Testing a Viscous Coupling

On the basis of what you propose, I'd go with "one wheel up" if you're looking at a clean (insignificant brake drag), nearly static (slow moving) viscous phenomena; where system friction and lubricant shear rate are not significant factors.

Thanks very much for the response. When you say "lubricant shear rate" - do you mean the lubricant in the wheel bearing & differential? I've been trying to understand the VCU and that is all about viscous fluid and shear rate between plates - but I presume in the context of this test you mean the lubricant in the diff etc.

Yes.

At any normal and reasonably warm ambient temperature there would probably be no noticeable skew in the overall phenomena you are looking to observe with your VCU component; if you'll not be subjecting the associated system lubricants to rapid shear during your test.

Once again, many thanks :)

Here is a test of the VCU.

https://www.youtube.com/watch?v=L0ytdkWyEsQ

It shows different weights being applied. Which gives some indication of how the VCU reacts under changes in load. The influences of the other drive componets would be about the same. They would effect the time but not so much the change with weigh changes.

Instead of a measured bar and known weights, may I suggest a dial torque wrench? The measured length and known weight in the above mentioned procedure creates a known torque therefor a dial torque wrench may provide the same results with a much simpler and possibly more repeatable setup

Thanks for the comment. Using a torque wrench is a great idea, and some people use one. Quite a few people don't have them though so the bar and weight concept is used as more people will be able to put something together. Mind you, not a lot of people have a 32mm socket either!

Some help here?

I did not view these before posting.

One Wheel Up Test Results - Viscous Coupling Unit VCU ...

Freelander(Land Rover) VCU (Viscous Coupling Unit) Test ...

Welcome to the forum.

No need to apologize. There are many non-engineers here.

Good luck.

Ya, like me. Hello New Zealand.

And me!

Welcome!

If this "Viscous fluid" is Non Newtonian, velocity is involved. Can this fluid be changed/replaced?

Yes I believe it is a 'Silicone' fluid that is Non Newtonian. The greater the 'shear' the greater the torque supplied by the unit is. The units are produced by GKN who do not say exactly what is in them...

http://www.gkn.com/driveline/about-us/Documents/datasheets/Viscous-engl.pdf

They quote in that doc...

• Alternating inner and outer plates are submerged in silicone fluid
• The relative speed difference between the plates produces a shear stress in the fluid filled gap
• The resulting shear stress transfers torque from one opposing plate to the other

The units are steel and sealed, the fluid can not be replaced. Commercial (and some home) reconditioners cut the units open on a lathe, clean the plates and insides of the old fluid, refill and weld back together. People without this equipment drill holes at either end, drain the fluid, tap the holes, refill and seal. The old fluid is often very reluctant to drain and resistant to thinning by all sorts of chemicals people put in to try and thin it.

People usually refill with "Polydimethylsiloxane" (is that Dimethicone? but I may be well off the mark there) such as...

http://www.ebay.co.uk/itm/Silicone-Oil-100-000-100000-Cst-400ml-Viscous-Coupling-Silikonol-Freelander-/231549816704?&clk_rvr_id=834161235171&afsrc=1&rmvSB=true

As I say though I don't know what fluid or 'additives' GKN originally put in them.

I agree that the simple "one up" test using known weights/torque wrench seems perfectly valid as a go/ no go test providing you are working from a practical and repeated start position for the weight. But have never had to do anything like this!

On the silicone fluid- this is something I am more familiar with, up to the 60,000 range. PDMS (yes, Dimethicone is a common trade name) fluids are usually distilled to certain endpoint, and the viscosity can be adjusted by mixing different ratios of different viscosities. So a 100,000 cst fluid has a range of higher and lower molecular weights that combined have an average test value of this viscosity. As you say these are sealed units volatilization is not likely a problem, but higher temperatures can also cause breakdown of the silicone fluid that could change viscosity; contamination may also have an effect.

If I had to clean one of these out, I would use a recirculating system. Drill and tap as you said. If you can get hold of some 100 cst fluid that would be best, otherwise odourless mineral spirits (also known as Isopar K or Naptha K) heated to about 60 C and recirculated through the tapped ports (warm the coupling as well)- at least 10:1, so 4 litres if the volume is 400ml, more is better. Use a compatible deadheadable pump with a pressure gauge; you would likely find the pressure will suddenly drop as it channels through. Let it run for an hour or so then drain well. This should clean the unit out far better than trying to drain.

Thanks JNB - I must have missed your post and am reading back over the thread.

The info is very interesting. If you don't mind I'll copy what you've said over to a thread on the Landie forum about refilling VCUs. If you're interested, here's a link to a thread containing peoples experiences. You should bare in mind people are not necessarily engineers - they people are trying to do the job in the garden shed with what ever tools come to hand and not really knowing a lot about the characteristics of the fluid they are working with. Its a long thread, so you may not bother or find it an interesting read - your choice :)

Freelander 1 DIY VCU service - Land Rover Zone

You are more than welcome to make use of it! As one of the many technical minded non- engineers here myself, I understand that some things just make peoples eyes glaze over....

I took a look at your link, up to page 7. You weren't kidding!!! Maybe I should make a few points:

1. I see there are a few different viscosities bandied about, they will definitely change the performance depending which is used. And also adding lower cst fluid would "loosen" up a unit, once it mixed in.
2. A big confusion on that thread is that heat causes the thickening. WRONG! All these fluids thin out as they are heated; non-Newtonian means they do not react in a normal linear fashion to shear forces, but either flow easier with shear (such as ketchup) or become more resistant to flow (which is what you want). I assume that these units have a fairly close tolerance between the plates, and the fluid- which is also quite sticky- layer between is not very thick. There is a formula for it, but basically you cause a shear force when the plates have a speed differential, and the thinner the fluid layer or the greater the speed differential, the greater the force. This shear force does create heat though, in a very thick fluid causing lousy heat transfer (in regards to the cooling comments). So I can definitely see that having a "tight" unit with different size tyres on constant highway use could cause issues...
3. Don't use ketchup to refill the unit.
4. If contamination or breakdown is a concern, air pressure will not clean the units out properly- they will channel and then take forever to drain anyways. Solvent washing will clean them out properly. Otherwise, it's probably simpler to use the air pressure and add back some lower cst fluid to loosen it up. With these viscosities, you can pump into through a zerk fitting but take it right out of the drain port. Even through a 2" drum bung, they still take forever to flow!
5. Finally, the 'looser" unit will have lower shear forces for an equivalent resistance, so for highway driving etc. there would probably be less wear on the drivetrain/tyres; there will definitely be less heat buildup for a given amount of slippage and the better fluid circulation may also help dissipate it faster.

lol, I thought you'd find it amusing reading through those comments.

I shall tell those that have added Ketchup that it is wrong! 'Ends of tethers' have been reached by many people trying to work on or even just understand these things. When this occurs it is usually after various arguments about what the fluid actually is and what CST rating to use and there will have invariably been many inputs such as 'fill it with XYZ' where XYZ might anything ranging from custard through baked beans to s**t! The the last post by the person at the end of his (I don't think we've had any VCU rebuilding hers yet) tether usually says, 'f**k it, I'll fill it with XYZ'!

Must admit, that I've been confused the the whole 'heat' thing. Reading various things about these fluids, it appears that it is the agitation that raises the thickening. There again I also see references to heat. I have been wondering if the heat generated is just a by-product of the thickening and/or the friction that this causes against the plates. You are right, the plates are very close together. With no fluid in the unit the plates basically free-wheel. With the fluid in they are restricted and the restriction increases with shear. Surely this restriction is friction and friction causes heat.

I have never opened a VCU, but I've rebuilt my IRD (front diff) due to transmission melt down caused by a tyre low on pressure causing the VCU to 'lock up' and trash it. When I took the oil out it was a grey/metalicy colour because it was mixed with the remains of the bearings being ground to a pulp. When I see pictures of VCUs that are being drained, the fluid often also has this grey metallicy appearance. This indicates to me that the plates have been worn through friction and contaminated the fluid. It think its this, and probably breakdown due to heat, that changes the fluid's properties and causes the VCU to become to tight.

I'll confer your comments to the group!

Yes, the heat is generated by friction- that being the internal resistance of the fluid to the shearing force. If you were to find a chart of typical shear forces through a laminar layer, you would see that it is parabolic with the highest forces at the plate surfaces, decreasing to a lower (often zero) force in the middle. These forces can be phenomenally high.

Also, PDMS fluids are naturally water clear. Now that I've looked a bit further and seen the stuff coming out of one, I'd be flushing it completely!!

Wow, those last 2 sentences are profound! I've read them a few times and believe I understand them. Even if I don't, they're fantastic sentences. Its a long time since I was at school, however I'm sure my English teacher told me never to start a sentence with 'And' or 'But'. But, even with that I shall copy those sentences and append them to many of my posts in the Landyzone forum :)

I would say starting a sentence with a "But" is pretty normal, even if some high school teachers don't like it.

As to starting a sentence with an "And", there are far fewer instances where it is considered correct, but it is still allowed under English grammar rules.

