Here's how to fix the viscous coupling on the radiator fan, overheating


There's been a lot of speculation over these couplings but i think we have an answer. If you've already jammed it up read on, you can fix it as good as new for just a few pounds. Better than reducing your MPG with a noisy fixed fan or giving Daihatsu 200 quid for a new one.

If you just want the gist -

The problem is most likely that the pin at the middle of the bimetallic coil has seized where it passes through the body of the unit. This needs to turn to open a gate inside that allows the viscous fluid to move around. If the pin doesn't turn when you heat the coil open the unit and tap the pin back and forth to free it. Close it up and test it. Job done! It really is that simple Smile

Or the long winded explanation -

The cooling system on my Fourtrak seems easily up to the job but crawling up a welsh mountain in 4WD low range i boiled it up. The dash gauge isn't very good as it was boiling up way before it was in the red. If your needle ever moves above it's usual position watch out!

The radiator fan would turn freely even when the car had overheated so i knew the coupling was at least part of the problem. As it only overheats with extreme load i'm pretty sure the rest of the system is ok, no need to re-core the rad etc. If yours overheats with normal driving your problem is going to be more than the viscous coupling, though it could still be part of the problem. Take out the thermostat and check it opens fully and at a low enough temperature. If that's ok, try flushing the radiator with the bottom hose off and then fix the coupling. IMHO DO NOT just put a new rad cap on before you've found the problem, that rusty old spring might be saving your engine by releasing pressure easily. It won't be the original cause of overheating as you'll only lose large amounts of water through it if you're overheating in the first place. If you put a stronger spring on there the pressure will try to find a why out somewhere else, an old hose bursting won't make your day any brighter...

By the way you don't need to stick a broom handle in the moving fan to see if the coupling's engaged Shok just try and turn it with your hand when you've just turned off a hot engine, you're not looking for fine degrees of stiffness, just pretty stiff or very loose. Or watch it stop when you turn it off, if it stops almost straight away it's locked, if it spins on slowly to a gradual stop it's not engaged. The difference is very obvious. It's only designed to come on when the engine is a fair bit above normal operating temperature though so you'll have to be sure the engine is properly hot, not just after a normal drive, or you'll not be any wiser. Best to use an infrared thermometer or the like and rev it up until the top of the thermostat housing is +90 degrees C. If it takes an age, and it probably will, you can be pretty sure the rest of the system is ok.

If you can't get the coupling to lock up you can fix this supposedly sealed, throw away, 200 pound unit for nothing in under an hour.

The coupling is a very simple design and unless it's been damaged there are only 2 possible things that can go wrong, either the viscous fluid has escaped which if the coupling hasn't come apart i think would be very unlikely. (There isn't much fluid visible normally so don't worry if you can't see much when you open it and the O'ring seemed too big a diameter but went squarely back in with a bit of fiddling.) Or the other thing is the bimetallic coil is not functioning properly, this is what happened with mine and is very likely the problem with yours.

I don't think the fluid gets old and stops working, i.e. the fluid is not some magic stuff that gets more viscous with heat. It is silicone fluid and silicone is known to resist heat very well, i think the secret is that it STAYS VISCOUS even when hot where other fluids would quickly become thin. (With a hot air gun on it the fluid still moves around at about the same speed as when it's cold.) It is the amount of fluid at the outer edge of the inside of the coupling that increases and causes the 2 parts to lock, it is not the increase of the fluid's temperature.

The amount of fluid is controlled by a thin plate (the blue steel bit in this one) inside the unit that rotates to open 2 small holes which allow fluid to flow outward and lock the 2 sides of the coupling. The plate should rotate when the bimetallic strip on the front of the case heats up, by hot air from the radiator, and turns a pin at it's center that passes through the casing. If this pin is seized, the plate will not turn, the fluid will not flow and the coupling will not lock.

If you take the coupling off the engine try heating the coil on the front of it with a hot air gun (only takes a few seconds) or a hair dryer might do it if you hold it close for a while. Watch the pin at the center of the coil, it should gradually turn by about 30 degrees and turn back when it cools down. If it doesn't move you've found the problem.

Before i started I got hold of some silicone oil from a model/hobby shop, used for RC car diffs, but i might not have needed it. This stuff will do it - but you have a choice. Toyota, who don't pretend their couplings should be disposable, quote using 3,000 cps for cold countries and 10,000 for Australia etc. I went for 7,000 to account for global warming Wink and ended up using less than half a bottle for what i spilled before i'd worked out what to do. If you've drained it and bunged it up with rubber you would probably need no more than a bottle but thats a guesstimate, i couldn't find a figure for it. If you add too much you'll find the fan is always on so you could drain some out and try again. (Actually that's probably a kinder fix than rubber if you want an always on option or you can't free the pin)

If you have the time you might be able to save all the fluid by keeping the closed unit coil side up overnight (it moves slowly this stuff) to drain the fluid into the other side which doesn't need to be worked on. I thought i'd be able catch it when it came out in case i needed it but it's harder than it sounds, like treacle and tricky to handle, i was glad i had some spare. Might be best to crack the screws loose before leaving it to stand. You need a number 3 crosshead/philips screwdriver for the 4 screws which will probably be plenty tight.

