View Full Version : differential for odd use

12-15-2012, 12:24 AM
Have an idea in mind which would require the use of differential gearing. This isn't something that hasn't been done before in some form, but I want to do it for myself. It involves the use of two electric motors driving what would normally be the axle shafts, and in this case the power would be coming off the ring gear instead of driving it. If the motors were both going the same rpm, but in opposite directions, the gears would all be whirring around except for the ring gear, which would be stationary. If the motors were turning at different rpms, the ring gear would be turning at half the difference between those rpms. If one motor was stopped, the ring gear would be turning at half the rpm of the rotating motor. If the stopped motor was then brought up to the rpm of the rotating motor, but in the same direction as it, then the ring gear would come up to the same rpm as both motors. It's a drive system based on a two motor/generator set.

My question has nothing to do how to make such a system work. I understand that. But a typical differential isn't made to have the spider gears constantly spinning at high rpm- they only rotate when you change from straight line motion. They only rotate at high speed when you have one wheel with traction and the other on ice and you're revving the living liquid bodily waste fluid out of it. But again, they're not really made for that.

What I'm looking for is a type that can handle that. The system can temporarily output more power than both motors together can provide (with my way of doing it anyway) so since the motors I want to use are about 2 horse each, I'm thinking the gearing should be able to handle 10 to 15 horse. I had been considering a garden tractor transaxle as a possible source, and I've also been looking at another type of differential using only spur gears. The spur gear type can use an inner-tooth ring gear for the driven sides, which might make for a little stronger unit, but basically the reality is that I'd need to be able to get a suitable unit in my hands. Any ideas?

12-15-2012, 01:36 AM
The toyota hybrid system does something similar but they use a planetary gear system.

http://eahart.com/prius/psd/ for a good explanation.


12-15-2012, 02:04 AM
Thanks Dan. That's a good write-up and interactive animation. It is quite similar to what I had in mind, and the planetary gear module would work for me.

12-15-2012, 03:32 AM
MY Pleasure, good sir.

At some point in the near future the Toyota patent will expire. I know that they were using that design in in 1997. It seems like the idea will scale up and down pretty well.


12-15-2012, 06:02 AM
And in a highly ironical twist, I have found a set of planetary gears sitting on a shelf. I shall have to check out the ratios and see if I can use them.

11 Bravo
12-15-2012, 11:41 PM
What speed are you planning on running the driving motors?

The failures I have seen in spider gears from one wheel spinning resulted from the differential carrier spinning so fast that the oil slings away from the spiders and they essentially run dry. If you think about it, there really isn't a great deal of load on the differential when one wheel doesn't have traction. I suspect the spiders could handle the load unless it is above what the differential was designed for. If the speeds were within reason and the gears stayed flooded with oil I bet they would handle what you are talking about. It would be an interesting thing to set up and find out for sure.

Not exactly what you have in mind, but the differential steer system Caterpillar uses on some of their track type tractors uses a single varying speed motor to run against a planetary ring gear. The motor can hold the ring, or run it in either direction at varying speeds to execute turns with both tracks under power. It makes the tracks rotate at different speeds, or counter-rotate. I have worked on these systems quite a bit and they are very interesting. I never thought about what driving the axles to get power output from the differential pinion gear would do, but it would be similar to what you have in mind, only different.;)


12-16-2012, 05:09 AM
You can build a differential using externally toothed spur gears.

The Artful Bodger
12-16-2012, 05:34 AM
You can build a differential using externally toothed spur gears.

Early Austin 7? Sturmey Archer adult tricycle diff.?

The Artful Bodger
12-16-2012, 05:39 AM
The failures I have seen in spider gears from one wheel spinning resulted from the differential carrier spinning so fast that the oil slings away from the spiders and they essentially run dry.

If I recall correctly the limited slip diff as used by GM Holden in 1960's had a friction washer at the 'fat' end of the spider gears which limited the amount of slip but if over done would burn the friction washers and cause problems. Presumably that friction washer could be been replaced with a suitable thrust bearing to get the high imbalance without burnout.

12-16-2012, 06:24 AM
It seems that three things determine whether a diff can handle continuous high differential speeds- one is the lubrication factor and another is the capability of the bearings to handle both the rpm and the centrifugal forces. The third is the ability of the teeth to mesh at high speeds without power loss and damage. I don't think that bevel gears are the right thing to do this with. There are of course spur gear diffs, and it's interesting to see the use in those steering systems. The action coming out is pretty much the same in every case- the rotation of the input, output, and secondary shafts are intertwined. An action on one causes a change in the others.

In my idea, the input shaft would turn at a maximum of about 3000 rpm. If I used the planetary gearset I have, the output shaft would have to rotate at 9000 rpm in order for the output shaft to be stationary. Dropping the rpm on the secondary shaft would cause some rpm to appear on the output shaft. Through control electronics I would tap power from the secondary motor and feed it to the input motor to augment battery power to keep the input motor rpm up. This is a case where the input motor and the secondary motor both can be run up to speed without the output shaft having to turn. In controlling the relationship of the rpm on the input shaft and the secondary shaft, you give a controlled spin-up of the output shaft. The input motor would be augmented to increase its flywheel effect so as to store an amount of inertial energy. This would supply the torque to get a vehicle from stopped up to speed without requiring a high current drain from the battery. In fact, braking the secondary motor would act to transfer more torque into the output shaft than either motor would be capable of either independently or together.

I will have to do some homework to figure out if there's a ratio in the gearing that is eminently suitable for this application.