I had a Grammar school teacher (1958, I was 11) that did not allow "and" after a comma, let alone after a "full stop".

I think she would have had a fit if I had written one after a full stop!!! But I did not know at the time a source of infos (school library maybe?) to prove it either way.....I never even considered it!!

That's an interesting point, but I think you're well beyond the rheology of a simple one component fluid such as they're talking about here! Apart, of course, from the effects of contaminants or metal particles in the fluid.

I have used several clay based products before, manufactured by Huber, that have to be dispersed into a water based solution and then the addition of an alcohol would cause it to thicken. These formed platelets or scales that would stack up as you say, due to the interaction of the charge on the surface and the hydroxyls of the alcohol causing them to separate slightly. They would then resist shear up to a certain point, above which the "stacks" were pulled apart. After that the product would flow easily, gradually thickening back up again as the platelet stacks reformed. I don't have that much experience with bentonites, but what you say definitely sounds feasible with the different particle shapes.

You may find the .pdfs referred to in post #10 to be helpful. It does appear that a) usage raises the temperature, which makes the liquid thicker and b) also expanding the liquid till it "humps" (fills the cavity 100%) and it then makes a 100% power drive thru

The PDFs are from GKN, the manufacturer

A search on google using "Viscous-engl.pdf" will also find them....they make very interesting reading I feel - thanks to post #10, I believe he was the first

The units are in usage on several makes and models of cars, for the last 30 years or so

When they need replacement, the liquid is difficult to remove as it stiffens up with age! (how useful!!)

Best of luck

I have studied that doc many many times and I can see no reference to "usage raises the temperature, which makes the liquid thicker". There is no doubt that usage raises temperature, but the only reference in that doc referring to heat or temperature is "The 'Hump' mode is activated when the coupling achieves 100% filling due to fluid thermal expansion thereby amplifying a hydraulic throttling effect between the plates". This doesn't imply the fluid is getting thicker - just that it has expanded to cover the plates completely, and possibly/pressumably is under pressure.

JNB

I take on board your comments about using thinning agents, however one of the problems on these units is that the volume of fluid ( and hence remaining air space) is, apparently, critical to the VCU working correctly. The reason most peeps just try and drain the unit, is that they have some idea of the volume of "new" fluid to re- inject.

That is a good point; unfortunately, unless you have a very accurate scale to weigh it before and after flushing and draining I can't think of an easy way around it . It's much easier to check 500 ml +/- a small bit than to do the same before and after with 5 litres or so when some might be left in the pump.

One thing to remember with these units is that the Silicone fluid is actually a dilatant fluid. i.e it thickens under increasing shear stress. It is not "just " a silicone fluid - when the two halves rotate in synchronicity, there is no shear forces between the plates and the fluid remains (relatively) liquid. As soon as either side of the drum rotates at a different rate, the fluid heats up to circa 100C extremely quickly and the fluid solidifies.

Upon solidification, the shear action is removed and the fluid regains its liquid state. the fluid then oscillates between both states until a natural equilibrium between both halves is reached ( there is no difference between the wheel speeds front and rear).

Interesting problem. Given the complexity of the fluid and its change of viscosity when in motion, it would appear that neither of the tests is accurate given that they are both under more or less static conditions.

The only "true" way to test these things is dynamically; to uncouple it, put a variable speed motor on the input side, put a Prony brake or dynamometer on each output, and vary the loading on each "wheel" to simulate the change in force as the turn progresses, hardly a DIY proposition.

The only alternative I could offer is a two step (static) approach. Get a bunch of people to agree that when the easy, on the vehicle test shows the unit is going off normal, take it out and subject it to the harder to perform test, and see if there's any clear, distinct correlation between the two. If there is then you could use the easy test as a proxy for the more difficult one.

Another approach might be to examine the wear element; i.e., the magic fluid, by withdrawing a fixed volume of fluid from say a hundred different units with varying mileage, and subject them to a dynamic shear test, record the results, and see if or by how much, the properties of the fluid change as it gets "used".

Who knows, you might even gather enough data to start a class-action suit against the manufacturer for coming up with such a stupid idea not providing a fluid whose life expectancy is far below that of the vehicle, thereby diminishing its resale value.

^^^^^^^ I think the the liquid is not affected by motion, per se, as if both halves if the VCU drum are rotating at the same speed, irrespective of that speed, there is no change in the state if the fluid. Changes only occur ( and very quickly - circa 90 degrees of rotation) when there is a difference in rotational speed, thereby causing a shear motion between the internal plates. For a fuller explanation see the first few posts here- http://www.landyzone.co.uk/lz/f38/definitive-freelander-vcu-testing-thread-99163.html

That quick VCU test is good enough as a go/nogo test.

If it passes the quick test it's most likely just fine. They either work or they don't.

I LOVE the wheel up idea.....but reading on the net tends to show this as not being a good idea. Also, if the VCU is too stiff, it can take the rear axle out as well. See the second weblink below. That would be expensive.

The problem is to take a bad one apart and learn how to fix it!!! Maybe its just dirty and/or has lost fluid! Could be easily fixed maybe, rather than just replacing....Then it would be good to monitor the life of a repaired one against new......see second weblink.

You may want to review this article if you have not seen it before:-

What-does-my-freelander's-vcu-do?

Where near the end you can read the following, showing that even worn or different manufacturers tyres can cause similar problems to a worn out VCU and therefore they should be checked first. But if its getting bad, DO SOMETHING!!

Because the IRD exerts a slight gearing effect between front and rear drive, incorrect tyre sizes can alter this to produce the symptoms of a seized VCU, and to load the transmission in a similar way. The critical dimension is the rolling radius of the tyres, measured from the wheel centre to the ground, with the vehicle's weight on correctly inflated tyres. Because actual sizes vary between tyre manufacturers, it's worth ensuring the same make of tyre is fitted to front and rear. If differing makes are fitted, or just two tyres need to be renewed, ensure the tyres with the greatest rolling radius are fitted to the rear. This makes life easier for the VCU and the whole transmission.

If the car was driven with incorrect rolling radius tyres, especially larger ones at the front, it could be that the VCU was "over used" so to say and over heated over long periods, causing the fluid part to vaporise to some degree possibly, making the problems even worse....

If the fluid part of the special liquid is lost to some degree, it appears that the VCU will start "working" at a lower speed difference....maybe someone can analyse the liquid and find out whatis is and replace the missing, to return the VCU to working properly again.....could be a lot cheaper than a new unit (though I have no idea a) if they can even be opened b) if the fluid is available!).

A cheaper alternative, for those who don't need to go off road, is maybe simply removing the guts of the VCU, or taking off one of the prop-shafts to the rear. Not only cheap, but a fuel saving alternative due to less drag and no windup!!

From the same article I read this:-

The VCU can fail in two ways. It's rare for a VCU to fail free, (ie, incapable of putting drive to the rear wheels). If this happens, there will be no indication during normal driving, but if the front wheels lose grip you'll become stuck because drive won't be transmitted to the rear wheels. Apart from becoming stuck, this condition will not cause further damage to the vehicle.

This website gives some other good tips that mention that the VCU is 9 out of 10 times the problem source:-

Bell Engineering.co.uk

It is a common fallacy that VCUs seize. In all our years of experience we have never come across a seized VCU. What actually happens is the silicon viscous fluid gets thicker and thicker with wear and slowly causes the viscous coupling to become stiffer and stiffer to rotate. More and more strain is therefore put on the gear train and failure eventually occurs to the IRD and rear diff.

For people in the UK, a new VCU can be purchased for only £395 from here:-

Freelander-brand-new-genuine-parts-UK

It appears that they need replacing at 70,000 miles or so, not much miles really...... they are also on Landrovers as well it seems.....

Best of luck!

Thanks guys for the input. Regarding tyres - the need for as near to identical sizes and pressures is well known within the Freelander 'community' - people (eg me!) often join the community because incorrect tyres have already trashed their transmission - so unfortunately its well known after the event!

Driving with incorrect tyres is similar to driving with a VCU that is far to tight - but many times worse as it causes serious wind up even in a straight line, not just on turns.

As stated LR do quote a service change interval of 70K miles on the VCU unit - possibly to cover class-action suits! In reality, they can fail after mileage much lower than that or go on fine many times that.

So long as they are not to tight and the tyres are in good condition, they work very well as a way of giving good traction to a vehicle, and without the need for any driver input or stopping to change setup. Also at £200 for a recon unit, they are not a particularly expensive item to replace when its needed.

All things though come back for the need to know when they should be replaced :)

Withdrawing fluid and shear testing it would give an absolute result, but unfortunately its not practical. The units are sealed, so would need adapting to make this possible. I'm not sure if fluid escaping is an issue. In fact I'm not sure if anyone within the 'community' really understands why they go tight. Its obvious that the fluid is behaving differently - when drain holes are cut fluid will eventually drain from one that is not in too bad a condition, but very little will drain from a poor one. It may be due to changes in the substance due to repeat shear cycles - either force or heat, possibly contaminant from fine metal particles worm from the metal plates, age, a combination of these or other factors.

Considering they are such a simple unit, they are very interesting!

GrunpyGel, What, if any, method is used to cool theses units? Could this be improved?

Then they would not work as the heat makes the liquid "set" and pass the drive through....

Yah sure......

If you look here:-

freelander-problems-how-a-viscous-coupling-vcu-unit-fails

You will read the following:-

The other thing to check, particularly if your tyres may be the problem, is the temperature of the VCU after you have driven it for a few miles. The VCU should remain cool to touch - but be careful, if you have mismatched tyres it can become very hot - so don't blame me if you burn yourself when trying to touch it, you have been warned! If your VCU is getting too hot to touch this is putting tremendous strain on the drive train - DO NOT DRIVE YOUR FREELANDER ANY FURTHER!

Thrashing a liquid with vanes, warms up the liuid, its unavoidable.....thats what eventually causes the failure of the VCU! The liquid simply gets thicker and thicker.....

I looked there, it is saying that heat is detrimental, thus my question about cooling.

I missed the question!!

As you obviously understand, the more often there is a strong differential between input and output, the higher the heating effect....which is why using differently worn/size tyres causes even more problems with overheating of te VCU....that was mentioned in some of the links I posted.

The units are not cooled - other than being in the open below the car with air flowing over them.

There is a lot of talk about heat in the units. In 'normal' conditions they should be warm to touch after driving - then there is debate about what causes them to be hot or cold.

The fluid clutches I am familiar with have cooling fins cast into the housings. Perhaps the units could be force cooled. At least at low speeds.

I am guessing a bit, but looking at the web sites, it appears that tyre wear is probably the single most probably cause of the overheating. They need to keep the largest circumference tyres on the back.....easy to forget!!!

Therefore I think that Freelanders need an extra gauge in the cockpit, one connected to an IR remote (does not need to have contact! See NOTE below!) thermometer that shows the actual temperature of the VCU all the time the engine is running..........So that the driver knows the temp of the unit under all driving conditions....it may even be possible to build a suitable unit for easier installation....

I think some Freelander drivers are in for a shock!!

Of course with some minor intelligence added and a connection to either mileometer or GPS unit, you could look back later and see what terrain affected the temperature.....

Also, maybe it may be possible to predict when a replacement is needed.....

At least the VCU unit appears to be very easy to replace once the car is jacked up (though I may have put my foot in my mouth with that statement!!), and at 200 UK pounds for a re-conned one, bearable in price....

NOTE

You may find a suitable circuit here:-

infra+red+circuit+for+remote+temperature+sensing

Best of luck!!

Interesting. In another post on this thread I mentioned I'd like like to check the temperature of the bearing in the differential pinion for signs that VCU wear or incorrect tyre pressures is putting the transmission under undue pressure - this could be an option.

I believe the pinion casing is aluminium - do you think heat generated on a bearing pressed into the case could be picked up outside?

I'd like to write an app as well that tapped into the ABS speed sensors to report when wheels are turning at different speeds (eg loss of tyre pressure), but unfortunately my early Freelander isn't ODB2 compliant and I can't find any device that will plug into its ODB port that I can talk to.

May I suggest a PIC to count the ABS signals from all sensors, some are SOOOO cheap, you may find that its better to have a PIC (4 in all) counting only a single ABS sensor and passing the values onto another one via RS232 connection(s), which then calculates for you....

I am not fully aware of your requirements, so its difficult to make better suggestions.

Your comment:-

Its obvious that the fluid is behaving differently - when drain holes are cut fluid will eventually drain from one that is not in too bad a condition, but very little will drain from a poor one. It may be due to changes in the substance due to repeat shear cycles - either force or heat, possibly contaminant from fine metal particles worm from the metal plates, age, a combination of these or other factors.

I feel is addressed in I believe the second link I posted, the liquid is designed to heat up quickly and go stiff, this drives off some of the liquid as probably vapor (Guessing!)....it is simply a part that does wear out.....

I bet the first person who reworks them into a new unit that is switched by hand (or even better with a 12 volt solenoid!), like on old locking hubs, so that you switch it in/on just when you go off road...or even just when you get stuck!! But as a re-conned unit is only 200 UK Pounds, why bother?

So, GrumpyGel, now you can see why it is said thusly:

"How many engineers does it take to screw in a lightbulb?"

"One: Unless there's a meeting in Conference Room 4."

Best to you in your endeavors --

And how do the engineer(s) get inside the light bulb ??

In which case a long, heated discussion would take place about the proper method of removing the dead bulb, the proper method of inserting the new bulb into the socket, which hand to use to turn the bulb to tighten it, how tight it should be, how to check the tightness, what type of ladder to use and the proper stance to assume when twisting the bulb into the socket.

Somewhere after the discussion is more than half way complete, and mostly settled, Andy will enter and proclaim that his way is the only right and proper way to do it. All other ways are wrong because only he is the one here who knows exactly the correct way to do everything.

The maintenance man will have come in, used one of those long poles to unscrew the old bulb and put the new one in and none of us in the discussion will even realize that the light has been replaced because we were all too busy proclaiming that only we knew how to do it properly.

When you say " Maintenance Man ", are you talking about me ?

Are you the one who changed the bulb? We were all too busy arguing to notice.

No, I drank the rest of the coffee, grabbed the last jelly doughnut and left the empty box on the counter.

I wanted that doughnut!

How about having those who insist on bench testing first perform the "one wheel up" method prior (and ofter if they replaced the fluid) and measure any difference (if any). Is there a linear relationship such as the weighted bar drops 20% faster on the bench than on the car accounting for drag?

They may then be able to eventually gather enough data to state it as:

Anything under 30 seconds using the bench method is "acceptable".

Anything under 36 seconds using the one wheel up could be "acceptable".

BTW, has anyone on your forum asked the dealer service department how they check?

lol, I suppose you are right absolutely right! To prove it you would need various VCUs is differing states of tightness, (1) workbench them all, (2) 1 wheel up test them all, and presumably (3) 1 wheel up test them on different cars - to prove or disprove that each individual car's components doesn't impact the result to greatly. If we're thourough, we should also do the test at different ambient temperatures and before/after the car has been driven a while.

The reality is that we're all individuals spread geographically and only really know each other as 'GrumpyGel' or 'The Mad Hat Man' (I posted this link on our forum!). Personally I have met 1 other member from NZ (in Chch airport for a cup of coffee when he was passing through) and one from France (in a pub for a pint when he was holidaying here). So its really difficult to organise this sort of stuff.

With regard to dealer service departments, you are assuming they test them! If you are lucky they might recognise that the car has traveled 70K miles and suggest replacing it. The dealers are more interested in the sexier (and more financially rewarding) Discoveries & Range Rovers - they tend to have a lack of understanding of the uniqueness of the Freelander's transmission.

With regard to the previous posts...

LR owners must be wanabe engineers because our threads on the VCU pile on and on with differing views - they're interesting, but do tend to go round in circles - and of course everyone is wrong except for the person that posted a post - we are all experts in our little world and luckily our geographical spread has result in no personal injury claims :)

Similarly as well, comparing to changing a light bulb, the maintenance man can just jump in the car and get a feel for how much 'braking effect' there is when you reverse the car of full lock. No light (to much braking) and I'll change the bulb (VCU).

All the links people have posted here are well known within the 'community'. Similarly the videos (most posted by members). I do hope I haven't just cross contaminated our OCD regarding Freelander Viscous Coupling Units into another communty!

Regarding replacing the unit or adapting it so that it can be engaged or disengaged - this creates some of the most long running threads on our forums. Many times individuals have come up with an idea along these lines and proclaim "I'm going to make something", the threads expand, then die and nothing is every produced. As said though at £200 for a replacement unit, there is no real need - so long as tyres are OK and units can be checked - plus the VCU is a very good AWD system - better in almost all ways to a switchable solution.

My personal thoughts are that we shouldn't bother regularly testing the VCU - a better solution would be to test the temperature of the pinion bearing on the differential to see if its getting to hot and thus indicating to much wind up and wear - then investigate why. I don't know if it would work or how you could do it, but it might tell us that the VCU and/or tyres are causing problems.

I understand that a weight on the end of a bar of known length will give torque measurement but this test calls for timing. Andy has pointed out that temperature causes the fluid to get thicker. A 'test' carried out straight after coming off the road will be different to one done first thing in the morning on an unroaded car. I would guess that jacking the rear wheel is used as the 3.2:1 ratio spins the prop shaft quicker and magnifies the stiffness. This means the bar only needs to move a short distance, and is therefore more convenient than using the front wheel. I would suggest trying the front wheel. Back off the brake pads and turn the wheel with a torque wrench and get a reading. By measuring a known good one and a known bad one you can get a standard. This can then be used as a yardstick to pass on to others. OR ditch the torque wrench and use a 'T' bar and spin the hub as fast as you can and see if it definitely locks. i.e. you discern a distinct difference between a slow turn and a quick one.

It may turn out that a simple and good test is to simply jack a front wheel and spin it slowly for a few turns then suddenly try to spin faster. If it is easy to turn slowly then gets hard/locks when trying to go faster; all is good. Even better is that the wheel stays on for this test.

