Quote from my mentor: "Keep things as complex as necessary, and not a bit more." - Mike [more]
 Chief Delphi Flipped Transmission By SteamPunk #1577
 CD-Media CD-Spy
 portal register members calendar search Today's Posts Mark Forums Read FAQ rules

# everything

Flipped Transmission By SteamPunk #1577

#### By: AmitShavit New: 08-11-2017 08:51 AM Updated: 08-11-2017 08:51 AM Views: 930 times

Steampunk 1577 has recently been working on a flipped transmission for 2 cim motors in 1:4.66 ratio.
The transmission is made of 3* 5mm (~0.2") aluminum plates, 2* 44t 20dp gears, 2* 12t 20dp gear (for cim) and a 56t 20dp gear.
It's designed in two levels - the first level decreases the speed (1:4.66) and the second level passes the rotation to the wheels' axis.

While designing this we used this formula in order to calculate the center distances:
C = (N1/DP + N2/DP)/2
Nx = number of teeth on gear x
DP = diametral pitch

We wondered whether should be any extras added to the result of the above formula?

Would like to get your opinion,
Amit Shavit, SteamPunk #1577, Israel.

• Guest

# Re: pic: Flipped Transmission By SteamPunk #1577

Looks good! It's pretty standard in FRC to add 0.003" (0.067mm) between gears to keep them from binding up. It looks like the pocketing could use a bit of work (especially in the inner-most plate), but other than that it looks like a nice gearbox.

# Re: pic: Flipped Transmission By SteamPunk #1577

Quote:
 Originally Posted by Ari423 Looks good! It's pretty standard in FRC to add 0.003" (0.067mm) between gears to keep them from binding up. It looks like the pocketing could use a bit of work (especially in the inner-most plate), but other than that it looks like a nice gearbox.
I second everything here. My two additions are:

1. I don't know anything about your team's manufacturing, but make sure you can hold at least +- 0.003" accuracy on your mill or router. If you are +- 0.004" or worse, you can go the "beneath" the 0.003" addition and really grind gears. This makes very inefficient gearboxes and very hot CIMs, not fun. Single speed does make manufacturing easier and this gearbox is well packaged, so that is good.

2. Your theoretical free speed with a 4" wheel is nearly 20ft/s which is pretty speedy, so make sure you leave room for drive practice, and avoid prolonged pushing matches. Even with some more conservatively geared 4CIM single speeds, you are in danger of really heating up motors or popping breakers, if your robot is heavy. I speak from experience. At our off-season event our shifting gearboxes got stuck in high gear (18ft/s theoretical max), and after a ~10s pushing match our main breaker popped. If your robot is heavy and geared aggressively, AKA not traction limited, you can stall CIMs and pop breakers quickly.

With those warnings out of the way, this looks like a good gearbox overall.

# Re: pic: Flipped Transmission By SteamPunk #1577

Quote:
 Originally Posted by Ari423 Looks good! It's pretty standard in FRC to add 0.003" (0.067mm) between gears to keep them from binding up. It looks like the pocketing could use a bit of work (especially in the inner-most plate), but other than that it looks like a nice gearbox.
I second everything here. My two additions are:

1. I don't know anything about your team's manufacturing, but make sure you can hold at least +- 0.003" accuracy on your mill or router. If you are +- 0.004" or worse, you can go the tighter than the 0.003" addition and really grind gears. This makes very inefficient gearboxes and very hot CIMs, not fun. Single speed does make manufacturing easier and this gearbox is well packaged, so that is good.

2. Your theoretical free speed with a 4" wheel is nearly 20ft/s which is pretty speedy, so make sure you leave room for drive practice, and avoid prolonged pushing matches. Even with some more conservatively geared 4CIM single speeds, you are in danger of really heating up motors or popping breakers, if your robot is heavy. I speak from experience. At our off-season event our shifting gearboxes got stuck in high gear (18ft/s theoretical max), and after a ~10s pushing match our main breaker popped. If your robot is heavy and geared aggressively, AKA not traction limited, you can stall CIMs and pop breakers quickly.

With those warnings out of the way, this looks like a good gearbox overall.

# Re: pic: Flipped Transmission By SteamPunk #1577

Looks like a solid design! Don't forget about wires when you're placing your cims. It looks like they may need to go a little further from the edge of the wheel just to be safe, but hard to say without CAD. Your lightening is decent, but it could use a bit of work on improving strength. There are a couple spots where I would add more struts, but I tend to be pretty conservative with lightening so you may be fine. I hate the wasted space with the sprocket, but it is kind of necessary with the length of the Cims. An interesting tradeoff here is using 3 minicims, but that also has its own problems like more pdp ports, weight (in pinions fasteners and extended plates), and height among others. Maybe you could fill some of that wasted space with an encoder mount just so its conveniently in the gearbox, or maybe some of that space can be taken up with shifting. I've tried over and over, and with shifting you can't get much better packaged than the WCP flipped cim, but it's interesting to see how you've retained this nice package without shifting.

Great job on the design!!!

# Re: pic: Flipped Transmission By SteamPunk #1577

Quote:
 Originally Posted by Peter Salisbury I second everything here. My two additions are: 1. I don't know anything about your team's manufacturing, but make sure you can hold at least +- 0.003" accuracy on your mill or router. If you are +- 0.004" or worse, you can go the tighter than the 0.003" addition and really grind gears. This makes very inefficient gearboxes and very hot CIMs, not fun. Single speed does make manufacturing easier and this gearbox is well packaged, so that is good. 2. Your theoretical free speed with a 4" wheel is nearly 20ft/s which is pretty speedy, so make sure you leave room for drive practice, and avoid prolonged pushing matches. Even with some more conservatively geared 4CIM single speeds, you are in danger of really heating up motors or popping breakers, if your robot is heavy. I speak from experience. At our off-season event our shifting gearboxes got stuck in high gear (18ft/s theoretical max), and after a ~10s pushing match our main breaker popped. If your robot is heavy and geared aggressively, AKA not traction limited, you can stall CIMs and pop breakers quickly. With those warnings out of the way, this looks like a good gearbox overall.
Thanks both of you Ari and Peter for the 0.003" tip.
Peter, we're pretty confidence about the speed, but if we see we need lower ratio so we will go with 3.5" wheel.

Quote:
 Originally Posted by marcusbernstein Looks like a solid design! Don't forget about wires when you're placing your cims. It looks like they may need to go a little further from the edge of the wheel just to be safe, but hard to say without CAD. Your lightening is decent, but it could use a bit of work on improving strength. There are a couple spots where I would add more struts, but I tend to be pretty conservative with lightening so you may be fine. I hate the wasted space with the sprocket, but it is kind of necessary with the length of the Cims. An interesting tradeoff here is using 3 minicims, but that also has its own problems like more pdp ports, weight (in pinions fasteners and extended plates), and height among others. Maybe you could fill some of that wasted space with an encoder mount just so its conveniently in the gearbox, or maybe some of that space can be taken up with shifting. I've tried over and over, and with shifting you can't get much better packaged than the WCP flipped cim, but it's interesting to see how you've retained this nice package without shifting. Great job on the design!!!
We're planning to add an encoder mount in the final version.

previous
next