BMW 325iX Transfer Case Schematic

3. Drive Shaft 8. Bearing 9. Shaft (left side is chain drive, right side sun gear) 10. Needle Sleeve 13. Cover 14. Planet gear carrier 17. Hollow Gear Wheel (gear ring) 18. Output Shaft 29. Shaft 31. Chain 32. Clutch

The engine applies torque to the planet gear carrier (part 14) via the driveshaft (part 3). The sun gear (center gear of the planetary set, part 9) gets 37% of the torque and applies it to the front driveshaft via the chain. The ring gear (outer gear of the planetary set, part 17) gets 63% of the torque and applies it directly to the rear drive shaft. The outer case of the viscous clutch (part 32) is splined to the sun gear via the cover (part 13). The inner splines of the viscous clutch are splined directly to the output shaft (part 18).

I finally figured it out. The diagram over the gears is *very* difficult to understand. Part 3 and 18 connection really threw me off. I only figured it out after I noticed the notch on the right side of the planet gear set (p/n 14).

This is how it works: The power input shaft (from the gearbox p/n 3) right-end splines connect to the right-end splines on the planet gear set, part 14. The right-end part of the sun gear, part 9, connects with the planet gears, and the left-end splines of part 9 connects to the chain going to the front output shaft. The outer gear ring, part 17, connects with the outer part of the planet gears and with the rear output shaft, part 18. (part 17 and 18 are locked together)

Now the input shaft turning the planet gear set cannot rotate without either the rear or front output shaft (or both) also rotating. Because the planet gears apply equal force to both the inside and outside gearing, the outside gear receives more power than the inside gears because the outside gear has a longer rotation path (power = force * distance). And since both front and rear output shaft turn at the same speed (if there is no wheel slip) then the rear wheels will receive more torque.

Now the viscous coupling connection thingy, part 13, is connected to the splines of part 9 hence also to the front drive shaft. Part 13 is larger than part 17 (not in the diagram though) so the right-end teeth stick over the right side of part 17, and connect to the visco coupling, part 32, which is also splined to part 18 (the rear output shaft).

So if there is a difference in rotation speed between the front and rear output shafts the visco coupling will start to lock the shafts more and more together. The locking % is dependent of visco-fluid temp, rotational speed difference and power transferred.

This is a very high-end center diff construction. Especially since the visco works between the front and rear output shafts (allows more preferable locking characteristics), not the input shaft and output shaft as in cheaper constructions. BMW really went all out with this :)

Transfer Case Shell

This shows the housing for the drive components. Note the vent at the top (part 17). On my car, overfilling the transfer case causes ATF to leak from here and drip down on the cat.

Comments

• Transfer Case viscous clutch failure: The viscous clutch is a sealed unit and contains fluid which hydraulically connects the front and rear axles. This unit can fail completely or by degree. Different explanations for this are plausible. For example, if the VC is overheated, it is possible that the properties of viscous fluid will be modified resulting in less lockup. Also, if the case of the VC fails and the fluid leaks out, the front & rear axles are no longer hydraulically connected at all (I have seen 2 examples of this type of failure). Assuming a total failure of the VC, only the planetary gears distribute torque. In this condition, the transfer case functions as an open differential distributing torque 37% to the front & 63% to the rear. This means if an axle slips the total available torque is limited by the amount being sent to the slipping axle.
• Chain Stretch: If the transfer case is run with old or no ATF lubricant, the chain will wear excessively and 'stretch'. If it stretches too much it will rub on the case making a great deal of noise. Also note: " a stretched chain will jump teeth on sprockets under load." Jef Rolison
• Stripped Splines on the driveshaft to the front differential: If water or other corrosive agent gets past the splash shield on the driveshaft and attacks the splines they will wear quickly. Eventually, they will not be able to transfer the amount of torque required of them and will skip one or more teeth. The sound that this makes varies depending on how many teeth are being jumped. It may sound like anything from a single clunk to a repeated series of clunks (like cards on bicycle spokes only more metalic and sinister sounding...). If your driveshaft splines are completely worn out and your viscous clutch has failed, your car will just sit in one spot making a grinding sound as the front splines skip.
• Using washers to 'fix' a stripped driveshaft: The driveshaft splines on my wife's car were stripped and the viscous clutch was ineffective, so the car would just sit there and grind as described above. As a temporary (1 year) fix, I lengthend the driveshaft the distance of 2 stacked washers (about 1/8 inch). I recently replaced this temporary fix with a new driveshaft. This is a picture of the old one after driving around for a year with the washers installed. In this picture, you can clearly see the contact points where the previously-unengaged splines had contacted the old worn ones. What you can't really see is that there was no significant wear to the newly engaged splines on the driveshaft. They are just bright and shiny.
• ATF in the transfer case leaks into the automatic transmission: Not likely. If you look at the diagram above, and remember there is 1/2 liter of ATF in the case, you can see that this fluid sits in the lower half of the case bathing the chain. Of course, the chain will throw the fluid around inside the case, but in order to make it to the automatic transmission, it has to splash up to the driveshaft and migrate uphill past a seal in the transfer case then past a seal in the autobox. No way, Jose. I have seen fluid get thrown out the top vent and drip onto the cat, where it makes a smokey mess. My wife's car has done this each time I have refilled the case.
• Q: Has anyone got their IX to move without a front driveshaft? (Response by Thing): Yes, I accidentally tried this. I was changing the rubber boot for the outer universal joint on the front left driveshaft. And when stuffing it back in place failed to insert the driveshaft all the way into the front diff. This caused the car to roll slowly while parked on a slight slope with first gear engaged. It got back in and drove it about 40m and could not feel any difference in clutch engagement or slip, except that the car had absolutely *no* traction on the gravel, with the rear wheels just losing traction with even slight throttle application. So the front wheel drive was obviously not 'coupled'. I got back under the the car and stuffed the driveshaft in all the way and problem solved. Now I could accelerate at full throttle again without the rear wheels just spinning up. So the center visco should be *very* stiff. If not someone has cooked it by using different diameter front and rear wheels, or swapped the rear diff ratio or something equally stupid.

  Note: if you happened to make the same error as 'Thing', but had a bad viscous clutch in the transfer case, your car would not move. Similarly, if you spun your front driveshaft (worn splines), and have a bad viscous clutch in the transfer case, your car would not move either.

• Judging the quality of your viscous clutch:
  • The viscous clutch should transfer enough torque from the slipping axle to the opposite axle to move the car. In the example above, Thing demonstrates that the car can be driven in this condition. If your car can be driven with the front driveshaft removed (or the equivalent -- a front CV disengaged), this means your VC is good enough.
  • The proper test for the VC involves putting the car on rollers. See message 7577 on Yahoo bmw325ix group: (Michael Helbraun): " in the shop manual the proper test for a VC, is on a set of rollers and a graph of permissible values per wheel speed is given. I'd think that if the VC either failed or not, there would be a simpler determination of good or bad rather than a permissible range of values. "
• If you have any additions or corrections for this section please let me know. thanks ... Ray

1. Crown gear set 2. Impulse sending wheel 4. Limited slip diff unit viscous 5. Hex Bolt 6. Bevel gear shaft 7. Bevel gear (2, one on each side) 8. Ball cup (2, one on each side) 9. Side gears (2) 10. Spacer ring 12. Screw, fillister head

For a description of how differentials work, take a look at the howstuffworks web site. In the above diagram, the viscous coupling (4), connected (via splines?, not apparent in the diagram) to each of the rear drive axles, detects differences in rotational speed. If one wheel slips, the coupling sends additional torque to the opposite (non slipping) wheel.

BMW E30 Home