Jim

The front wheels turn the front diff, which has a 3.21 ratio drive via a pinion to the front prop down to the VCU - this way the front and rear props are turning 'at the same speed' so that the VCU does not need to slip (much). The fronts therefore turn at the same speed relative to the props/VCU as the rears. As well as the added complexity of the pinion there's more going on up front as well such as splines into the gearbox that are connected to shafts in there. So I'd presume there's more to pollute the tests.

Even with a VCU in the acceptable range of viscosity, you can not spin any wheels by hand - unless the car is raised. By hand, even with all your weight, strength and the leverage of the wheel, they will only turn at a very slow rate.

You are absolutely correct in saying that the temperature of the unit/fluid makes a sizable difference to the results of either testing with a weight on a workbench or 1 wheel up. If you take the car for a drive and test, then test at regular peiods as the unit cools, the timings do get longer.

If you take the car for a drive and test, then test at regular peiods as the unit cools, the timings do get longer.That's an interesting observation. It implies that heat thins the fluid. It means that another process is at work. Later posters talk about "stacking" of stuff in the fluid and indeed the .pdf talks about "humping", this may be the same thing.
Sorry about the front wheel suggestion; i got it that the front wheels were a 1:1 ratio. I couldn't rationalise the different turning speeds that would result. I just shrugged and thought that is why the VCU is there.
It would be good if the distance between the plates could be opened up to allow for ageing of the fluid. I did once pull apart a viscous coupled fan. I did so because it stopped working when it got hot. This ties in with your statement above. The fluid in the fan was still clear and very sticky when cold. This seems to imply that your guess at contaminants causing the lock up may be correct. Perhaps units should be fitted with magnets like the old diff plugs?
Jim

VCUs are consumables with a scheduled replacement interval like any component subject to wear.

I tend to think that this unit, being relatively easy to replace, especially in a shop, (and nobody has jumped down my throat saying that its hard to do), is just one of the many parts of a car that has a limited life span.

These include Oil, oil filter, tyres, brake parts, air filter, spark plugs, even exhaust systems if not stainless, to name but a few.....

One must not forget that it is well documented that once you start getting "wind up" for any reason from the VCU, you are putting other more expensive parts of the drive train at risk, not forgetting that the vehicle becomes less easy to keep on the road....(Landy with 4 wheel drive engaged on tarmac experience of mine!!)

So why do the Landy owners make such a big deal out of the VCU, are they just being just being extra cheap?

Bite the bullet and replace the unit with either a re-conned or a new one, whatever suits your pocket best. I did not even manage to find someone saying that the re-conned units have a shorter life, but as they are about 50% cheaper anyway (UK prices)........slightly shorter life is acceptable....

There are some other possible actions like:-

1) disable it and stop going offroad

2) add what is basically a locking hub instead, lock it when going off road, have a warning "light" or similar on the dash

3) Replace it with what the old Landys have, a clutch, operated by the driver, while stopped, to connect the front and rear transmisions together.....just like the original....., but have a warning "light" or similar on the dash.

4) what would appeal to me personally, would be to replace the prop shaft behind the VCU, with some sort of mechanical clutch as in 3), so that the VCU is still there, it is turned, but the fluid does not get "thrashed", therefore heating up and expanding, and getting "old" in the process. It will just spin and otherwise do nothing. You only engage it when offroad.....or maybe do this to the propshaft in front of it! Or to both? An engineering company could make a fortune with a properly developed version and here I am, just giving the idea away...... A proper version could even be electronically controlled, so once the driver has selected 4 wheel drive, it gets "mated" the next time he comes to a stop.....

The VCU's job is to only drive the rear axle if the front axle has one or two slipping wheels, therefore in normal usage, the fuel consumption remains that of a front wheel driven car, so over 70,000 miles, it is probably mostly paid for by fuel savings anyway....

Furthermore, "proper" 4WD cars have a central differential, a high cost item, that the VCU does for far less money and weight!!

At car shows, VW many years ago (90's?) demonstrated a Golf/Rabbit with the same system, climbing a ramp with ball bearings fitted to part of the track, showing the car moving over them effortlessly, without even a pause as the front driven wheels started to spin slightly.....very impressive even by today's standards....

I found a VW (rather long) advertising video from 1986, using the same sort of VCU, which states SPECIFICALLY (for the un-believers here) that the VCU action is due to the silicone fluid HEATING UP, which causes it to lock. Watch from about around 6mins 38 secs....NOW YOU KNOW. it gets hot!!

VW Video:-

VW Synchro Ad

I am sure that other methods are also possible to reduce the wear on the VCU.........so get your thinking caps on Guys!!

It appears that the system that we are discussing here has fallen out of favor at Landrover anyway, probably for the reasons we all know here and the modern versions have oil pumps that lock the front to rear drive when needed.

See here for some interesting 4WD stuff:-

VW-Synchro- and-4motion

Haldex_Traction

VW-my-tiguan-4-motion-to-2-motion/

Best of luck.

lol, I agree with almost all of what you say. Its is a great method of making a car AWD/4WD and it has a life shorter than the vehicle and as such is a service item.

Some service items on a vehicle will always exceed their service life - eg oil. Others may fail early - eg spark plugs. When a spark plug fails, you know immediately and, although it may take some investigation to determine exactly what is wrong, it doesn't take anything else with it - you just replace it earlier than expected.

Unfortunately the VCU is an item that can fail earlier than expected - ie 70K miles. Unfortunately as well, when it fails, you may not know and it takes other much more expensive items with it. Hence the need to test it regularly while it is working well to determine if things change and it should be replaced early.

I ultimately believe more transmission failures are caused by mismatched tyres rather than VCU failure. Generally members of the forums don't get failures, its 'new members' that sign up because its happened to them be told about the importance of matching tyres and to a lesser extent how a failing VCU can cause problems. However, members ability to keep their cars in good condition is due to the collective discussion about VCUs and the increased knowledge on how they work and how to test/monitor them.

Having spent some considerable time researching VCUs on the web (with several others here!), I am also of the opinion that not having the right tyres (which includes worn and wrongly placed) is the main problem for the VCU.

This is a very possible reason for an ELF (early life failure) of the VCU. Everyone concerned should try and avoid such failures.

That is why I mentioned in a post some days ago that an IR (remote sensing) thermometer, with a dash display, would probably be the best way to see what is happening with the VCU, especially while driving, just by reading the current temperature of the VCU. I even supplied some schematics that could form a basis of such a unit for a DIY person to make such a thermometer....

It appears to me that such a method, might have been a good idea for Landrover to have installed from new....

For instance, you are driving on flat tarmac, steady speed and the VCU is heating up for some reason, so either wrong tyres, or the VCU is already past its working life.

(My previous post today demonstrates clearly that it is the HEATING effect that causes the silicone to thicken up and passes the "drive" through the VCU to the rear wheels (in the case of front engined vehicles), or to the front wheels on rear engined vehicles! Both types were shown in the VW video)

If some sort of flexible tape measure is used to measure the circumference of each wheel, that can simply rule out tyre problems or not.

Assuming that the tyres are good, then all thats left is a shot VCU as far as I can tell......easy going....

As already mentioned, a shot VCU can take out important and expensive other parts of the drive chain - not good!!

Axle windup can even be dangerous enough to throw a car off the road on corners, as it tends to negate steering, depending upon speed and angle of the steering. If you have ever experienced it, you will fully understand what I mean....so its not just a cost item per se....

A couple of comments here.

It is the shearing action if the fluid which creates the heat. This only occurs during differential rotation of the 2 halves of the VCU. So a hot, or cold VCU is not necessarily symptomatic of a good or bad unit.

It appears that thickening of the fluid is what transmits drive to other parts of the drive train, causing its early demise.

The VCU can fail in either "seized" - is very stiff, or "open circuit" - no coupling. The latter is rarer, fail safe and not normally noticed. Some "repair" houses supply " no drive" VCUs - normally considered a "con" as to the unknowing everything looks OK.

The problem is no it, per set, the VCU, but being able to determine its state PRIOR to drive train failure.

Alternative methods of replacing the VCU have been suggested, and in some cases trialed, but are generally significantly more expensive than just replacing the standard unit..

As previoiuysly stated, I believe the triuck is to regularly test, and then replace if any significant change has taken place. Your idea, Grumpy, of monitoring IRD bearing is a known service procedure and would be very effective if it could be done cheaply enough.

And ( starting a sentence with and) yes, Landrover owners, particularly owners of aging vehicles are usually as tight as a ducks...... :)

i believe you are misunderstanding just how the VCU operates. If you watched the video from VW in Post #58 I believe, the first video, it actually states exactly how. I even mentioned the time on the video to start watching....you should look and listen. Try at about 6mins 50secs or so....

It states categorically that when there is a big in speed difference, (this is for a rear engined car!) the rear rotor turning faster, shears the fluid, the "shearing" of the fluid/silicone causes it to heat up and it becomes thicker, this thicker fluid "causes the driven plates to pull the others along with them, driving the second (front) rotor with it. I repeat again, this is for a rear engined car.

The opposite is of course true true for a front engined car......that is, the drive is then from the front to the back, but in each case the "shearing" of the fluid causes it to heat up and thicken...the heating is a requirement, no heat, no thickening, no drive.....

