of petrol engines is somewhere around 85-95C, so mine's maybe on the lower side but as you've mentioned previously - more horses for your buck!

I can also confirm that the clutch disengages at around 4000RPM as previously mentioned by myarmar, and will re-engage too but I failed to note at what RPM - I'm sure it was around 2000RPM.

However the system was designed, my vehicle's cooling now operates superbly and the car is, as usual a complete joy to drive; hence I got carried away and forgot to note some RPMs.

Let's hope today brings more sun

LeaUK · #15 05-30-2004, 04:15 PM

I'm now starting to wonder if I have a faulty B13 temp. gauge sensor. My viscous fan operates as per testing procedure (thanks to myarmar for instruction) but engages at 85C not 95C (as IC gauge reading). My Aux fans engage at 110C (I assume all 119 engines are the same and thus should operate at 100C). The common link is what I'm using to measure temp, the IC gauge; now where did I put that thermocouple?

Has anyone got the transfer function (temp Vs resistance) for the B13 sensor - maybe it's the same as the B10/8? Currently (car has been still for a while) reads 1.02Ohms across the B13 and the IC gauge reads 65C, but my fingers dipped in the coolant say around 75C.

Jim - BTW my aux fans seem to be 'stageless' ie only on or off. You mention that the SClass they're three stage, do you expect all 119 engines to be the same?

Lea

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode