Solved School Me (and you?!) On Viscous Couplings (BW4404)

[gavin]

What is a viscous coupling?

The viscous coupling is a 2-piece part, filled with a viscous fluid and clutch packs, which is pressed together and sealed to prevent any fluid from leaking.

What does a viscous coupling do?

It provides variable output-power to the front output shaft, the amount of power dependent on how much slippage there is in the rear.

How does it work?

Under normal conditions, the inner and outer portions of the coupling are spinning at the same speed. When there is slippage in the rear axle, the rotation of the 2 parts change; since the front tires are spinning slower, the inner portion is rotating slower than the outer. Because of this variation, the viscous fluid inside gets warm (due to clutch packs "mixing up" the fluid), and becomes less fluid, and more solid. As the fluid thickens, it creates friction, by putting more force on the clutches inside the coupling. This then causes more power output to the front tires.

What are the common signs of a bad viscous coupling?

Unfortunately, there are no truely common signs. Also, the signs differ depending on which way the coupling fails.

There are 2 real ways the coupling can fail:
First, the seal can blow, allowing the fluid to leak out and into the rest of the t-case.

Second, the coupling can "sieze" up. Basically, the fluid never liquifies as much as it should, so there will always be more power output to the front driveline than originally designed by the OEM.

How do I know how my coupling failed?

In the first situation, you would notice it failed when you changed the transfer case fluid. You would also lose all power output to the front driveline, since the fluid is lost, no friction can be made to put more power to the front.

In the 2nd situation, it's a little harder to tell. Since there will always be more power output to the front, it can almost be like a driving a manual t-case in 4wd all the time. We all know this is bad on the driveline. Yes, it can cause the driveline to bind. This driveline binding could cause vibrations in the vehicle, regardless of speed.

As you can (kind of) see in this picture, viscous coupling is the large part that is sitting in the huge "cup" that goes to the rear output. The gear right behind (above) the viscous coupling, is the gear that connects the front output. The viscous coupling, gear, and bearing are also on a separate shaft than the input shaft.

The rear output is directly connected to the transmission output, via the t-case input shaft which has teeth at the rear output end, which fits between 4 little gears on a larger gear assembly.
In this pic, you can also see how the viscous coupler, gear, and bearing are (fairly) press-fit onto another shaft.

This is the just the viscous coupling.


Is there any way to test a viscous coupling?

From what I've found and read, there just may be. You would need a very flat, solid, smooth surface. You will also need a couple chunks of 2x4, and an easily-rolling floor jack.
What you need to do, is jack the rear tires off the ground. Stick a 2x4 in front of the front tires.
Start the vehicle and put it in gear. Let off the brake. At idle, the truck should not pull itself over the 2x4's with only the front tires putting power to the ground. If it does, that means the coupling has seized, and will need replaced. It will (could) be causing driveline bind which can harm other parts of the driveline. (this could be inaccurate, as this "test" I found on a site regarding the VW Vanagon which, under normal conditions, only provides ~5% power out the front)

You could also test it out on surface with less traction; ie, dirt or gravel road. If you accelerate quick enough to spin the rear tires, you should see the front tires also start to spin.

Another test, if at all possible, is during winter. If you can get in a situation where you can get the rear tires on ice, but the front on a surface with traction, you should still get the truck to pull forward.


Why does the truck creep in park, with the front driveshaft removed?

The vehicle is able to "creep" in park, without the front driveshaft, because without the shaft, the viscous coupler is able to "slip." When you have the front shaft installed, to get the vehicle to roll, the viscous coupling is moving as a whole. That is, the inner and outer portions rotate at the same speed. This cannot happen because of the parking prawl in the transmission.
Without the front driveshaft in, there is nothing keeping the coupling "together." That is, because the inner and outer portions of the coupler are not held together (essentially by the front and rear differentials), the coupler can slip. 1 half of the coupler can rotate without the other half rotating also. This slipping happens by design, of course, because the slippage is what causes the fluid to heat up and basically "lock" the 2 halves together.

NOTE:
From what I've read, it is more common for a viscous coupling to seize up, rather than leak.
Also, 99% of the information I found, was all based on the Volkswagen Vanagon. So I have no clue how much will actually compare to the Explorer's BorgWarner 4404. Although I can say that all viscous couplings have the same basic parts and same basic working-concept, so it should be similar enough.

Where I got most of this info:
The Viscous Coupling
Viscous Coupling and Couplings In The Vanagon Syncro T3
How Differentials Work
Viscous coupling unit - Wikipedia

I also got some of the info based on the Jeep's Quadra-Trac which also uses a viscous coupling for AWD-capability.

Only last note: I've also read many many times that these viscous couplings do have a limited lifetime. I believe the Vanagon said to expect around 100k miles or so. And apparently some VW 4-Motion cars have a life expectancy of around 40k miles. (Technical Q&A | Porsche Club of America)


There haven't been too many posts on the viscous coupling, and especially none with a whole lot of real info. And due to issues with my truck, and the viscous coupling being the only thing that has not been replaced, I did a lot of research. Figured I might as well share my findings for anybody else who may find this useful

 

[IZwack]

Hey fantastic writeup gavin -- made it a Sticky!

 

[gijoecam]

Let me preface this by saying this: I've never had a t-case apart, so I don't claim to have been there and done that. I'm strictly going by my understanding of the inner workings of the case, based on the OEM service manuals, and the experiences of others on this board. I'm not an ASE certified mechanic and I don't play one on TV. My name is GIJoeCam and I approve this message.

That being said, I think there may be a flaw in one of your statements:

The rear output is directly connected to the transmission output, via the t-case input shaft which has teeth at the rear output end, which fits between 4 little gears on a larger gear assembly.
In this pic, you can also see how the viscous coupler, gear, and bearing are (fairly) press-fit onto another shaft.

It's my understanding of the 44-04 that the input from the transmission drives the VC, and that the VC then proportions the output to the front and rear outputs. In other words, there is no direct mechanical link between the input and the rear output. This is also evident by the fact that an Explorer with a functioning VC will creep in park when the front driveshaft is removed. If there was a direct mechanical link, the creeping in park would not be possible.

With the 44-05, which does, in fact, have a direct drive to the rear output, we see that removing the front shaft does not prevent the vehicle from being parked on an incline.

Further evidence of this is supported by the text description of the unit and operation description from the OEM Ford service manual (accent and comments added):

The all-wheel drive (AWD) transfer case is a two-piece aluminum, chain driven, viscous clutch type unit. This produces a system in which all-wheel drive is always activated. All-wheel drive transfer case is automatic and has no external controls.

The viscous clutch is a non-repairable, torque distribution device. The internal construction of the viscous clutch consists of alternating plates that are connected to the front and rear outputs of the transfer case. The viscous clutch is filled with a high viscosity fluid which flows through slots in the plates. The resistance to shear causes the plates to transmit torque at the needed ratio. The ratio that torque is transmitted at is approximately 35% front and 65% rear.

A front differential compensates for the difference between the inner and outer wheels. However, when one driveline component travels farther than another, there will be driveline or torsional windup that must be released.


Operation

Torque is transmitted through the input shaft to the planet carrier assembly. Torque flow continues through the gear ring to the rear output shaft. Torque also flows from the planet carrier assembly to the sun gear shaft, which is splined to the drive sprocket. The drive gear is connected to the driven sprocket by the drive chain. Torque continues through the driven sprocket to the front output shaft flange. The viscous clutch provides the connection between the gear ring and the sun gear shaft.

From what I can see from the exploded diagram, the input drives the sun gear of the VC. The Sun Gear drives the VC. The Ring Gear (on the outside of the VC) drives the rear output, and the Sprocket on the front of the VC drives the chain for the front output. In both cases, the power passes through the VC first, then to the front and rear output...

But like I said, I've never held the parts in my hand... I'm strictly going by the OEM service manual and the experience of others on this board...

-Joe

 

[gavin]

gijoecam
you may have schooled me

even with the case apart, it's hard to tell exactly what goes on when the case is assembled.

I have no kind of OEM service manuals on this case either. I'm just going by what it looks like it does.

If you look at this picture:

You can see the output end of the input shaft has splines; the sungear(?) has splines where the shaft fits through.
So in theory, anyway, if the input shaft was rotating, it would rotate the sungear, which in turn would rotate the output, correct?

The VC is it's own beast which is a press-fit onto a separate shaft with the other gear and a bearing.

In this picture:

You can see splines (these are part of that separate shaft that the VC is pressed on to) that fits between the 4-gears in the sungear(?), although the outer teeth of the VC also fit into this cup.


Obviously, I could be wrong on some of these parts. I'm no engineer, nor am I a certified mechanic/auto repair.

I'm just going by what it looks like it should be doing when working properly.


edit: I believe the reason for the creep, is due to the fact that the front end is not helping prevent the VC from "slipping" under an extremely low load. But as I've said, I very well could be wrong.

 

[gijoecam]

I think I got it figured out from those two pics, but I need to ask you a couple more questions... let me get home this afternoon so I can do a screen shot and point to some specific things and ask about them.... But I think I finally get it....

If what I'm seeing makes sense, the input drives the planetary set's sun gear. The sun gear drives the planetary gears, which drives the ring gear/output shaft. That ring gear is also splined to the outside spline on the VC, correct?

I think I finally get it.... I might need some more pics to better explain it... Can you easily remove the output from the rear case half? If you can, set it on the planetary and snap a picture of it from the front output side of the case (i.e. driver's side of the vehicle). Then, slip the output off, and the planetary off (if it's not held on) and conform that the shaft just passes through the center of the VC (i.e. the input is not splined to the center of the VC). If so, then this all makes sense... I'll be able to draw some arrows to better explain it when I get home...

 

[gijoecam]

p.s. Thanks for doing this... I've been wanting to get my hands on one of these for years to see what makes it tick... Just never had the opportunity!!

 

[gavin]

well, my case is together.
I took these pics when I rebuilt my case a couple months back.
That thread is located here: http://www.explorerforum.com/forums/showthread.php?t=218430

Here's the album with all the pics: http://www.explorerforum.com/photopost/showgallery.php?cat=3727

If what I'm seeing makes sense, the input drives the planetary set's sun gear. The sun gear drives the planetary gears, which drives the ring gear/output shaft. That ring gear is also splined to the outside spline on the VC, correct?

yes, this is exactly what it appears to do.

 

[gijoecam]

well, my case is together.
I took these pics when I rebuilt my case a couple months back.
That thread is located here:

Gaaah!! Day late and a dollar short... story of my life...
I'll have to check out the gallery...

yes, this is exactly what it appears to do.

OK, see, the rear output then needs the resistance of the front output to turn. Without the resistance of the front (say, the front driveshaft is removed) when it's in park, the rear driveshaft applies torque to the same part of the viscous coupling as the planetary gear does when the input is driving it. With the input fixed in place when it's in park, the torque from the rear output shaft is free to make the planetary gear set roll around the sun gear, and the VC turns unrestrained by the front driveshaft. That's why it's able to slip...

The more I think about it, the more I marvel at the design...

 

[gavin]

OK, see, the rear output then needs the resistance of the front output to turn. Without the resistance of the front (say, the front driveshaft is removed) when it's in park, the rear driveshaft applies torque to the same part of the viscous coupling as the planetary gear does when the input is driving it. With the input fixed in place when it's in park, the torque from the rear output shaft is free to make the planetary gear set roll around the sun gear, and the VC turns unrestrained by the front driveshaft. That's why it's able to slip...

The more I think about it, the more I marvel at the design...

I think this is basically what I was trying to say. Just in a different way. Maybe.

Basically with no holding force on the front output, the rear can slip due to the design of the VC (the VC itself is slipping).

With weight on the front, the VC would need to move as a whole, rather than "slip," which cannot happen since the transmission is in park. It would have to move the whole t-case input shaft/gears/etc.

Right? :scratch:

 

[gijoecam]

Yeah... we're both trying to say the same thing, and it's not coming out... I think if I play with my fancy new screen capture program I have at home, I can describe it a little more clearly. I'll try and find some of your pics and use some arrows and text boxes overlaid to better describe what we're trying to say.

-Joe

 

[gavin]

added another common question, and an answer that is to the best of my ability. The dreaded "truck creeps in park with no front driveshaft!" question

 

[gijoecam]

I totally get it now.... It just hit me looking through the pics again...

The input shaft drives the planetary gear set (but NOT the VC).
The outside ring gear of the planetary gear set drives the rear output, while the sun gear drives the center of the Viscous Coupling.
The center of the VC is also what drives the chain for the front output.

So, essentially, the planetary gear set drives the front and rear outputs simultaneously. That allows the planetary gear set to act as the center differential.

Now, the rear output is splined to the outside of the VC, and we know that the inner part of the VC is splined to the front output. That's how the VC is able to regulate the the amount of slippage between the front and rear outputs.

It's all so clear!! Seriously, I FINALLY get it!!

Pretend for a second that there's no viscous coupling there... The planetary gear set would still drive both the front and rear outputs. Now, if you remove the front driveshaft, the input tries to drive the planetary gears, the force goes to both the sun and ring gears, but with no resistance, the sun gear turns freely, and the rear output stays stationary. (The same way a broken axle shaft works in the axle)

Now, if you stick the VC in between the front and rear outputs like it does, without the front driveshaft, the VC resists but does not stop the differentiation between the front and rear outputs.

I wish I had some better pics... Wanna take yours back apart and get some pics for me?

 

[gijoecam]

Pics that might help clarify the description:

Does that make more sense now?

 

[gavin]

makes perfect sense

 

[lower4x4]

I think that this might be whats wrong with my truck, when i went wheelin last my friend was taking it out of 4wd and puting in in 2wd constantly and after that it started blowing drive shafts off. i just havent put it in 4wd. i havent been able to see if it will still do it becasue my clutch went, i am currently in the process of putting a new one in. do you thing i need the vicious coupling?

 

[IZwack]

I think that this might be whats wrong with my truck, when i went wheelin last my friend was taking it out of 4wd and puting in in 2wd constantly and after that it started blowing drive shafts off. i just havent put it in 4wd. i havent been able to see if it will still do it becasue my clutch went, i am currently in the process of putting a new one in. do you thing i need the vicious coupling?

You dont have a viscous coupling in your Explorer -- it only came in 2nd generation Explorers with the v8 engine (not te v6)

 

[gavin]

I posted on a local-to-me board, on diagnosing a VC, just because I've been curious.

Get on a flat dry surface and steer to lock in either direction and drop it in gear...if it can make a circle at idle it should be good. With that said, a fluid change will change things for the better.

You should be able to do figure 8's without applying hardly any throttle, if it binds up then your VC is toasted.

 

[dkchrist]

Totally unrelated, but the VC works as there is a shear-thickening fluid which as the fluid is moved past itself (i.e. flows) it begins to form a solid. The faster the flow the harder (more solid) the fluid gets. There are very cool applications for this type of fluid. For example, certain transmissions actually use this fluid to drive the vehicle. If you search for toroidal transmissions, there is a little info on them but not a lot. They are very efficient and have a ridiculous range of gear ratios.

Sorry for the rant, it is just a little cool info that I have learn from some extracurricular college activities (SAE Mini Baja).

 

[gavin]

Totally unrelated, but the VC works as there is a shear-thickening fluid which as the fluid is moved past itself (i.e. flows) it begins to form a solid. The faster the flow the harder (more solid) the fluid gets. There are very cool applications for this type of fluid. For example, certain transmissions actually use this fluid to drive the vehicle. If you search for toroidal transmissions, there is a little info on them but not a lot. They are very efficient and have a ridiculous range of gear ratios.

Sorry for the rant, it is just a little cool info that I have learn from some extracurricular college activities (SAE Mini Baja).

already been covered

How does it work?

Under normal conditions, the inner and outer portions of the coupling are spinning at the same speed. When there is slippage in the rear axle, the rotation of the 2 parts change; since the front tires are spinning slower, the inner portion is rotating slower than the outer. Because of this variation, the viscous fluid inside gets warm (due to clutch packs "mixing up" the fluid), and becomes less fluid, and more solid. As the fluid thickens, it creates friction, by putting more force on the clutches inside the coupling. This then causes more power output to the front tires.

 

[gavin]

gijoecam, in your many manuals, does anything say how much pressure should be required at a front tire to get the viscous coupling to "slip?"

I'm still trying to determine if my VC is toast, and possibly causing me all the rideability/drivability issues my truck has had for over a year now, and it's starting to get worse (truck consistently pulls to right; had alignment done, no change. tire pressure good. half-shafts, wheel bearings, and brakes appear good).

Was dinkin around on the truck yesterday... for giggles, had the truck in neutral with rear tires blocked. Lifted up pass. side front tire. I was able to turn the tire, which in turn rotated the driveshaft, but it certainly took a lot of force. Which, of course, may or may not be normal.