You actually get a good tip as when this fluid is "worn out", it stays thick, even cool!! Then its simply useless.....

The silicone fluid when in "use", reacts the opposite to the usual when heated, motor oil when hot gets thinner for example, but this silicone oil is designed to get thicker.....and increase its volume.

You may like this picture better:-

You see this, with an explanation here:-

http://www.google.de/imgres?imgurl=http%3A%2F%2Fwww.syncro.org%2Fsitephotos%2FVCdiag.JPG&imgrefurl=http%3A%2F%2Fwww.syncro.org%2FVCTest.html&h=208&w=350&tbnid=kYjFLVv1aRpnSM%3A&zoom=1&docid=hY8RPhuacPr-0M&hl=en&ei=_alhVeOaJofTU7SJgeAB&tbm=isch&client=browser-ubuntu&iact=rc&uact=3&dur=1828&page=13&start=274&ndsp=24&ved=0COkBEK0DMEw4yAE

Best of luck.

erm, he knows as well as anyone (outside GKN) how the VCU works.

I posted earlier based on my own knowledge and experience with silicone fluids- I have none with viscous couplings. If the link you post is accurate, it's probably not a straight PDMS. I will check next week, doing a viscosity test on 30,000 fluid at different temperatures and post the results.

Just an aside- 30,000 PDMS is like cold honey anyways, and just about as sticky. Looking at the diagram posted I can easily see it transmitting a huge mount of torque without any further viscosity increase, and due to the nature of the fluid the "clutch" action would strengthen with increasing speed differential.

What is different?

We are both saying the sàme thing. The shear forces induces a temp rise which then causes a change of state of the fluid, as per the temp\viscos8ty graph I posted earlier.

I've watched the video - my dad had one of those G2 Passats - the posh GL5 version :) People complain about Freelanders running 2wd getting wheel spin - they should try driving one of those! He also had a couple of G1 Passats, good cars.

That, what would now be called an infomercial, does clearly state that the shear of the plates causes friction which causes heat and the heat causes the fluid to "thicken".

That is not necessarily though the normal behaviour of "Silicone oil", see...

http://www.shinetsusilicone-global.com/catalog/pdf/kf96_e.pdf

We do not know what fluid goes into a GKN Viscous Coupling - so we can not say anything for sure!

I'm sure your concept of tracking the temperature on a VCU would give signs that there are problems with the tyres on a good VCU. I'm not sure it would show whether the VCU is going tight or not though. Another thing we don't know for sure is what causes them to go tight. Be it chemical breakdown or contamination through worn metal particles, or what ever - the 'normal' heating process though will not occur - for example there will be less shear in the VCU and more tyre scrubbing.

I'm not sure if any car manufacturers still use Viscous Couplings, the Haldex units you link to have taken favour. The Nissan implementation (possibly others) monitors axle speed differences in normal conditions and there is a warning light on the dash if tyres are not matching.

Our Freelander's primary function is as reliable family transport. Unfortunately my fishing and skiing come lower in priorities. When our transmission imploded, I rebuilt it as 2WD - for reliability and cost reasons. Unfortunately, I can not test any of these suggestions on my car! I do have my eye on another 'hack' Freelander that will be mine as opposed to the family's. Assuming I can actually do it, I will have temp sensors rigged in - I'll put one on the IRD & the VCU to see what I get!

You are not convinced when a world car manufacturer tells you what happens, you are not convinced when the people on the Freelander website tell you, your are not convinced by anyone....!!!

Naturally its your own free choice what you believe (Ghosts?), but I would say that to take a different idea is really going against simple physics and many knowledgeable people....

From Wiki:-

Dilatant fluids(Slightly edited)

Dilatant, or shear-thickening fluids increase in apparent viscosity at higher shear rates. They are rarely encountered.

A possible use is in a viscous coupling in which if both ends of the coupling are spinning at the same (rotational) speed, the fluid viscosity is minimal, but if the ends of the coupling differ greatly in speed, the coupling fluid becomes very viscous. Such couplings have applications as a lightweight, passive mechanism for a passenger automobile to automatically switch from two-wheel drive to four-wheel drive such as when the vehicle is stuck in snow and the primary driven axle starts to spin due to loss of traction under one or both tires.

You may find this link to be very informative with regard to temperature and viscosity:-

Viscous_coupling_unit

Where you will find the following mentioned under Design:-

Alternate plates are connected to a driving shaft at one end of the assembly and a driven shaft at the other end. The drum is filled with a dilatant fluid, often silicone-based. When the two sets of plates are rotating in unison, the fluid stays cool and remains liquid. When the plates start rotating at different speeds, the shear effect of the tabs or perforations on the fluid will cause it to heat and become nearly solid because the viscosity of dilatant fluids rapidly increases with shear.

Are hope that you are convinced now!

Simply put, this type of Dilatant liquid gets hotter with shear (I personally do not know of any liquid that does not increase its temperature with shear, though I never say "never"!), making it thicker...QED

If not, you will have to remain as not understanding from my point of view......as I have done my best for you.....

I'm not convinced either way. I'm not going to be convinced by VW promotional video nor the font of all inaccuracies that is Wiki.

If you look at post #44 in the second point I'm told by someone who appears to know what he is talking about, that these fluids get thinner when they are heated.

So do I believe you or him?

Don't believe me, just do some good reading and you will find out for yourself whether it gets thinner or thicker....

There are enough websites that say that when they don't work anymore, the liquid is almost solid....that might be a tip!! Its hardly likely to get thinner and thinner and then suddenly thick....

But, as I have already said, you must decide as to what you believe.....rightly or wrongly!!

But its obvious to me personally, that you have not read any of the links I have posted, or simply not understood them, or not wanted to......you can pick the reason, I have no further interest in searching for further good information.

I believe I have a good understanding of how it works, thats the important bit for me, nothing else really matters....

Remember the Bible, PBS!

Andy - can you state who you are replying to, as I find this thread rather difficult

Just find the thread number in the top RH corner. If there is a thread number directly below it, thats the one its replying to, If there isn't one, its not a reply to a specific post.

This one is #74, replying to yours which was #73!!....OK?

Grumpy....they get thicker. That ius the thing about dilitant fluids. They work in the opposite way to "normal" liquids.

Grumpy,

Please keep in mind no one has identified FOR SURE that these use PDMS from the factory. Based on what Andy is posting, I'd say they don't. PDMS is a commodity fluid, made a handful of big guys including Dow Corning, GE and Shinetsu. They (or their customers) also use PDMS for base materials to produce a raft of other specialty fluids such as amino or hydroxyl functionals. But let's take a look at the Shinetsu document you provided- which is excellent, thank you; I can probably use itr for my own purposes as well!

Section 3 (looks like I don't have to do the temperature test after all) states that PDMS has a lower viscosity rate change than typical mineral oils. It also shows both a table and graph of this, clearly proving that there is still a continual drop in kinematic viscosity from below zero to over 200 C from all these fluids, as well as a couple of different types.

Section 5 refers to the relatively high degree of thermal expansion, which corresponds with other posts.

Section 12 is very interesting. It refers to the thermal stability of these products in air, with an example that shows the accelerated breakdown at higher temperatures, causing crosslinking in air (and the gooping previously mentioned). There is a typo on the second 200 C, but regardless- IF I were refilling one of these with any type of silicone fluid I'd purge it with argon or nitrogen first.

Section 14- surface tension- indicates that it would spread readily between plates, which would be desireable here.

Section 19 shows that there is a continual drop in kinematic viscosity with increasing shear; VERY interestingly, the percentage drop is much greater with the higher fluids.

Unfortunately, Andy's link does not have any values to compare to this, just the graph showing low to high. It doesn't list what the upper or lower values are other than saying "solidlike", but if that is correct I would say this is not a PDMS as I personally would trust the Shinetsu document.

This doesn't mean the VCU's wouldn't function with PDMS fluid, just that the torque variation response would be different and there would also be no lockup point (assuming solidlike means literally solidlike, capable of transmitting all the engine power to the rear wheels- I'm sure you Freelander owners could figure out a way to prove this), so continued operation above this level could continue to raise the temperature and accelerate breakdown.

With a sample of the fresh OEM fluid I could easily test it and supply the results, but we don't have Freelanders here in Canada!

wow, you've done some serious studying there. Unfortunately its getting late here and my mind isn't keeping up with what you say, but a couple of things rang some bells.

"Please keep in mind no one has identified FOR SURE that these use PDMS from the factory" - this is true. GKN won't divulge what goes into their units. My brother works for and is quite 'high up' at Land Rover - and he won't tell me either! I'll have a dig around the German site referenced earlier to see if they let on to what they use in their units.

"IF I were refilling one of these with any type of silicone fluid I'd purge it with argon or nitrogen first." There's quite often talk of 'additives' in the GKN units, I'm not sure if that's them saying it, or 'us' throwing it out as a possibility. What you say, whilst not being an additive to the fluid as such, could be seen as such.

I'll have another read in the morning

Firstly, you are making a very invalid assumption, that all Silicone fluids are the same or similar.

