FRC 5987 Galactic Swerve

I’ve been getting a lot of responses both here and on the Discord telling me i should’ve used a Falcon or NEO 550 for the turning, with the main arguments being less reduction and weight. After looking at it more closely, i just wanna point out that neither of these motors would actually mean less reduction, at least in terms of stages- my module has 3 stages of reduction for the turning, which is really just a mistake on my part.
Looking at it again, i could’ve just as easily done the same ratio with 2 stages
A 2 stage ratio would in hindsight be better in every aspect- efficiency, weight, space, complexity etc. I made this mistake because i’m so used to the 775pro 12t 32dp pinions and didn’t even think of using a 20dp. If i get around to making a MK2 this is probably the first thing i’ll change.

But either way, this does mean that using a Falcon or NEO/NEO 550 really doesn’t help much since it would still require the same amount of stages in reduction and either just needlessly complicates the software side of things or costs too much

1 Like

Eh, REV’s API really isn’t any harder to use than CTRE’s, and a NEO 550 + SPARK MAX is $10 cheaper than a 775pro and Talon SRX. But while I wouldn’t use the 775pro for this since it’s beaten on weight, packaging, and price, you know your team’s needs and operating procedures better than us, and that should probably play a larger role in the steering motor choice than anything else.

This isn’t a commercial design, after all; if your team prefers 775pros to NEO 550s when the latter has enough power, then more power to you. There’s valid reasons for that, despite my earlier comments.

If the question is can you cantilever the pinion, yes you can. As far as the can the bevel gears be held by 2 separately machined parts, sure as long as there is a way to adjust depth and backlash. There really should be a provision to adjust gear depth and backlash similar to automobile ring and pinions. Looks like there are shims on the axle.

It looks like there are 4 bearings on the pinion. If the 2 end bearings are ball bearings then that’s all you need. The ball bearing closest to the pinion will take the trust load. Also note that no thrust retainment is required in the other direction if that helps making assembly easier. People freak when they see that the only reason our pinion shaft doesn’t fall out is because the bevel gear on the wheel is in the way.

KHK gears

The biggest issue is machining them to fit the wheel. We TIG weld the pinion to the shaft.

1 Like

While I wasn’t a part of the design, we already have 775pros, Talon SRXs and Mag encoders as Ari said. So buying eight motors and motor controllers (for two robots) is very expensive and probably not worth it.

I haven’t looked at the option of KHK shipping to Israel but that might be worth checking out

We have shipped to Israel before. The trick is not to use USPS because it takes a very long time to get through Israel post customs. I have had the best luck with DHL.

REV pushed out an update to allow for an alternate encoder.


Nearly every swerve design out there can use various drive motor options (brushed or brushless) and various steering motors (brushed or brushless). Changing between various CIM-class drive motors would be relatively easy (although some design choices in the module might restrict your choice of motor pinion and if motor pinions were not available at the correct PD for the motor you wanted to use, then you would need to tweak the design, I guess). Changing between steering motors is sometimes less simple, but with many designs, it is not that hard.

Our module is designed for a CIM class drive motor (although we did use 775s in 2017) and a 550 class steering motor. When we work with other teams that want to use our design or develop their own design using some of our design elements, we help them understand how they could use the various motor choices. Typically, each team will select their motors based on what they are most comfortable with and what they have in inventory. We switched from MiniCIMs to NEOs for drive last year without changing the design and switched from RS550s to NEO 550s for steering this year with the only change being the gear ratio of the VP gearbox. So the module design itself is able to handle these options. There are pros and con of each choice and ultimately, the choice to use the REV ecosystem for our team may not be the right choice for other teams.

Each team works within the options that they have.


Quick update: turns out i geared the encoder wrong :sweat_smile:
I replaced the encoder gears with custom 3D printed ones to make the ratio 1:1 with the turning pulley, seeing as i needed a size not available on Vex. Other than that, i’ve been hearing peoples’ criticisms and might make a Mk2 later on if i have the time for it.

The Grabcad files have been updated to reflects the changes

1 Like