pic: Gerrish Gearbox MKII
 

Hmm, looks interesting. Perhaps some 2D orthographic views though? I can't quite make out exactly what is going on...

Ortho-What now?

The standard in presenting mechanical parts. 2D Top, Front, and Side views....

A very interesting concept indeed. There is a lego differential that looks exactly like that, just minus the motors on each side and the final drive gear on top. But since that is a differential, it would compensate for the rpm differences in the motors. I'm just not sure how well it would work with a 14,000 rpm difference.

Thanks

But, Where Am I going to find a 14,000 rpm diffrence?

He's referring to the difference between the Atwood's speed and the drill's. The Atwood is supposed to be ~5000 rpm, and the drill is supposed to be ~19000 rpm, hence the 14000 rpm difference.

...I'm not sure I understand how this would work.

Shouldn't a differential have two outputs? I'll admit, though, I'm really not clear on how they work.

But, again, since the motors would each have varying RPMs. . . well, how does it work? The Drill output is 19,470 RPM or something. The Chiaphua is 5,500 RPM. They're each connected to a unique shaft and miter gear? Then those two shafts input into a single miter gear there at the bottom? That does nothing to compensate for the difference. Then, I guess, one of those original two miter gears is attached to the larger miter gear, with then gears *up* to the output?

I don't know that I follow what this is meant to do at all.

Perhaps an explanation coupled with a few less fancy drawings is in order?

I did some tests with this theory out of legos.

On one side of the differential I geared a Lego motor up (40 tooth on motor shaft meshed with 8 tooth gear going to differential). On the other side I geared a Lego motor down (8 tooth gear meshed with 40 tooth gear going to differential).

The differential spun at the speed and torque of the geared up motor. The geared down motor did not have any effect on until I held the differential until the geared up motor stalled. That is when the geared down motor kicked in and spun the differential with the speed and torque of the geared down motor.

So from my findings it seems that this does not work, unless anyone can prove it correct.

I studied it for a few minutes.. and I think it's more for a pretty picture with 3DMax than anything else.

Well, I guess differentials don't work in reverse then where there are two motors and one output instead of one motor and two outputs. It was an interesting concept to toy with though. I wonder if there would be better results if the speed difference was not so much.

Yeah it really doesn't work that way. It is just the way that a differntial is designed. That same problem can be rather quite problemsome backwards also if you loose all traction in one tire. Btw After watching a few episodes of junkyard wars can someone tell me what it means to lock a differntial??? From the way it sounds it might just fix the above problem though it would probably defeat the purpose of it.

Locking a differential is accomplished by physically connecting the two, otherwise separate output shafts so they act as one. In situations where an open-differential would result in getting stuck. . .where one wheel is sleep and the other is getting no torque, locking the differential would make the rear axle act as a solid piece, giving torque to the wheel that's not stuck.

Sorry
 

If it is not clear how this device works i will make a more clear and understandable render with labels for now i tell you it has two inputs (the motors) and one output (the shaft on top)

Re: Sorry
 

After spending the last few hours reading a lot about differentials and their hellspawn, I'm pretty sure that what you're attempting here will not work.

The entire point of a differential is to vary the speed and torque of two output shafts . By reducing it to a single output with two inputs, you're not creating a differential anymore. After all, what's "different" about the output(s)? Nothing. It's a single output operating at single speed at any given time.

Now, though, I think I understand what you're trying to accomplish. You're using the concept of a differential in reverse to mate two motors, and then you need two of these modules on a robot to operate both sides independently.

Based on this the mathematical relationship at work in a differential with 1:1 gearing is input = (outputA + outputB) /2.

So, for example, let's say input is 10. Therefore, it's 2(10) = outputA + outputB. While driving in a straight line, outputA and outputB would be equal at 10 each -- or the same as the input. If there is resistance against one output, more power/speed/torque will go to the output with lesser resistance. This might end up being a 12 + 8 or a 3 + 17 or even a 20 + 0 scenario.

Your idea here reverses this concept. So, therefore, out equation would read more like this:

output = (inputA + inputB) / 2

So, again for example, let's look at the motor RPMs for the chiaphua and drill under no load.

output = (5500 + 19000) / 2
2(output) = 24500
output = 12,250

Simple enough, right?

Well, now let's look at this in terms of torque.

inputA = 2.2 Nm
inputB = 1.2 Nm

2(output) = 2.2 + 1.2 = 3.4
3.4/2 = output = 1.7 Nm.

So, as you can see, the output of your differential is actually less than what you'd get from a single motor.

I'm unsure if there any circumstances where this wouldn't be the case, but it doesn't appear that there are.

I'm also unsure how things would be affected if the gearing were something other than 1:1.

If I've made some grossly wrong assumption here, please correct me. Thanks.