The one used in the VCUs is "Dilatant" and cannot simply be bought, over the counter, as you appear to think.

As some of the Freelander sites bemoan, they cannot buy it anywhere....it appears to be only made for and sold to companies who use versions of a VCU!!

Do not mix it up with these other hundreds, maybe thousands of silicone fluids, from many different companies....which are usually formulated to give a stable viscosity over a wide temperature range....this stuff for VCUs isn't the same!!

Best of luck.....

Andy,

That is exactly my point. That the silicone fluids being discussed on the forum Grumpygel linked to were PDMS fluids, which I am familiar with and which exhibit certain traits throughout the "family" of molecular weights. While I can see them acting in a drag function that would definitely turn a viscous coupling, they are not a dilatant or shear thickening fluid- rather, the opposite. My apologies if my post gave the opposite impression, that was not the intention.

Therefore, either the information you and others have posted about the operation of VCUs is at least partially incorrect (which I am not saying! I know about silicones, not VCUs and my interest is usually in shear thinning not thickening), or the correct fluid is NOT a PDMS- or perhaps more importantly, at least NOT AN UNMODIFIED PDMS. Therefore any attempt to replace it with the straight commodity fluid could lead to incorrect- or at least a different method of- operation. Full stop. If anyone can provide a third alternative I would be very interested.

This is backed up by the data provided by a large manufacturer of these fluids, hardly an assumption.

Perhaps this link http://www.google.com/patents/EP1443097A1?cl=en will help- some more from our friends at Shinetsu!!

So it would appear that this particular patent is for a PDMS (or similar) fluid, that incorporates particulates to cause the dilatant behaviour; one of the named applications is for a viscous coupling.

The only remaining question I see relates to heat causing the thickening; as the only thing I personally know of that exhibits this behaviour reversibly is molten sulphur I wonder if that is a misunderstanding, or if there is yet another fluid in use!

Looks like a GA to unredundant for identifying this mechanism previously!

You've obviously got the bug to understand exactly how these things work!

I think you are correct in that some of the information put together is "least partially incorrect" - and definitely contradictory.

I described Wiki in post #70 as the "font of all inaccuracies". They have a page specifically about Viscous Couplings in cars...

http://en.wikipedia.org/wiki/Viscous_coupling_unit

In the first paragraph on that page it states "When the plates start rotating at different speeds, the shear effect of the tabs or perforations on the fluid will cause it to heat and become nearly solid because the viscosity of dilatant fluids rapidly increases with shear." The graph shown on this thread in post #35 contradicts this. That graph is taken from a German engineering company rebuilding VCUs for Opel cars. I believe that graph is from the machinery they use for testing VCUs. It clearly shows that when cold the fluid is "very" viscous and its viscosity drops as temperature increases.

That same graph then goes on to show a very sharp rise viscosity (the Hump which various VCU related material refers to) but attributes it to pressure inside the coupling rather than temperature.

We don't know what fluid the German company use. I've had a look through their website and can find no mention of it. I had a look at the many pictures as well to see if I could find a bottle or barrel in the background, but no!

I'm beginning to think the Dilatent properties are only in effect when in Hump mode. Technically Dilatent means it thickens with shear - is thickening under pressure the same thing?

I suppose also we should consider what the VCU can achieve in a transmission. In a "Permanent 4WD" transmission there are 3 differentials. The center differential distributes power 50/50 to the front and rear differentials and they distribute power 50/50 to the wheels on their axle. So all 4 wheels get an even distribution of power. With an AWD (VCU) transmission this is not the case. Basically the front differential (in the Freelander's case) gets all the power, the rear axle gets 'trickle feed' power. In the Permanent 4WD system though, when a wheel loses grip, it gets 100% of the power. In this scenario in AWD a front wheel has lost grip, so....

At this point the plates in the VCU are now rotating at quite different speeds. From the information in this thread I'd say there is no Dilatency (shear induced increase in viscosity) or temperature increased viscosity as such, the fluid heats up increasing pressure to the critical point that induces 'Hump' mode. This passes 100% (or possibly up to a maximum amount given the info from the German site) of power to the rear axle. The rear wheels will (hopefully) get grip.

From that point on various factors are at play. With the rear wheels gripping and the VCU in Hump mode, this brings the speed down of the spinning front wheel - as the axles are 'locked'. However, the front wheel would not have spun freely in the human visible sense. Below is some info I calculated and posted on our Landyzone forum, it wasn't challenged - but there again it might have got lost amongst other threads...

Lets say the fronts are spinning at 10MPH - that's 15KMH So....

15KM = 15,000M
15,000M / 2M (roughly tyre circumference) = 7,500 Wheel Revolutions Per Hour
7,500 / 60 = 125 Wheel Revolutions Per Minute
125 / 60 = 2.1 Wheel Revolutions Per Second
2.1 * 3.2 (roughly Freelander's diff ratio) = 6.7 Prop Revolutions Per Second

So if a front wheel loses grip and is turning at 10MPH faster than the back wheels, the plates in the VCU will be turning at 6.7 revolutions per second difference. That sounds to me like a massive difference that could heat the fluid to Hump mode almost instantaneously. There is no need for shearing Dilatent effects, the pressure effects will kick in immediately.

MHM has talked about an 'equilibrium' effect in the VCU where the temperature fluctuates rapidly (I believe) as the fronts lose grip, rears gripping and bringing the front speeds down, but they still don't have grip so speed up to once again be brought under control by the rears etc...

All this would tie in with a lack of need for dilatent behaviour - if indeed it is pressure within the VCU that is causing it to 'Hump'.

In the LZ thread, I have mentioned (cannot remember where I got the info from now) that the VCU "locks up within a quarter of a rotation - as you said, that is very quick.

You forget that the Freelander (at least) has ABS and would "lock" the spinning front wheel very quickly as well. The combination of ABS and the VCU enables the system to work very efficiently (when it is working as it should)

I am going to make some assumptions now!

If you go back to your very useful Shinetsu doc, you'll see that it contains info on both the thermal expansion and compressibility of the base fluids- both of which are comparatively high. I am going to GUESS that the outer casing is the drive portion with the internals driving the rear (is that correct??),which would optimize centrifugal force. So when you start off cold there would be a natural imbalance due to the fluid sitting on the bottom, which would then disappear as it flowed to the outwards- does a malfunctioning VCU ever show vibration in a quick accel from sitting, as thickened fluid might take longer to flow out??- and there would be a bit of a reservoir around the plates. It is possible that the OEM fill would only cover part of the plates at this point, and as the liquid is sheared it heats up and expands inwards to fully cover the plates and provide more driving area. This would seem to be the Hump action; as well, this will compress the air in the center of the VCU, which will then cause an increasing resistance to further expansion (see the fluid compressibility)- as well as pushing out past any defective seals (how often does this happen??).

Also, the thing to remember is shear is a measurement of the hydraulic forces caused INTERNALLY through the cross sectional profile of the fluid by resistance to flow. It is NOT a measurement of outside or interfacial speeds, although these are usually what causes the shear. So the tabs and perforations will enhance shear as they "grab" the fluid, but even adhesion to the side of the plates will cause this. So basically the speed differential of the plates causes the particulates in the fluid to jam up- think of the front door to your one of your local stores if the have a fantastic sale just before Christmas!!- with the internal shear also creating heat throughout the fluid (this is like tempers rising as everyone jams in!!) as well as the friction of the plates sliding past- solidlike does not mean it's welded to the plates. As this happens it expands and covers more of the plates, which will cause a reduction in slippage and shear, until it reaches an equilibrium (unless the engine is too powerful for the VCU), like everyone getting inside the store and calming down; this continues until the shearing force is removed. The exception here is that that heat is continually produced during this process, and the temperature of the unit will continue to rise if it is not removed fast enough. So heat is a side effect of the operation, not a trigger (otherwise there would be a noticeable response difference between driving in the Arctic and the Sahara. The pressure also has relatively little to do with this other than possibly limiting the continued unrestrained expansion of the fluid, which would keep the volumetric ratio of the fluid/particulate more constant and therefore also keep the dilatant effect more consistent.

So if this is how it works:

1. Dilatent behaviour is what causes the VCU to "lock up", not heat. Regardless, we know that heat and dilatency are both caused by shear; heat does not cause dilatency through the discussed mechanisms. Note that heat is an instantaneous product of shear; it's essentially internal friction.
2. Hump mode is caused by heat, with the expansion allowing for an increase in locking potential. This would occur after the initial locking effect, or at least after a degree of shear that would create enough heat even without triggering the dilatant behaviour. OR I am misreading the definition and Hump mode relates to the viscosity "hump", not the degree of plate coverage. In this case it simply is controlled by the amount of shear, and is the critical point of 1). Please clarify which it is!
3. Pressure has nothing to with the dilatency, other than possibly controlling the fluid characteristics.

Casing : Yes, the outer casing is welded to the output shaft.

Vibration : I don't think there have been reports of vibration accompanying a failing VCU, but I may be wrong. My one is virtually siezed solid and is still fitted to the car and I do not notice any vibration from it (the connection to it within the front diff is disabled, but the rear wheels drag it around). The VCUs are fitted with dampers on the diesel & 4 cyl versions, but not the v6.

