2910 MK2 Swerve Module Release



In preparation for the 2019 season, Team 2910 Jack in the Bot is excited to release the CAD for our MK2 swerve module.

2910 MK2 Swerve Module, Versa Wheel .x_t
2910 MK2 Swerve Module, Machined Wheel.x_t

When viewed from the top, the module is pretty much symmetric across it’s diagonal. This makes it so there is no need for a mirrored version. The MK2 module’s mounting bolt pattern and ground clearance are the same as the Swerve Drive Specialties MK1 module. The mounting plate is designed to easily integrate with VersaFrame. The standard configuration is to mount the module on top of the frame. However, like the MK1 module, the module can be mounted below the frame for extra ground clearance.

The MK2 module also leverages the following COTS parts from the SDS MK1 module:
Both Wheel forks
Wheel spacer
60T Bevel gear

The weight of the module is 4.9 pounds.

Using COTS gears, the available drive gear ratios result in unadjusted free speeds of:
12.5 ft/s
13.5 ft/s
15.0 ft/s
16.1 ft/s

Like the MK1 module, the MK2 module uses a 4" VersaWheel. We also have a custom aluminum wheel designed for 1" width nitrile tread.

The gear ratio for the steering of the module is 18:1. This ratio is pretty low, but the NEO should have more than enough torque to overcome this. We will likely dial back the steering NEO’s output in some way to make the PID more controllable.

We are not sure if this module will be viable for the 2019 season because the NEO and Spark Max are very new and we have not tested how they will work in this application. We figured we would release it just in case.

The module can also be used with 2 CIMs or mini CIMs with a CIMcoder to measure driven distance. This makes the module much heavier and quite a bit taller. At that point the MK1 module probably makes more sense to use.


Here is a picture showing the steering encoder better.


that machining is GORGEOUS! What machines did you use?


Thanks, All the machining was done on a Tormach PCNC 440.


Is there a reason that you used specifically a Neo for the turning motor? It seems like it just makes the module much more expensive and throws a lot of unnecessary power into the system.


I’ll hazard a guess that they didn’t want to have to go for the comparatively larger reduction of a 775pro. Additionally, it lets them get away with having 2 of the same motor (i.e. making it easier to keep spares). With the extra stage required, I wouldn’t be surprised if the 775pro version were significantly heavier.

If I were to nitpick this design, it would be really cool if the drive and turning motor shared the same pinion, so that replacements could come from the same reserve.


The main reason for using the NEO is that the higher torque, lower speed nature allows for the simple 2 stage reduction. The NEO weighs just a little more than a Bag or 775. It also makes the module look nice and symmetric. It would never be used to its full power potential in this application, but that’s ok. There really isn’t anything lost by using a over powered motor.

The Neo + Spark Max is comparable in price to a BAG + Talon SRX. If the NEO lets you get away without a VersaPlanetary then that is a big cost saver.



That totally makes sense, I was mostly just thinking that a lot of teams have SRXs, BAGs, VPs, etc, but as we move further from the release of the NEO that becomes less and less relevant


@sanddrag Thanks! Full CAD is available at the top of this thread if you want to check out the module in detail.

The large bevel gear is from Swerve Drive Specialties. It is a machined down khk SB1.5-6015.

The big steering bearing is a X-Contact thin section bearing equivalent in size to a Kaydon KA035XP0 or Silverthin SA035XPO also from Swerve Drive Specialties.


This is really a phenomenal design, thanks for posting it in so much detail! When the end result is so simple and elegant, it’s hard for the casual observer to see how much work went into it. It looks absolutely bombproof. Have you put any thought into a 2 speed version?


@s_forbes Thank you. We went through quite a few concepts and designs before making one we were happy enough with to designate as the MK2, figure out all the manufacturing for, and actually build.

I haven’t put much though into a 2-speed version, but the layout of this module with the extra drive gear stage seems much better for 2-speed then the MK1 module.


I really appreciate that you suspended your module with (what appear to be) picture framing wire for the photo shoot. Simple but thoughtful touch that made the presentation of this post even more professional. Seeing the module freestanding in the photos is really cool.


Have you done any driving testing with these yet?

I’m specifically curious on the effects of the NEO requiring zero speed before direction changes due to the limits of the spark.


FRC Blog - 2019 Motor Controllers and MXP read from post 349 on.


See their post in the 2019 motor options thread for a video of the NEO/Spark steering.


We have not done any actual driving tests yet. We just started testing yesterday with getting the steering of the module up and working. REV has sent us some firmware modifications that they expect will help with the direction change issue we are encountering at the moment. The plan is to work on this tonight.

Depending on if the game is good for swerve or not, all development work for this module will either continue or come to an abrupt halt until the post season.


Fantastic, thanks.

I love this module, and I’m hoping that this game is good for Swerve. The NEO opens some interesting packaging options that weren’t quite doable with CIMs.


With the game being pretty swerve friendly I’m happy to say that development with this module has continued. The new Spark Max firmware has fixed the direction change issue we were seeing. We quickly threw together this chassis to get our programmers going. It’s not pretty, but it is up and running and working great.


Those modules are VERY compact. Having lots of space inside a base for things other than driving is pretty excellent!