Once you got the halves apart place the top half coil side up, spray some WD40 under the coil and tap the middle of the coil with a hammer, it should move in a mm or 2. Turn it over and hit the other end of the pin using a punch or head of a nail etc so it moves back through, turn it over and repeat until you can push the coil side back in with your thumb, until it's loose. Try heating the coil, if it now rotates add oil if you lost some, screw the 2 halves back together and get it back on the motor and try it out. When the engine gets up to temperature, took ages at fast idle, it should suddenly come on with a roar. A very satisfying fix Smile

I think the problem is that this coil so rarely needs to function, only with towing or heavy off-roading, it has a lot of opportunity between times to seize up. Might be worth occasionally getting some heat on it or letting the engine get some heat up.

Great car these Fourtraks, well worth keeping them out of the scrappy for as long as possible Wink so i hope this fix helps someone out!

All the best

Viscous coupling

I am very grateful to you Abdllah for sharing that information with us. I have mechanically locked up my fan coupling to overcome the problem, but as you rightly say, it is now noisey and inefficient (and I did wonder also about the possible damage to the water pump as there is now no flexible drive between that and the fan). Also with winter coming the fan is not needed except when towing hard. I was therefore looking at the possibility of fitting an electric fan but there is not much room and this could also be expensive unless you are lucky at a breakers. So, once again, thanks for that information, I will give it a try.

Sorted mine by...

...very gently rocking the end of the bi-metal coil out of its anchor slot (in the coupling casting), then nipping the end of the pin with mini-mole grips, a squirt of WD40, and mucho wiggling clockwise/anticlockwise for a few minutes to free the pin off totally.
I dried the assembly out overnight in the airing cupboard (wrapped in a blanket to protect the laundry! Smile ), checked it was still loosely rotating, re-seated the coil tip in its slot, and job done! Not one warm-up during a run to the Isle of Man (including some bouncy off-roading & hill-climbing over there)...

Good idea. Now we know the

Good idea. Now we know the problem and the pin is accessible from outside it isn't needed to dismantle the unit. The pin can travel in and out a few millimetres so a gentle tap and pulling it out again a dozen times might do it without even having to remove the coil if you have fine enough needle nose pliers?

viscous fan

Very useful info Abdullah - I believe this is a common Fourtrak problem.

I thought I would share a related observation:-

Firstly don't believe anyone who tries to tell you that the fan plays no part above 40 or 50 mph. From my experience on an F78, this is not so. Admittedly I have aircon installed which provides additional resistance to frontal airflow. During the 2006 UK heatwaves I had regular overheating with the engine at full load on hills even at 50+mph. Turned out that the fan was the culprit. I found that the pin was inititally reluctant to move, but even when freed the fan was still not engaging sufficiently under the overheating conditions. The cure was to prise out the end of the coil spring which attaches to the housing and manually 'bend' the coil so that the pin rotates a few degrees allow the viscous fluid port to open slightly earlier. Only a very small adjustment is needed. On the first attempt I went way too far and the fan was enagaged pretty much all the time. Instantly stopped the overheating though but was noisy and fuel consumption dropped by 1.5 - 2mpg! I would guess that locking the fan completely would would generate yet more noise. That would be OK to get you out of trouble but not good if you do lots of miles..

I noted while doing this that the coiled spring was quite rusty on the surface. I guess that this plus many thousands of thermal / stress cycles had gradually 'weakened' the spring into a new position. The fix seems to have lasted OK so far (~14 months) - temp gauge never budged since.

viscous fan

Just as a further observation, I noticed when the coupling was dismantled that the pretension on the bi-metallic coil is adjustable by way of the two elongated slots. Is it possible that the coil gets 'weak' with age and even when hot is not able to overcome the pretension to open the ports enough? I have adjusted mine to the minimum position in the hope that this cures the problem.
In my case, the cooling system works fine when the fan is mechanically locked up, the fan pivot pin was not seized, and there is no evidence of viscous fluid leaking out.
In my experience (with 2 fourtraks), the cooling system copes well (even under hard conditions) when the fourtraks are young but after about 8 years old they have trouble managing. In one case even changing the radiator did not improve the situation, so maybe, just maybe, a weak coil could be a factor.

How Much?

I tried a 2oz bottle of 10,000 oil and it just locked up the fan.

Has anyone figured out how much oil to use?

And I would need to clean out ALL the old oil before adding a different weight correct?

Fan oil replacement


Just curious, why did you use the 10,000 oil, as thats suggested for
use in Australia . hot climates !!!. Where are you in the US.

Thought the oringinal post was quite a revelation, and far better to try it than just locking up the fan with rubber sheet or card.

I am in the UK and when motorway driving at the limit (70MPH) or plus, I have to watch the temp gauge, and put the heater on full with windows slightly open to get that extra bit of cooling, I know my truck was not built as a motorway cruiser, it just does concern me a little. As I have a replacement rad , new thermostat, and change the fan belt every year.

It would be interesting if others would post to say they had tried this option.