Fluid pushed out : Once again, I don't think this has been reported or at least is not a regular occurence. However, its highly likely that unless it left residue on the outside, it wouldn't be spotted anyway.

So if this is how it works : I don't think the Hump can simply be more plate coverage. There are too many references to the Hump creating a dramatic increase in "lock-up". I would even go so far as saying that until the Hump, the VCUs do not put out any greater torque - it may increase some, stay the same or possibly drop. Only once the plates shear at a sufficient rate to create the Hump - thats when you start getting a real increase in power to the back wheels. I may be wrong in this - but the more I read (and its difficult to balance all the input) that's the way I'm tending. I'm also convinced that pressure does play an important role. I've also just stumbled back upon this page...

http://www.easy2design.de/stuff/visco_sae.pdf

If you look at section 3.5 it discusses thermal expansion creating pressure. Then again in section 5.3 it states "if the pressure is relieved, or if, due to a leak, it is unable in the first place to build up, then there is no stiffening ("humping")"

I haven't read the whole document, but there is a "Conclusion" section 8. It summarises (if I read it right) that the humping is actually the plates being forced together.

JNB, The "incorporated" or inherent particulates are what I am curious about. To simplify for my own benefit, What would this fluid look like at a microscopic level?

My thinking is that there is a change in particulate shape and thus impingement, caused by breakdown. This impingement could also cause a change in the plate surface due to abrasion. This would cause reduction in intended performance even if the fluid where replaced. Note: Just a theory.

Off this topic: There was an early echo sound effect device that used silicon oil with ferrous particles, one stationary and one rotating plate, with magnetic transducer and pickup. The oil was called "Miracle Oil".

Guess how it worked

Thanks for the GA!

That's interesting!!

Regarding the structure- most of the time when working with these formulations I'm using emulsions, so this is a bit different. Based on the Shinetsu patent, these are slightly above micron size (which usually seems to be the upper limit for colloids), but are sized so as to minimized settlement. It doesn't mention density but I would imagine this would be a concern for centrifuge action in the VCU, so the closer it is to the fluid density, the better. Any clay based material or similar cationic would have a dispersive effect that would help, and the patent also mentions a dispersive fluid or treatment.

Going by this, I would expect to see a distribution of sizes from the 2 to 30 micron level, probably like a flattened bell curve with the production methods listed. So you would have a random arrangement of these spherical particles, although a microscope will likely show them to be rounded crystalline as opposed to actual spheres. Concentration would of course depend on the fluid ratios.

The particulate materials listed in the patent do have a range of physical properties, and there are probably others here that could explain them better than me. If the were ball milled as mentioned, that tends to reduce any embrittled areas quickly as well as removing corners and sharp edges. This would give an OEM fluid that already has maximum "hardness" and minimum abrasiveness for the particulate matter, which would reduce both abrasiveness to the plates and further particle size reduction. Having seen what wear does to stainless steel emulsifier discs though, I can attest this still does happen! Note though that these discs typically run from 3,600 rpm to 10,200 rpm, so if the plates are made from a hardened alloy steel at the sub 1,000 rpm differential Grumpygel referred to previously this probably wouldn't be much of a concern for the plates and rebuilding. There could even be a shot peening effect. Fluid contamination from it would seem to be the greater issue to me. Over time the average particle size will be reduced as well, especially in a unit that sees repeated severe usage (usage being the amount of time operated in the locked mode).

Now that's interesting!

From reading only, not firsthand knowledge, it would appear that the fluid in the VCUs ages, and changes "back" less easily when old and used up....which is why the VCUs have limited working life!

Some speak of it almost being solid when it fails to release and you get "Windup"....a thick gel probably.

Also, the Swiss company pdf shows what happens if it gets burned.....I am assuming that can happen with a tuned car and lots of fast starts on loose ground, or maybe any car when the tyre sizes are wrong. But I am just guessing!!

From reading around on the net, the correct "Dilatant" fluid is relatively hard to come by, I have not found a company willing to deliver it to anyone.

Filling it with "any" silicone fluid may be a way of stopping the VCU ever locking up again ever, which might be a way to get 2 wheel drive, relatively cheaply!!!

Though I am interested in such technical possibilities, I myself do not have a car with a VCU. But I find it very interesting to learn more about any technology....which is why I really like CR4!!!

Learning that the maximum amount of power transmitted through the VCU is not only a function of the fluid itself, but also of its temperature at any particular time and the volume of fluid inserted/used, was a bunch of pretty interesting fact(s).

Reading that the customer can order them filled from that Swiss company, to transmit a certain maximum power was alone a very interesting fact on its own......that had not come out anywhere else, so the Freelander amateurs, trying to repair one of these units, are actually possibly creating a dangerous situation for themselves....where the front to back power balance could be badly and wrongly distorted.....assuming they get the right liquid of course!!

So if nothing else, the ones reading here, with a VCU on some car/truck, will have learnt that.....plus the fact that a properly made and setup VCU will probably save on fuel costs and tyre costs and is at the end of the day, is relatively cheap to replace.

But I do sincerely feel that NOT having some method of remotely reading the temperature of the unit while driving, particularly on flat tarmac (the temperature when off road could well be higher - correctly), is bad news. You get no indication that maybe the tyres are causing a problem, maybe you even wreck your VCU and do not know why!!

It need not be expensive or difficult to build/install......but it could save a lot of extra wear and tear, which of course equals money!!!

Also, having driven the old Landrovers, with no VCU or middle differential, AND occasionally forgotten to switch back to 2 wheel drive once back on the road, I know what "windup" is! It could even throw you off the road on a corner, it nearly did me once.....luckily I only went on the grass, but if that had been a mountain road, with a sheer drop!!!!!! Who knows?

I have heard of this happening, never happened to me but here are a few guesses (someone else here will probably "know" far better than I do!):-

1) Different fluid, maybe a different manufacturer that when it gets old, gets thinner and thinner with age? Easy to test by warming it up maybe? I would test with the engine NOT running, by spinning it with one finger occasionally and warming up the middle with a hair dryer or heat gun. It should get harder and harder to turn against the stopped engine.....

2) Loss of some or all of the fluid?

3) One "side" of the unit is mechanically damaged or broken, so that even when warm, with the thickened fluid, no "braking" effect is noticed.

4) Maybe it does not work in the same manner as the VCU does? Some are electronically controlled for example!

Looking on Wiki here:-

Cooling_Fan_clutch

It appears that there are some further fundamental differences not well explained:-

Types

Most fan clutches are viscous or "fluid" couplings, combined with a bi-metallic sensory system similar to that in a thermostat. Some clutches are electronically controlled (instead of bi-metallic strip). These provide the potential to control the level of engagement depending on any number of inputs. Common controlling factors might include engine oil temperature, transmission oil temperature, coolant temperature, AC system pressures and ambient air temperature.

I hope this helps.

Thanks for replying. In my case the fan was on a 2T truck and replacement was $A700 plus postage from Hindustan where the last one in the world was located. It is so common for the fans to fail here that the fix is well known. To whit; remove radiator, drill a 25mm Tek screw straight into the coupling. This gives direct drive.
I am not keen on bastardising OEM tech so i dismantled the coupling and saw two plates with grooves in them and a clear very sticky goo. By the way, when cold the fan is difficult to rotate but not so tight as to enable engine turn over by moving the fan. I do recall seeing a 'reed valve' which i know now to be a bimetallic strip (thanks ). Once the fan was hot it was too easy to turn over causing overheating. This is the opposite of the VCU fail. Of course it is possible that MORE fans failed to the 'stuck on' mode but because that doesn't show up as overheating it goes un-noticed.
I just realised that my fan was sticky cold and less so when hot AND the OP has observed the VCU is stickier when cold than when hot. This may be the simplest way to test. I don't have a car with a viscous fan in it so i can't tell whether a good unit is easy to turn cold and harder to turn when hot. If the answer is YES then that is the easiest test of all. Simply jack up A wheel and turn it, then drive around and jack up the same wheel when hot and turn it. Harder = good Easier = bad. If it is very hard to tell, a torque wrench and set procedure may be needed. Such as; place socket on a wheel nut so that the torque wrench passes over the centre of the wheel hub. Thus maintaining a constant radius. Read maximum torque required to move wheel in both cases.

I don't understand what is going on here but it may be that when the fluid is stuffed it behaves as most other fluids ( thin when hot ) UNTIL shear is applied, then it rapidly Humps up, ramps up, and locks. This may even be caused BECAUSE it is thinned. Just guessing.

Jim

The VCU liquid is thin when cold, thick when hot and when aging, it "fails" where it remains almost solid.

It is a "Dilatant" liquid (note that the the CR4 spell checker does not know the word either!!).

Lots of misunderstanding here, mainly from the non-Engineers as far as I can see, you are not alone!!

With regard to the engine fan, I would simply remove it (I have actually done this years ago after one got broken, the plastic simply flew apart), then go to a breaker's yard and buy an electric cooling fan and the cowl, and add them to your car's radiator. You can add them in front or behind the rad, depending upon space and the direction of rotation of motor/fan blades.

