Torsen Limited Slip Differential

by pizzachef Aug 16, 2016
Download All Files

Thing Apps Enabled

Please Login to Comment

The worm gear can spin the worm wheel and the worm wheel can spin the worm gear. Technically not a worm gear/wheel. Thus, it functions as an open differential. If you are like me and wanted a torsen differential to learn how it works, this will leave you scratching your head. Other than the main function, this thing is beautiful. The files were great and printed like a dream.

Hi, are the shafts the same a qspoc's differential design? I'd be interested in using the yoke instead of a third knob

I added a new part called PinionShaft_forYoke.stl so you can attach qspoc's yoke. You may have to print the yoke a little oversized if the fit is too tight.

No, the pinion shaft is a little bigger and bolts on differently. Give me a day or two and I'll see if I can edit the yoke to fit, or make a new shaft to fit the yoke.

Can you put up a picture of where to add the motor? Ive printed DriveGearUpdated.stl (which I think you refer to as Drive Spur?) and its not clear how to integrate with the rest of the piece. thanks for the great model btw!

I'm not sure why the file DriveGear_updated.stl is in the list, it is not used in the model. That is from the remixed model, may have been a mistake when I uploaded. The two gears used to drive with the motor are DriveSpurGear.stl and DriveSpurPinion.stl. Have a look at the model from qsopc:

He has pictures with the motor attached. The pinion shaft has a little ridge in the side of it to act like a key, and the DriveSpurGear has a mating keyway. The DriveSpurPinion has a hole with a flat spot to mate with the motor shaft.
Good luck! Let me know if you have any trouble.

Functional Differential Gear System
by qsopc

Hi! Just finished printing and assembling this thing. Thanks :)
Overall the design is really good and convenient to print.
It seems it doesn't work... I mean - it works as a differential, but not as a Torsen :) - it doesn't lock at all. As a Torsen it should lock if you try to stop one axle sharply, or if you try to spin up one axle very rapidly - Torsen reacts to torque / radial acceleration.
I think the problem is in gears. "Worm screws" and "worm gears" have almost identical helix angles - which makes them to behave more like normal helical gear pair, not like a worm drive.
In a real-world Torsen differential worms are more skewed which in fact allows them to work as worms.
It's simple to test: try to rotate the "worm gears" (red on your picture) by hand and see if "worm screws" (ones on axles) rotate - in my case they do. They shouldn't.
So it seems gears need to be redesigned... do you know any opensource tools able to design helical gears & worms?

Hi vitalif,
Glad you like the differential. About the gearing, I didn't design the torsen part, I remixed it from a previous design. But the helix angle of the worms seems to match those of real life torsen diffs...see Miata example here:

On the one I printed, the torsen part worked too well, it would not let the two wheels rotate independently, so I added grease on the worm gears so that it would slip a little. You don't want the worms to lock up completely, just give some friction to resist a total "freewheel" if one side has no traction. If you adjust the helix angle, you will get more or less friction, I suppose it depends on how much you want to lock the wheels together, and how much you want to allow some relative rotation for going around corners and such.

I used Inventor to design my models, and that is not open source. If I can get some time, I can try editing the helix angle for you, if you'd like to try something with less slip. How drastic would you like to go?

This is awesome, could you upload the source files, I would like to make it a functional Diff. and STL-files are tuff to edit, and some of the dimensions make it difficult to find bearings and other external parts.

Just did a quick pack and go from Inventor, it should have grabbed them all. The files are Inventor 2017 format. Let me know if you have any trouble with them or if something seems to be missing.

ABS anyone? because you could treat with acetone for extra slip. (make smooth) maybe even try 0.1mm layers??? (may cause wobble though... i've had better prints @ 0.3mm)

I don't use ABS much, and have never done the acetone back for smoothing. But I'd love to see how it comes out if you try it.

I think that there is a problem with the BevelGear_Housing file. Please check the file.

I was able to download it and slice it with slic3r. Can you describe what problem you're having?

I was using Simplify3D software that failed to print the model correctly. I tried using PowerPrint software and it worked (I hope because it's about 65% done). The PowerPrint software uses the slic3r software.

Thanks for the heads up. I don't have S3D, so I wasn't able to test it with that slicer. If you know what was going wrong, I can try tweaking the model.

Is that grease I see on the worm gears? Does it help? I know I said something about the use of grease in my description but I've never actually tried it.

Thanks! Yes its white lithium grease. The limited slip was so limited, it needed grease to let one side spin when the opposite side was held. My worm gears didn't come out super smooth either, so there was some extra friction from the slightly textured surface. I tried some worm gears in PETG and they seemed to come out a little nicer.

That is awesome. I love it. Great job!