h a l f b a k e r y

meta:

account: browse anonymously, or get an account and write.

 user: pass:
register,

# Fluid Coupling Four Wheel Drive

An elegantly simple way of transmitting power to both axles.
 (+1, -1) [vote for, against]

Here we go:

Fluid couplings have seen much use in automobiles in transmissions. (Torque converters are a modified fluid coupling utilizing a third element.) However, to the best of my knowledge, they've never seen any use in a transfer case.

Here's what I propose: use a three-element fluid coupling to transmit power to each axle of a four wheel drive vehicle. Drive one element (the pump), and from this element, drive the other two elements (the turbines). Attach one element a piece to each axle.

With a fluid coupling set up at fairly tight tolerances, this would allow each axle to spin at slightly different speeds, though it would still mean that torque would transmit to each axle under low traction conditions. It would be mechanically much simpler (and therefore reliable) than the complicated viscous-clutch systems now in use, and would require much less machining than gerotor pumps as used in Jeep four wheel drives. Current production techniques are readily adaptable to make fluid couplings (how many cars have automatics these days? 90 percent? 95 percent?), and it wouldn't take much experimentation to figure out the optimal clearances to allow enough "slip" on each axle without losing large amounts of driveline efficiency.

Possible problems are: fluid couplings are somewhat inefficient; cooling this thing could pose a problem, and there is a likely fuel economy penalty. As well, without some sort of additional gearset, it provides no low range. The size of fluid coupling needed to achieve this would be rather large, and so it could pose packaging problems. As well, this would provide only a 50/50 torque split; though I think it could be adjusted somewhat by creating looser tolerances on one axle or another (albeit at another fuel economy penalty). I myself like 50/50 torque splits on four wheel drives, so that's not a problem for me.

Overall, I think it could be a simple, elegant solution to the problem of driving both axles and still allowing a vehicle to turn, though.

 — Dirty Luke, Aug 01 2004

Syncro.org http://www.syncro.org/VCTest.html
[elhigh, Jun 13 2005]

 Why not take it a step further and remove the driveshaft from the picture? Mount the pump to the end of the transmission, and large hydraulic lines to the turbines located at the pinion shaft of the axle. This would allow for any amount of suspension travel(limited only by the length of the lines) without worrying about u-joint angles and driveshaft extension/compression.

Low-range can be achieved by mounting a Klune-V underdrive unit to the transmission.
 — destructionism, Aug 02 2004

[destructionism], so, those turbines/hydraulic motors you'd be adding to the unsprung mass of the vehicle would weigh how much?
 — bristolz, Aug 02 2004

Getting rid of drive-shafts by the use of a hydraulic motor on each wheel is not new (see "Fluid Coupling Four Wheel Drive" in this category, and the Bateman 24M farm tractor).
I wonder whether [Dirty Luke] can explain how the two output elements of his device both receive drive from the single input element. A fluid coupling has the input and output elements facing each other, like two halves of a flattened sphere; the close relationship is what provides the drive.
 — angel, Aug 02 2004

 [bristolz] I don't know. It was just a thought. 4-wheelers dont really care about that stuff.

My idea(well, what I had in mind, anyway) would work much similar to say, power steering, with turbines instead. One hose to supply the turbine, one to take the fluid back afterwards. It's possible to have two separate hyrdaulic pumps for 4-wheel drive. Actually, now that I think about it, a gear pump would probably be the best bet.
 — destructionism, Aug 02 2004

 I really didn't have hydraulic motors in mind. They have been used in 4WD applications (though not in a car) many times over the years. They work well for some applications where a constantly variable ratio and large amounts of torque are needed, but they really aren't all that efficient, not to mention they tend to heat hydraulic fluid to a large extent.

 Instead, I was thinking more along the lines of a three-piece torque converter (without the stator reaction member). These same basic units (a fluid coupling, no stator) were used in some Chrysler cars of the late 1940's (Fluid Drive) backed by a two speed manual gearbox (some had provisions to shift on their own).

 What I propose is a three-element fluid coupling, consisting of one driving element and two driven elements. Buick used a similar concept, but instead used a "staged" principle whereby different pitched vanes took over drive of the vehicle at various speeds (the Dynaflow). Instead of this setup, though, I propose to use two concurrently driven elements from one driving element.

 The arrangement would look something like this: (|), with the parentheses being the driven elements and the straight line being the driving.

Again, I'm no fluid engineer, so I don't know what sort of vane arrangements it would take to make this work, but it at least seems feasible in my head. Let me know if it seems feasible in yours.
 — Dirty Luke, Aug 02 2004

Uhm, have you ever heard of a viscous clutch? That is a fluid clutch. The Subaru WRX (not sti) use it. So does the AWD Chevy astro van and awd blazer. You talk of viscous as it is different than a fluid coupling, but is in fact one and the same. If you are talking about a hydrostatic drive system like tractors and dozers use, that is cool also, but to heavy and contains to much loss for auto use. Great idea, but baked a long time ago... By the way, a viscous clutch uses heat to alter the coupling. The fluid used thickens when it gets hot from friction. This causes it to stop slipping. This makes a great setup...
 — dlapham, Aug 03 2004

 For years, I actually drove one of the pioneering American vehicles to use a viscous clutch full-time four wheel drive system, the AMC Eagle. I loved them. I am well familiar with all versions of the viscous clutch systems in use today, as well as the various other systems. This idea was actually inspired by the Jeep Grand Cherokee system which uses gerotor pumps to transmit torque.

Instead of using a clutch pack and special fluid, though, this system I propose uses three "turbines", one driving and two driven. No clutch packs involved, no special fluid. And it's most definitely not in use on anything today.
 — Dirty Luke, Aug 04 2004

 Luke, So this is basicly a torque converter attached to each driven wheel? Would you have a lockup converter that locks the wheels from slipping at a certian speed differential between the drivetrane and the wheel? Like current converter lock between the crankshaft and transmission imput shaft?

I suppose that these converts could be smaller and lighter since they are driving only one wheel. You could also do away with the converter between the engine and transmission since these convertes would induce the slip required for takeoff and stops. You could use an electric lockup and even an electric dual pitch vein (like chevrolet once used in the 70s) that interact with the transmission in order to lockup when cruising and unlock when pulling, again much like the converter now does between the engine and transmission. All this being said, I don't see the advantage of this over a viscous clutch. Doesn't it do the same thing while weighing more and using up more power?
 — dlapham, Aug 05 2004

Probably. And that's why it's probably not in use. I really like your idea of using a lockup feature, though - that's definitely a step farther than I had considered. I suppose if you really wanted to get fancy with a computer, you could develop adaptive strategies to lock up the couplings when the vehicle was going nearly straight and unlock them for turns in order to minimize slippage. You could even play into steering-angle and wheel speed sensors to calculate specific lockup strategies. The complexity goes on and on...
 — Dirty Luke, Aug 05 2004

 What is the status of using viscous fluids clutches where the fluid is altered using electical current?

I thought I saw womething in Popular Mechanics once that this was going to be used in this sort of application - to vary the torque transmission???
 — sklaus, Aug 13 2004

I wonder if ferro-fluid could be used.
 — Ling, Jun 13 2005

Ferro fluid is currently employed in some shock absorber applications. The cost of the fluid, however, is quite high.
 — BMCCUE, Sep 26 2005

 [annotate]

back: main index