Generally the normal 12 volt motors (my experience) have separate fields, so to reverse rotation, reverse either the commutator or the field (not both). Check that the brushes run with little or no sparking, especially if you have reversed it.

Maybe some modern motors in cars are brushless, but may then be more difficult to reverse, but I cannot offer any helpful thoughts on that....

A front engined, front drive car is more likely to have an electric one, though many modern cars have them as far as I am aware.....Note the exact make and model just in case you ever need a replacement!!

I simply kept an eye on the temp gauge and switched it on when required, usually when in slow or stopped traffic.....while driving it was never ever needed and although I never ever checked it, I would guess that I saved some fuel through that change....But that was not a hot country per se (UK).

But a friend of mine here actually added two such fans individually controlled, (I don't know if automatically or not!) for driving to Turkey one summer. He also said afterwards that he needed them rarely.....

Don't forget that the engine thermostat, if working correctly, will keep the engine at the right temperature even if you leave the fan on longer than needed, you are simply wasting a small amount of electricity/fuel.....so do leave it on if there are any thoughts about whether its needed or not.

Also, nowadays, it must be really easy to automate it's control far easier than when I did it probably 40 years ago.....electronics today are fantastic!!!

Do pick good quality motors as they usually turn while driving and are wet in rain and frozen in winter!!! So do not forget a fuse!!

Best of luck.

These units definitely work by the fluid in them thickening - of that there is no doubt.

I'm convinced that its a Dilatant fluid, however, I have not been convinced that heat causes thickening.

In post #69 there is a link to a definition of Dilatant fluids...

http://en.wikipedia.org/wiki/Dilatant

That page makes Zero references to "heat". It makes 1 reference to "temperature", it is an equation that is far to complex for me to understand, but I wouldn't mind betting that an increase in temperature will reduce the viscosity.

The first sentence on that page quotes "A dilatant (also termed shear thickening) material is one in which viscosity increases with the rate of shear strain. " There is, as I say, no reference to heat.

So I wouldn't mind betting that when front & rear axles run at different speeds, the fluid thickens due to shear between the VCU plates, this increases torque transfer, but friction also creates heat which will reduce the amount by which the fluid thickens, thus reducing the torque transfer gains. As said before though, the heat will also cause expansion, thus covering more of the plates (GKN's Hump mode) and once again improving the net torque transfer. So in no way will heat increase viscosity/thickness.

There is also a difference between working and failing VCUs. In failing VCUs the fluid is to thick - once again, there is no doubt about that. However, it is to thick regardless of temperature. It may be due to chemical breakdown, contamination, going square, going round - I don't know - but it is to thick regardless of its temperature.

In fact tests have shown that on a failing unit, increasing temperature reduces viscosity - this is an example using the 1 wheel up test...

http://www.landyzone.co.uk/lz/f9/vcu-data-280972.html#post3455422

This may be impacted by other factors, such as temperature of oil in the differential, but not enough to confirm that the VCU/fluid gets thicker the cooler it is.

Grumpy - have seen the graph above which shows quite clearly shear force against temperature. You can see that shear force\ viscosity reduces with heat, up to about 100C then ramps up extremely quickly. This takes less than 1/4 turn.

Hi MHM, I've seen the graph on Landyzone and recognised it on this thread. I've not known where it came from. I've just done some digging to find it - well maybe not that exact one, but similar graphs from the same people for different VCUs. I'm looking at ...

http://www.dorfbrunnen-garage.ch/englisch/pdf_e/viscokupplung_e.pdf

They don't attribute this 'hump' in viscosity to heat (per se) - but the increased pressure within the unit caused by expansion of the fluid.

I can run with that :) As I said in #52

So we're all right and all wrong.

Reading that article - a few interesting similarities, but also significant differences.

They state that they can change the max torque being transmitted through the VCU.

They state they can change the "split" ratio between front/rear from 2/3 to 1/1.

They state that the shearing action causes "warmth" which increases pressure, which causes the increase in viscosity. Now, I am not sure I, necessarily, agree with this, but it would account for why it has been stated elsewhere that the fluid volume ( and air gap) is critical to the VCUs working characteristics.

Yeh, I was wondering the same about the tuning of them along those lines, I presume that's what GKN is talking about in their VCU doc.

Knowing how easily these VCUs go "out of tune" and tighten, it does make you wonder whether those specific statements relating to power split can be realistically achieved though.

Actually they do, you need to read the text on the last page as well as looking at the last graph of temperature against torque. Heat is mention as "Warmth" in the translated English version, "hot" is also mentioned at the point of when maximum power transfer takes place as the liquid expands....

You can even see in the graphs that "hot" is at a temperature of around 160°C, then the unit is effectively locked. I feel that the graphs are largely self explanatory , and you can see the effect quite clearly..

(The original German document is slightly more specific, its linked below if you can read or translate German...)

The company ( Dorfbrunnen Garage in Switzerland), correctly say, that the VCU was set up originally for standard cars (various) made by Opel, for up to a specific amount of power transfer, not a (percentage), to the rear axle. This is a max between 68 to 90 Nm..(probably depends on model, but it does not say that specifically in either document)

With a tuned engine with more "horses", that max amount does not change automatically. The VCU has to be remade/refilled if another specific up to a maximum amount of power needs to be transferred to the rear axle.

The customer can specify up to a max of 50:50 front to back power transfer by them tailoring the filling of the same unit, at his/her wish..

You can read most of this in the English pdf yourself, I have underlined the relevant parts and added in parentheses, what I feel is probably the more usual English wording, after reading the original document through:-

With a normal visco clutch (in hot status)(160°C) only (between) 68Nm to 90Nm are transferred (to the rear axle). It's not correct (to assume) that during increase of power of the engine (tuning) that (extra) power will transfer to the rear axle. Because the visco clutch (will) still only transfers the (same max) 68Nm to 90Nm. Therefore, also the visco clutch should be adapted (tuned) to the new (engine) performance (torque = Nm).

With (When there are only) small differences in rpm between input rpm at the sun wheel and drift to the drive shaft the back resistance is overcome by the small (lower) viscosity of the silicone liquid in the form of light slip (slipping).

With larger differences in rpm (front wheels slipping or possibly wrong tyre sizes) the silicone liquid (fluid) between the lamellas is cut (sheared). Thus warmth (heat) develops and the pressure in the housing of the visco clutch rises.

By (Due to) the increase of (in) pressure(,) the viscosity of the silicone liquid increases rapidly, that means, (that) the silicone liquid can (only) be cut by the lamellas with more (greater) difficulty.

The visco clutch begins to lock. (in the German version it says a bit more than that. It says, in my words) "at the Lamella,(input to output) a power transfer takes place, even though they are not actually physically touching!)

In German "An den Lamellen erfolgt eine Kraftübertragung, ohne dass sich diese direkt berühren."

The English translation was not made by someone with English as a first language, but its still fully understandable I feel, even without my added words.

The original German document for those who can read German can be found here:-

http://www.dorfbrunnen-garage.ch/pdf/viscokupplung.PDF

I hope this clears up any problems.

The company has quite an extensive and interesting web page in English here:-

Dorfbrunnen-Garage

Look under "Transfer box" for further infos for various models of VCU for Opels. This company are nowadays the sole manufacturer of this unit for Opel models, so I tend to feel that they know what they are doing and how the VCU and its fluid REALLY works.

It appears to be at least physically, as good as identical to the Freelander units, even if it does not transfer the same amount of power....

Have a great day, I am!!

We have a Bank Holiday today and we are going to a new Turkish restaurant this evening with good friends!!

Afiyet olsun. بالهنا و الشفاء! 慢慢吃 (mang-mang sik)

I hope the Chinese one is not a forewarning!!

Good info Andy. Its interesting that they focus so much on power distribution. They presumably use their couplings in these Opels for slightly different reasons than when installed in Freelander. Both would have safety requirement benefits, but I imagine the Opel implementation is also after (speed/acceleration) performance benefits compared to Freelander's off-road benefits.

Enjoy your Turkish meal. I had a Turkish meal the weekend before last - a Kebab after the Spandau Ballet concert in Auckland :)

I love Turkish food too, my Father served in the British army in WW2 and got sent to Turkey by Churchill for over 4 years!! We as kids were taken by our parents usually to a Turkish restaurant, so he could keep his Turkish fit for use...

He spoke fluent Turkish, sadly I don't, big failure of mine.....

He worked, after retirement, for the UK justice system representing Turkish lorry/truck drivers with badly maintained vehicles....

If you haven't been there, try to, the people are really great and the old ruins and the like are fantastic....it gets hot in summer, which I love! Over 50°C is often to be found....

Take a sailing holiday there on a big yacht for a week, very cheap and brilliant....away from lights, the stars are almost touchable...I could wax on for hours!!! Sorry!!

I have been to Istanbul a few times and found it a fantastic place. The Egyptian Bazaar was awesome and dining on the shores of the Bosphorus, very memorable!

I have visited the other side twice, I am hoping to get to Istanbul the next time....