Searching through the old threads, apparently none have answered the question relating specifically to the toughbox, only to banebots and drillmotor gearboxes. Just how much input torque can the toughbox handle before failing? We used one with two CIM’s this year for our winch, and the thing performed like a charm, easily outputting over 100 ft-lbs. So what if we could have doubled up the toughboxes together to improve the efficiency of the motors operating it? Would it be able to handle such loads, even if it reached the top and stalled?
Why you would ever need two CIM’s into two stacked toughboxes, I do not know. You are lifting, something easily done by one CIM fairly quickly.
While I have never tried it, my team had a single CIM feeding a DeWalt feeding the guts of a Toughbox (re-made side plates and new input gear, same gears, bearings, and the long shaft with a bearing block on the other end), and after about 6:1 chain reduction we flipped ourselves over using the vertical pole. That is a lot of load. And it was happy. Our math says the end joint (with 6:1 chain from the toughbox output) had around 450 working ft-lbs of torque. Stall on the gearbox output is around 180 ft/lbs. It lifted the robot and held it while it wiggled (the DeWalt had anti-backdrive pins). Here’s a pic of the prototype:
http://www.chiefdelphi.com/media/photos/34788
Because both the DW and Toughbox have around 12:1 reduction, this would subject the toughbox in this pic. to around as much load as the second in two toughboxes (but this only has 1 CIM). We made sure in software that we never stalled, but on the one rare occasion that it happened, the CIM began to get… unhappy. Had we run it longer, it would have smoked before anything mechanical gave. So, with one CIM, you will be fine.
The toughbox is… tough.
I am a little confused on your setup. Are you asking to run two CIM’s into one toughbox, then the output of that into another (as I assumed above), or something different.
That could lift your robot pretty fast. Probably not a benefit this year over 1 CIM but it’s certainly usable.
You’ll want to go to steel shafts in high reductions, but I’ve seen AM gears being used in extremely high reductions.
Yes, I’ve been considering an idea for next years drivetrain. Unfortunately, the motors would only be capable of outputting power in one direction, and was thinking maybe a modified toughbox would become capable of shifting between forwards and backwards with an idler gear installed somewhere. The thing would essentially go along the lines of 3-4 motors->toughboxes->output shaft->directional toughbox. I was curious as to whether we’d need to build our own heavy duty gearbox to handle the torque involved.
Why do this when you could reverse the voltage to turn?
With my potential design plan I’m looking at, power can only be transmitted in one direction. Reverse the motor direction would do nothing to turn the drive shaft, like on a bike. Only goes forwards, not backwards. Actually, a bike is exactly what I’m looking at. It has 21 speeds to choose from, and the only problem is modifying it to work in reverse as well. :yikes:
-
the only reason a bike system won’t go in reverse is because the chain tensioner will only take up slack in one direction, and there is a one-way bearing on the input shaft to prevent loss of chain tension. The active chain tensioner is basically required because the sprocket size changes depending on gear. If you tensioned the chain in both directions, you could remove the one-way bearings and not worry about loss of tension in the chain.
-
Do you need 21 gears? I don’t think so. We built a 4-speed in 2004, and we didn’t really need more then 3 gears. It was automatic, and it hardly spent much time in 2nd before shifting to 3rd.
-
Have you heard of a CVT (Continuously Variable Transmission) and it’s cousin, the Thunderchickens CCT (Crazy Chicken Transmission)?
- Thanks for the tip. I’ll have to look into that.
- For the win/fear factor. You got 4 gears you say? Well, we got 21, [exclamation here]. Ha. It’s mostly for fun bragging/pimpin’ our design to other teams/awards, but because of the huge differences between low and high gears, it’ll allow us to choose extreme torque and keep a setting for decent, maybe 20 fps+ speed.
- I have, but in terms of design, This is essentially continuous while being a COTS item. Plus, the possible friction losses appear to be higher than the chain design. (maybe)
My team is looking at using a type of CVT transmission next year. I did not know of any team trying something similar. If there is any information out there I would like to know.
