pic: 114 New Gearbox Design



I figured it’s been a while since I’ve uploaded any of my work, so here’s a bit of a preview on my current project. This gearbox uses the Large CIM, is a ball lock, and eventually will shift via Servo.

The total weight for the gearbox without the motor is 2 pounds, as I plan on using Hardened Aluminum gears and shafts. The box is nice and small, as well as decently simple to machine and to mount.

High speed for a 4 inch wheel bot is 9fps, and low is 3. This gearbox may soon be designed to attach to the Outback Tracks that I’ve been looking to use.

Questions and comments welcome.

Looks good.

Just as a heads up, to save yourself some effort, you really don’t need to harden the shafts. 7075 or 7068 both work fine as shafting, in an unhardened state.

Where are you getting the gears from, or are you having them custom made?

Also, why the decision to move to one large CIM instead of the dual small CIMs like in most of your previous designs?

and lastly, I’m assuming the lower speeds are for use with the tracks? (IIRC, you had 15fps and 5fps this past year)

Thanks, didn’t know that. That should save some time and cost.

The gears will be custom made. I’m still looking into where, who, and how much, but it doesn’t look like it will be that bad.

One large CIM ends up being more efficient, and when coupled with tank tracks, is plenty to shove just about anything around. That also leave 4 high power motors for a beastly manipulator.

Yes, the lower speeds help make this gearbox nice for the Outback track system, with the higher gear set as default. Then, when you need to get into a pushing match, you downshift and “politely escort the other bot across the field.”

The small CIM is a good deal more powerful than a the large CIM, at half the weight as well. Why not change the design to use one? You’d still have two small CIMs left for manipulators (I can’t imagine needing more than that).

Can you show me the numbers that prove that statement? All of the numbers I’ve seen (the ones FIRST gives us for the small CIM are at 100amps of draw) point to the Large CIM being more powerful. Plus, with only one motor, there’s no torque fighting or loss form having two motors.

On page 8 of the Guidelines, Tips, and Good Practices document you will see that the peak power of the small CIM motor is significantly greater than that of the large CIM motor.

Interesting. I’m still seeing higher torque on the 3 inch CIM… However, I’m emailing the source for the numbers I’ve been using, as to find out where they came from. Here they are, in case anyone cares:

----------------FP801-005 3.0" CIM---------FP801-001 2.5" CIM
Torque---------200in-oz--------------------100in-oz
Power----------245 W----------------------250 W
Efficiency-------59%------------------------45%
Current Draw----34A------------------------37A
Shaft speed-----1650 rpm-------------------3800 rpm

Again, not sure where this comes from, but I’m investigating that right now.

Those numbers are totally wrong. The document Dave mentioned has the correct specs. I’ve never seen these before.

Yup.

Also, of course the big CIM has more torque at the posted speed… but that speed is half the small CIM. If you were to match the speed, the small CIM has more torque.

Also, the small CIM draws 133 amps at stall… further confusing me about the numbers you have.

To reliably shift a ball-lock mechanism you need to apply quite a bit of force and apply it very quickly (speaking from experience). Too little force and the plunger won’t push the balls into their slots. Too slow and ball bearings skip out of their slots. This is why every ball-lock shifter I have seen (including ours) uses pneumatics, and I can’t see how a servo would work.

My suggestion is to make the gearbox have a mounting that can work with both servo and pneumatics. This interchangibility will be very good, as well as making your design more appealing to other people.

(Deleted: duplicate post)

These are the performance statistics for the motors at or near 40A, not at stall or at free speed.

Also, though at 40A the larger CIM is producing more torque, it’s moving much slower while doing so, as others have mentioned. A better indication of which motor is “more powerful” is to look at, as you might guess, the power. The smaller CIM has a power of 250W – more than that of its larger cousin.

That’s what I was thinking. These numbers (I recently found out) came from some IEEE phD who was asked to find all the specs on these motors when running in between 30 and 40 amps.

From my basic knowledge, wouldn’t the higher torque at lower speeds mean that my gearbox will end up smaller, less weight, and will still push just as much as an equivalent box with a CIM?

Also, only having one motor eliminates the issue of torque fighting when there are two motors on a gearbox.

I’d be surprised if the weight difference would be that major, and it might even go in the other direction. Since the big chip weighs ~2lbs more than the little one, and you’re using aluminum gears/shafts it would seem to me that the weight of another gear reduction or just a larger reduction on the first and/or second reductions isn’t that different than the weight difference of the motors.

EDIT: I plugged the numbers into JVN’s spreadsheet, and I got the following for the big chip @ 40 A puts out 276W, and the lil chip 275.7W so the difference is barely noticeable. And, yes, I realized the big chip is putting out more power than the peak power rating FIRST supplies, so I ran a quick power calc on the FIRST supplied NLS and stall torque numbers and came up with 285W being the max power of the big cim. So, something is amiss in either FIRSTs numbers or my calcs. I suspect FIRST’s numbers are the culprit because using the same calc I got the right peak power for the lil chip.

At any rate, the 40 amp breakers don’t trip at 40A anyways–it’s significantly higher for a short period of time, so the torque would increase as well.

I am slowly trying to learn something other than controls, and I am having trouble understanding this “torque fighting” you are speaking of. Could somebody explain this phenomena to me?

Thanks,
Pavan.

Motor “fighting” can be explained in one word: nonexistent. It comes up a lot in FIRST, I don’t know why. It simply is not true. As long as the motors are spinning in the same direction, they are both contributing useful torque to the cause. Now, if you have two different motors geared together, their load sharing may vary over a range of speed, but they will easily strike a happy medium. While you will not end up with the full torque of each motor added, you will have significantly more torque than just one of the motors. The efficiency of the system is likely to drop a bit though.

Looks a lot like our gearboxes from this year for our mecanum drive. haha. Spur gears were 20.5PA 24Pitch and so they’re hiding behind.

http://inlinethumb44.webshots.com/9131/2870343260099908742S600x600Q85.jpg

Are you using big cim to save money on gears? If it’s for weight, it wouldn’t really help. The weight for a gearbox using a small cim motor and same final speeds will end up probably in the same ball park. But using two big cims in the whole drive, heaviest motors nice and down low, would allow the small cims to be used elsewhere up higher. But running at equal speeds, the small cim is more powerful than the big cim. I can’t find the spreadsheet I used, but I remember that it was a decent loss, around 20-30 ftlbs, although when it really came down to it, it was traction limited. How much torque will the gearboxes be outputting before stall?

Are you still intent on using just the Big CIM?

Also, you mention Outback manufacturing. Is this THE gearbox they will sell? If it is, it definately needs to support 2 motors and be able to use small CIMs to stay competitive on the market.

I’m still baffled by the choice though… One small CIM is more power for half the weight (remember, the motor can draw 133 amps [at stall] and the electrical system WILL supply it for a short time. It doesn’t just hit 40 and stop), two small CIMs is double the power for the same weight.

Also, One big CIM geared to 9 fps takes a little over 1.5 seconds to reach top speed… For 9 fps, that acceleration is rather slow.

Why not design it for One Small CIM and one Large? That leaves open two small CIMs for manipulators like you desired, and still provides a competitive level of power.