Re: Manipulator Torque (A Bit Scary for Us)
 

I don't know what your robot looks like, so it's kind of hard to figure out what to suggest.

maybe send me a sketch?

Re: Manipulator Torque (A Bit Scary for Us)
 

I have some simple CAD screen shots with no drive train I could show you. The CAD is still in progress...



do those help at all?

Re: Manipulator Torque (A Bit Scary for Us)
 

If you can help support the arm thru the part of travel where it is sticking straight out, and maybe 30 degrees above and below that point, it would probably help quite a bit. Maybe make another shorter arm that is supported by the gas spring, that the main arm can rest on, and as the arm travels way up it lifts off it

Re: Manipulator Torque (A Bit Scary for Us)
 

I think I see what you're saying, that would make perfect sense too. I guess you really don't need to support it all the way through the arc do you, just the perpendicular where torque is greatest.

Re: Manipulator Torque (A Bit Scary for Us)
 

If you do use a gas spring, keep in mind that the rated force is at one end of it's travel--so a 50 pound spring will have more than 50 pounds force when it is compressed. I don't recall what the force is when fully closed, maybe 20% higher? I think there's one or two in the garage I could play with on the scale...

Re: Manipulator Torque (A Bit Scary for Us)
 

Okay, sure. The information you have given me is extremely helpful, Thanks so much for helping me out. Though I think I'm going to get some sleep, I'm on EST so it's 12:53.

Re: Manipulator Torque (A Bit Scary for Us)
 

We had an interesting situation last year, the first time we just calculated the torque to lift the arm at a constant speed. When the smoke cleared, we decided that torque calculations also need to account for acceleration of the load.

This essentially doubled our numbers, but on the second try it all worked.

Re: Manipulator Torque (A Bit Scary for Us)
 

Useless Math: no real need to read......

Alright, I think I have this figured out. We plan to go with two super shifters with two FP motors on the AM Planetary Boxes (to simulate a CIM) as well as two CIMs for drive. Then we use another CIM for he arm. The CIM according to the guidlines manual section operates best at 4350 rpm and 64 oz-in in torque and 27 amps (pretty good). Then I converted that 123.9 foot lbs I need to 125 ft lbs or an even 24000 oz-in. Soooo this is where it gets interesting, you use the CIM with the planetary (184:1) gear box. That gives 11776 oz in, nooooot quite enough (184*64). So we decided to use a short length of #35 chain rated at 269 lbs with a 23 tooth steel hubbed sprcket on the planetary and a 54 tooth on the axel for the arm (which would probably be a live axel with the sprocket dead mounted). That gives 432:1 gearing ((54/23)*184) or 27648 oz in. That is 15.2% over engineered, so it should be perfect. That would raise the arm in 1.92 seconds, which is also perfect (60 seconds/((4350 rpm/432 gear)*(360 rotation/115.97 degree arm arc))).

My Succinct Question:

Math now works the question is mechanical. How do you mount this? From the bottom up the plan was this....Dead mount the 23 tooth sprocket to the planetary with key and pin. Then a short #35 chain to the 54 tooth sprocket. That sprocket is dead mounted to a one inch axel (keyed) and then the two top bars of the four bar are also dead mounted to the axel with a keyed hole and then bolted to two steel hubs (keyed) on either side. Kind of like each arm in a hub sandwich about 6 inches apart with the sprocket in the center. This entire thing might be assisted by a gas spring for 60-120 degrees in the arc of the arm. But I figured I would design it to work without the spring and then if weight and time allows the spring would only help. Do you see any obvious problems with this? Or improvements that we should look at? I have to thank imensely anyone who read this entire thing, I realize it is a bit of a thome.

Re: Manipulator Torque (A Bit Scary for Us)
 

Yeah good point, this is a ten pound ball moving at speed. We are hoping that with a 2 second lift that is just enoguh to not be painfully slow and just slow enough not to make acceleration a huge problem. Plus we can put some ramping in the code for the arm so it doesn't just jolt right up.

Re: Manipulator Torque (A Bit Scary for Us)
 

Remember that the arm and ball are starting at rest.

Our numbers are in the 2007 notebook over at the school, but roughly I seem to recall that it was almost the same amount of work to accelerate the mass as it took to lift the weight. And they add.

Re: Manipulator Torque (A Bit Scary for Us)
 

After thinking about it I realized my torque calculations for the "normal torque" are for the perpendicular ( the arm is parallel to the ground) or worst case scenario. They are also set to the maximum efficiency rating of the CIM. Because the arm starts at about 30 degrees it needs only a half of that previously calculated normal torque (vectors and such, 30-60-90) which means the way I have it set up you have at the beginning of the arch a 12000 oz in normal force to overcome and the motor producing 18720 into acceleration torque. I don't remember how to calculate acceleration from acceleration torque but I think that will be plenty. I hope.
So my big question now is how to mount the dang thing to the shaft. We figured we would key and pin the 23 tooth sprocket directly to the CIM planetary gear box shaft. Then stretch 35 chain about 12 inches up to the 7.7 inch diameter 60 tooth sprocket. But that's about where I start drawing blanks. That 60 tooth sprocket needs to be hard mounted to some type of axle which is also hard mounted to the two top arms of the four from the CAD above (the bottom two arms spin freely) and then that entire axle assembly with sprocket needs to be free spinning. But the axle must be something that can withhold substantial stress (though the closer I put the arms to the sprocket the easier it is on the axle), the sprocket must be mounted in a way that can take that 125+ foot pounds and the arms must be mounted in a ways that can take 62.5+ foot pounds. I have not even the foggiest of what diameter axle to use, whether to key it alone, to use hubs, for it to be steel or aluminum, to weld the sprocket to the axle, to use a threaded axle. it's kind of mind boggling, but this is the single most important joint I have to deal with. I was told belt could actually better transmit the torque but that in this particular application it's difficulty to replace makes it completely not worth the effort, is that correct also?