March 28, 2013

Another Differential

Back when I built the nifty spur gear differential for my electric tricycle, I chose to buy, well, spur gears rather than bevel gears like differentials traditionally use.  I did this partly because McMaster's spur gears are much cheaper than their bevel gears, and partly because I had the great LOLriokartdiff off which I could base my design.

A couple weeks ago, while reading through a build thread on RoboWars Australia I found pictures of some bevel gear sets originally intended for angle grinders, which were purchased from Ebay.  A quick search revealed that a number of Amazon retailers stock similar gears.  I ended up getting two sets of these, so I could find out how useful they were.

A few days ago, these came in the mail:

First impression:  They are very small.  The big one is ~1.8" diameter.  I didn't measure the pitch, but it looks pretty similar to the 20 pitch spur gears I used in the trike differential.  They're almost certainly sintered, so don't know if I'd trust them to withstand a melonpower.

In a differential, they would be assembled roughly like this:

I spent the better part of night morning machining the housing for the differential.  Because of laziness and oh god randomly sized helical bevel gears, I decided to not bother trying to CAD anything.  I basically just found some chunks of aluminum and winged it.

For output shafts I used some 1/2" keywayed shaft left over from my original differential.  Since the gears were bored out to 11mm, I turned a shoulder into the shafts.

The gears have some ground down keyway stock stopping them from rotating independently of the shaft, and a big countersunk bolt stops them from moving axially.

To make the housing, I started with some absurdly large timing belt pulleys that were basically solid aluminum round about 3.4-4" thick by 3" long.  The housing is composed of two parts, that are identical except for the six holes.  One side has clearance holes for 8-32 screws, while the other side is tapped.  The cylindrical grooves in each half retain a 8 mm shaft that goes all the way through, and holds the smaller gears.  I made the grooves by bolting the housing together with some paper shims separating the halves, and then drilling holes sized exactly for the shaft.  When the shims are removed, the shaft gets squeezed between the two halves, since the groove profiles are not quite full semicircles.

All the parts of the differential laid out:

All together.  To drive it, you would just bolt a sprocket or pulley through the same six bolts that hold everything together.

I sealed it up with a bunch of grease on on the gears.  And you know what?  Despite my lack of planning, it works beautifully.  Sure, there is a little bit of axial play in the output since I bored out the housing to the wrong depth, but that can be fixed with a washer or two.  Also, the steel-on-steel interface between the shaft and two small gears is definitely questionable.  The motion is significantly smoother than the spur gear differential, and there is absolutely no binding.  It is also a bit more radially symmetric than the spur gear differential, so it should work better at high RPMs, since it won't shake everything apart.  Which may be very important, depending on which of the few ideas I have for this I end up going with.


  1. Wow Ben!

    That's really clever. I've been looking for a good way to build an inexpensive, small differential for a while, but I haven't really been able to wrap my head around spur gear diffs. This looks like a good option. Thanks for the inspiration!

  2. This doesn't look right to me. How do the output shafts rotate independently, as required by a differential? It looks as though the housing rotates to move both small gears, which transmits to both output shafts, which is correct. But if the housing is stationary, the output shafts are locked together by the small gears, and can only rotate in opposite directions. In your spur gear diff, the spurs ride on bearings which are independent, and can therefore allow the output shafts to rotate freely when the housing is stationary.

    Although, maybe I am missing something. I have been staring at it trying to create the motion in my head, and it isn't working.


    1. If the housing is locked, the output shafts can rotate in opposite directions to each other, since the small bevel gears can spin freely and independently on the shaft they ride on. This produces the exact same movement as is present in the spur gear differential.

  3. Correction: the spur gear diff allows ONE of the axles to rotate freely, independent of the housing; that axle which is not on the power transmission side.


  4. where did u use it ?can i use this in a go kart?

  5. Cool!
    Basically a nice simple diff for picking up regen with an appropriate electric drive train, if I have pictured it right.
    Have you had a chance to test the sintered components torque range/capacity?
    Will / Have you add bushings to the small gears?

  6. Great post! Thanks you so much for the share. You are really a master & look our site local web design company in Round Rock & Austin -

    web design round rock
    austin web developer
    austin web design

  7. I must say this is very creative and truly genius, i plan to try and replicate something like this with maybe a couple differences. great inspiration here and a wonderful tool to teach people how a diff actual works.

  8. I was looking to build a differential unit on my Kawasaki Vulcan 750 motorcycle. Many trike manufaturers
    including Voyager and others offer 2 side wheels retaining the OE rear wheel and not introducing a true differential gear set ups. Your posting got me motivated to manufacture a differential unit for my project
    and I leared few tips from this. THANKS IN MILLIONS!!!! Shiv Sivaram