We’re looking into designing custom gearboxes next year to move away from planetaries in high load situations and also just in places they would package better.
There are many 20DP pinion sizes available, but I feel like the most common ones I’ve seen are 14T. What is stopping teams from using a smaller pinion on their first stage reduction, like the 8T Falcon pinion? Does pinion choice affect load capacity or backlash? How does one go about choosing a pinion size for a specific purpose?
I’ll let others dive deeper into tooth strength topics, but some things to contemplate:
(Worth noting that the REV pinions were designed so the 10/11/12T gears have the same pitch diameter, and so do 13/14T. That means you wouldn’t have to swap the spur gear to make those changes. I think the VEX/WCP ones have something similar going on but haven’t researched.)
Think you are mixing things up a bit.
The 20DP specify the Dimetric Pitch of that specific gear. DP basically defines how the teeth geometry is setup. In order for gears to be compatible, they must have the same DP (or Modulus if using SI).
Whereas the “T” specific the number of teeth. Which basically will translate into your gear ratio but this also will specify the diameter of your gear as well.
I think they’re talking about various tooth counts of 20DP gear and asking why people don’t always start with the smallest number?
ooops. sorry my mistake.
Thanks for your input! Love hearing that I can swap in different pinions that use the same pitch diameter. Intuitively it feels like swapping in a smaller pinion would add slop to a system, it’d be interesting to hear people’s experience with that.
Worth noting as well that the really small pinions (9t, 8t) may have the same pitch diameter due to addendum modification, so you may be able to swap from a theoretical 9t to an 8t to get more reduction without changing plates.
As always get your numbers for pitch diameter and ctc distances from the manufacturer/distributor page and dint overly rely on generic ctc calculators, especially when working with small pinions.
EDIT: for more detail see the post below 
In the current era / using readily available pinions - it’s hard to pick the “wrong” one.
In general, depending on the vendor, you have the following options, all phased as “tooth count on XX Center”. Center refers to the effective center to center distance. So a 13 on 14 Center is a 13T 20DP gear that is functionally interchangeable with a 14T. List below:
8/9T on 10T Center
10/11/12 on 12T Center
12/13/14 on 14T Center
14/15/16 on 16T Center
Looking at the spread, you can essentially cover a 200% spread in tooth count from the bottom to the top - depending on the motor - Neo’s (V1, V1.1) are limited to 9T minimum, and even then the 9 is usually a press fit. Vortex’s, Krakens and Falcons can also run 8T minimums using either a Falcon or Spline XS Shaft.
Vortex’s also have an oddball pinion - that’s the 7T integral shaft, which acts like an 8T pinion.
Smaller motors 775’s, 550’s, BAGs are typically used with 8T and 6T pinions, or application specific (not 20DP) options. Usually you’ll have more luck in 32DP if you want to be really meta - 5mm/4mm/3mm shafts are common in the RC world, so you can more or less find whatever pinion you want, though they typically aren’t profile shifted, so you need to either make an adjustable mount or get the ratio right. (It’s been years since I personally have done a “peak power” ratio on anything, but there is a way to gear a small motor so that your operating speed is literally where the motor makes its peak power. This is where you start getting into obscure single tooth pinion differences.
Based on the options above, essentially you let the packaging of the gearbox drive your “input center to center, or input tooth count” - more literally, if you’re shooting for packaging within say, a 2x1 tube envelope (so the box doesn’t extend above or below the tube) that means the biggest input gear you can run - is a 38T input. If you agree that we call 2” a nominal clearance (I’m doing this because math is easier) then it’s a 40T input. (pitch dia on a 40 is 2”, the real OD is 2.1” but 40 is a pretty number)
Pair your largest input gear (40T) with your smallest pinion, and that will determine your input spacing. In this case, it’s 8:40, with the 8T being 8 on 10, or effectively, it spaces as a 10. That sets your design at 50T on the input spacing. From there you can use:
If you look at the spread above, you’ll see that you can cover a spread of ~235% with something like a 10% step or less - meaning that you can adjust the ratio in 10% adjustments with each pinion input pairing.
The Beauty of this is that it can be done with relatively few parts or changes. Realistically, you can pick “somewhere” in the spread for your design, then get the components to cover +/- some amount from the initial configuration, then adjust as needed, if needed.
From there, controlling the ratio at the input on larger reductions acts as a “coarse adjustment.” For things like swerve, or an intake, pinions will provide fine adjustment of the ratio. But when you’re compounding reductions - say doing a 3 stage, the minor adjustment at the input creates a large swing at the output. (Example, if your stage 2 and stage 3 are 25:1, shifting from 5:1 to 3:1 at the input is the difference between 125:1 and 75:1. So on big boxes, pinions aren’t the best place to fine tune, if that’s what you’re after)
Not all gears are available from all vendors, and for all motors, so you need to plot out your resource pool and see what aligns.
One last item of note:
Thanks to WCP’s and SDS’s most recent product drops, you can run (5) different pinions (12 through 16) with the SDS MK4/MK4i - which essentially allows you to run a higher over all gear option “slower” if you find that your DT is geared too fast. (Or draws too much current)
Personally, I like to “add speed” to a mechanism, rather than removing it. So typically I will design around the smallest pinion in a certain effective tooth count - usually the 10T on 12T - and then know that if the mechanism needs to be faster, and the current is reasonable, a quick speed boost is a swap away.
(Slower mechanisms are typically easier to control, and easier to diagnose. One you get it working well at slow speed, it’s easy to retune / optimize at high speed)
This is awesome, thanks for the writeup! There’s definitely a lot more flexibility added when factoring in swapping the input gear, which I forgot to do earlier. It does also make sense to gear slow in early design, I’ll take note of that for sure.
Picking pinions, I really just go for what works best for the application or what is currently on hand. A lot of the time, one or two teeth might not make a huge difference (imo), especially on big gear reductions like in a WCD gearbox. If they do, the other stages can be adjusted before manufacturing. Just run the numbers first and everything should be fine.