With my potential design plan I’m looking at, power can only be transmitted in one direction. Reverse the motor direction would do nothing to turn the drive shaft, like on a bike. Only goes forwards, not backwards. Actually, a bike is exactly what I’m looking at. It has 21 speeds to choose from, and the only problem is modifying it to work in reverse as well.
How would you make it shift gears? In my experience bikes only shift gears while moving. If this is the case you could concievably get stuck in the really fast gear and be unable to shift into the gear with more torque and could end up not being able to get moving again.
I’ve known about bike-like limitations. I believe it would be feasible to incorporate into the reversable gearbox a neutral position, which would act as a type of clutch, allowing the sprockets to spin freely without any connection to the wheels/treads/legs/etc.
That would probably work but it could be rather inconvenient to have to put it into neutral to change gears to start moving.
It would be a really cool design challenge though.
You would have to design it so it could not get stuck in gear…
-
You would have to ensure you have enough torque at the highest gear to move the robot from standstill. If you want to go 25ft/sec, will your motors be able to get you going if you are not shifting up.
-
If you get pushed and want to shift down, you will either have to back up to shift, or have a pneumatic clutch release the wheels so you can shift
Another problem you have is that 21 gears is far too many for a driver to handle (ours has a hard time with 2 sometimes). But, if your programmers are up to it, fully automatic.
I see any system based on a bicycle derailleur gear as being far more complex then it is worth.
Our four-speed was based on two two-speeds in series (see whitepaper), so you could add multiple two-speed transmissions in series until you had “enough” gears. We found four to be enough, and we have not used more than the three provided by DeWalt transmissions enough since, and have recently used AndyMark transmissions which have two gears.
About the CVT, the 217 “CCT” is not a true CVT, it provides infinite variability but does not actually change the output torque - It is a planetary transmission with a motor on the sun and ring gears (with the ring driven by a non-backdrivable input, they used a worm gear). The sun provides the power, and without the ring spinning, it is at a fixed ratio. They then powered the ring gear to increase the ratio, but since they are adding more power (to spin the ring) to make it go faster, the torque is still the same.
Again, why?
Once you become traction limited, you’re pretty much done torque wise. Every gear lower than that just decreases current draw. When you get higher, you get incrementally worse acceleration and greater current draw for incrementally more speed.
Maybe a better track to start on is what about a two speed transmission leaves you unsatisfied? Top speed too low? Not traction limited at the low speed? Want another gear for smoother acceleration?
About the CVT, the 217 “CCT” is not a true CVT, it provides infinite variability but does not actually change the output torque - It is a planetary transmission with a motor on the sun and ring gears (with the ring driven by a non-backdrivable input, they used a worm gear). The sun provides the power, and without the ring spinning, it is at a fixed ratio. They then powered the ring gear to increase the ratio, but since they are adding more power (to spin the ring) to make it go faster, the torque is still the same.
Thanks for explaining that.
The trouble is getting to the point where it’s traction limited. This could be around 150 to possibly even 200 pounds of tractive force with the right gripping setup. Even with 4 motors, it’ll be difficult to get a usable top speed once it’s geared down all the way. Furthermore, a second gear with a higher top speed for this would probably create a useless power well, and would have to be tested for optimization (calculus would come into play and simplify things here, but considering that only about 10% of my school takes a beginning calculus course, it’s out of the question). Last, the normal two-speed shifters are simply too expensive for us, and would also require pneumatics, something we’ve began to abhore due to it’s weight and volume requirements.
Traction limited with the most common tread material drawing 40 amps of current can get you to about 4-5 fps. Not fast but a reasonable speed for a shifter.
Making a 21 speed drive train with a mechanical reversing secondary gearbox is going to be much, much, harder than fundraising the $800 or so two shifters and extras would cost, for almost zero benefit.