Okay, so I come here from the issue of having a very small hanger and thus not being able to fit two big motors onto it.
The idea of a differential with a single, large (80mm) motor and a small diff was conceived, but after looking into the matter, I found that that is just not gonna happen.
While looking through the web, I found this post, which got me thinking:
This ideal might be hella stupid and purely theoretical, but I am curious what you think about it:
what if you used two motors with different gearing ratios going into a differential, which is connected via drive shaft to another differential, which is connected to the wheels. The motors and first diff are mounted under the deck, the rear diff is mounted on the trucks inbetween gear drive adapters.
The ESC is set up so that with increasing speed, less power is sent to the high gear motor and more to the low gear motor, essentially working as a pseudo CVT (continuous variable transmission), meaning both high torque at low speeds and high speed.
I know this ideal is not feasible in reality - too many parts and complexity for an issue that doesn’t really exist. And I am sure the same thing can be achieved more easily, but I can’t think of how.
Having driven a “fixed axle” eskate around for months, I didn’t found any issue with it. No slipping etc. Etoxxs new insane build has also a fixed axle drive. I’m stern that you don’t need a differential at such short axle lenghts. The difference in turning radius is just negligible.
I have considered a belt drive design with a certain amount of slip in the large pulley to prevent shocks from tearing the belt. You could use that idea with the 1 motor 2wd to effectivly not have a fixed rear axle.
My idea is a workaround to allow the same belt/pulley as without the slip feature. You could get the slip effects by just using v/round belts but then it would be effected by wet weather also not parts we use.
Any CVT with the reliability that would be acceptable would be expensive, possibly the cost of double the motors and ESC.
I think an easy way you could do a CVT is using a spring loaded conical pulley and v belts. They’ve done it for FRC robotics, and the premise is basically the pulley has two halves, and as belt tension increases, it forces the halves apart against a spring, reducing the diameter and increasing the ratio. Controlling the tension and thus ratio could be done similar to Kaly’s electronic dampener.
I could be wrong, but that sounds like it would give you the average of the performance of both motors, rather than the best of both worlds. For example, in the extreme case where you have one motor running and one fixed, the differential acts simply as a 2:1 gear reduction, outputting the average speed.
Correct, but it uses some sort of electronic to mechanical actuator to tighten as it goes faster. In this case, the actuator would move the idler, presumably at the same steps as what the Kaly’s dampener is programmed to move in.
Yes I’m aware, I was talking about the electronic to mechanical translation part of it. It’s easy to translate whatever signal that valve receives to a servo or solenoid or something, but I have no clue how it gets the signal in the first place, hence the reference to Kaly’s solution which is speed dependent, exactly what I need for this.
Usually smaller things that use CVT like carts and minibikes use a centrifugal governor to regulate belt pressure. The faster it goes the more it presses the plates together increasing the ratio.
Yeah I was thinking about a centrifugal one, but I couldn’t figure out how to make one small enough to fit onto a motor pulley. With a servo one, it should be easier to bolt onto an existing motor mount with idler system, and sync both sides which allows the use of dual motors, where a centrifugal one might have differences on both sides based on wheel speed.