Curiously, this post re the viscous fan, has been the only post that the original poster has ever made on the site ?????, as far as I can see . Hmmm !!!!Hmm!!!!

Edward (ews) '92 Fourtrak 2.8 TDX

Well all they had was 10,000

Well all they had was 10,000 and I thought it wasn't far off from 7,000. The 10k was way to thick I think. I may try something around 5k and less oil. I'll let you know.

My rocky used to overheat horribly. 2 head gaskets, water pump, radiator, rebuilt head, thermostats e.t.c

Now I can't get it above C on the temp gauge. Can't even use the heater its so cold!

Funny from one extreme to the other! Better than overheating though!

But yes if you just fill it up with oil it will lock easily and keep it cool.

Is yours a Sporti Ews?

Is yours a Sporti Ews? Either way I find your need of heaters worrying. I have hade F80 and Sporti and F78 past the legal for more than short runs plenty of times with no ill effects. Apart from the the driving licence. Oops.

Any veiws expresed in this thread by me are purely from my own experience, and (sometimes) falible memory. Hope my comments help, but please don't take them as gospel.

Any veiws expresed in this thread by me are purely from my own experience, and (sometimes) falible memory. Hope my comments help, but please don't take them as gospel.

She is a F75 .91 2.8 TDX

Hi Nev.K

had this heating problem for years, constant driving at 70 mph causes rapid water temp increase. put the pedal down to the metal, and you need the cabin heater to keep the temp under control, no problems up to this speed, though really long motorway inclines need an eye on the gauge, thats my partners job !!!!!. we never let the truck go critical, just a niggling problem. All the usual tricks tried, replacement rad, thermostat, clear grill area, fan locks up with broom test. She will top out at 95mph as indicated but then she was never built as a motorway cruiser.

Edward (ews) '92 Fourtrak 2.8 TDX

Fan oil replacement

Biggrin Sorry, if i had any other revelations i'd have posted them!

It's easy enough to test if the coupling is engaging on the vehicle as per original post and it sounds like you've tried everything else. Other than the water pump? It might be that it's not really overheating to a problem level. You can find out by watching the dash gauge as your partner calls out the temperature of the water, i found it was easiest to measure from the top of the thermostat housing which seemed to accurately show the water temp coming from the block. If the gauge moves above normal before the fan comes on, and the fan comes on at 90 ish it might be that there's nothing to worry about. Though no warranty comes with this information Wink

I wonder if i might umbly suggest stickying this thread? I bet half the Fourtraks out there have stuck pins, i had no trouble from mine literally until I'd got to the top of a welsh mountain, and that was after driving from Suffolk (though I've never been over 65mph in it). Since I fixed it i've heard it engage only a few times when at an off-road pit and only for a few seconds a time. I think it's possible to have one of these wrong a long time and not notice it until it's really needed and then it might be too late. Might save a few heads (and a divorce if the wife was supposed to be watching the gauge Biggrin ) I now get it up to temperature to check it hasn't re-seized as a service check as well.

How much?

I don't think you need to worry about mixing weights but if you didn't get all the old out and you added a whole bottle you've probably got too much in there. To check that try leaving the case coil side down overnight, if when you open it you can see any oil around, you've overfilled it. It works by pushing the oil back into it's reservoir behind the holes with centrifugal force, through (i think) 2 always open holes at it's circumference. If there is any oil that can't get back into the reservoir it'll be trapped between the edges and the coupling will be always on. With a few bolts, a bit of wood and a drill you could spin it up with it off the engine instead of leaving it over night. If someone has a dead one maybe they could cut it in half or knock a hole in it and measure the volume of the reservoir? Or fill it with a thinner liquid and measure that. Otherwise it's impossible to give an exact quantity.

It might be worth checking those side holes aren't blocked as that would keep it locked, but i can't imagine how they'd get blocked.

It is normal for the coupling to be engaged for a few seconds when the engine starts as the oil will have levelled out into the coupling through those open holes. If you rev it you'll hear a roar when it starts from cold which should then subside as the oil is forced back into the reservoir. I've noticed that on other cars as well.

Heavier weight oil will definitely NOT make the fan come on sooner or be always on as it is the coil turning that opens the gate to let the oil flow that sets when it locks. Heavier weights would only give a slightly firmer lock.

we used to tow a caravan

we used to tow a caravan with a reneault 19 1.9 diesel fitted with an electric fan. i fitted a switch accross the sensor in the rad, the temp gauge would start to rise on hills, even on hot days manually switching the fan on at the bottom of long devon hills would actually drop the engine temp noticably below normal before getting to the summit.a lot better than hoping the fan would cut in before the engine cooked. an electric fan with a manual override has to be better than a self locking coupling which may or may not work when it thinks it should and cant be overridden by a thinking driver , bit like auto hubs in snow, least you know manual ones will work before you're stuck.

Old old problem. New

Old old problem. New solutin. Thank you.

Any veiws expresed in this thread by me are purely from my own experience, and (sometimes) falible memory. Hope my comments help, but please don't take them as gospel.

Any veiws expresed in this thread by me are purely from my own experience, and (sometimes) falible memory. Hope my comments help, but please don't take them as gospel.