Hello CD! I have seen posts about teams using the AndyMark Cim Encoder. Our team wanted to try it out this year, but the stocks were out during the build season. So, can anyone provide feedback on the encoders? Were there any issues? Thanks in advance.
I’ve heard a lot of complaints about them here on CD, but my team used them successfully the past two years. They are pretty convenient in that they’re pretty much plug-and-play mechanically.
Electronically, they leave some things to be desired. The tiny molex connectors are easily damaged and have no business being on FRC robots. The encoder disk is only held on by a tiny set screw, and can come loose with vibration. The included cable has single header pins (instead of groups of 3 like normal PWM cables), which mean you have to be careful that they stay connected
We had one failure with them this season that messed up out auto routine and caused us to lose semifinals at isde2. I didn’t help fix the issue, but I know that it was fixed before we packed up and hasn’t failed again through 3 competitions. Overall, I’d say they’re pretty decent so long as you take a lot of care to make sure they’re properly installed and maintained. If you have the resources to use a more reliable encoder* I’d probably recommend that instead.
*I’ve heard very good things about SRX encoders
I’m partial to the Grayhill 63R encoders that VEX recommends with their 3CIM ball shifters. They’re easy to install and easy to connect to a Talon SRX using the Universal Breakout board from AndyMark.
In my experience the CIMcoders are hell to wire and take a lot of fussing with to get any useful data out of them.
We tried CIMcoders on our shooter back in 2016 and never got them to work. Thinking back, not sure if it was due to an inexperienced team or actual issues.
We’ve used CIM Encoders because there is on good place to hang an encoder on our swerve and we like where they package on the front of the motor. The 2 problems we’ve had is loose set screws and unglued magnets. We do the following to make them more robust:
- Broach a keyway into the hub of the magnet 180 degrees from the set screw.
- Replace the set screw with a slightly longer one.
- Cut a key stock that is the full length of the keyway in the CIM motors shaft.
- Make sure the magnet height is correct and it doesn’t drag on either side of the housing. Slop in the thrust of the CIM can affect this.
- Use a tiny bit of pink or blue loctite on the set screw.
We don’t count on the set screw only. With the key extended down into the magnet, the magnet will turn no matter what, even if the set screw comes loose.
We think the magnets coming unglued is from rubbing on the housing, usually because a set screw has come loose. That’s why it’s important that they don’t rub.
I don’t like to speak badly of products but this is the only one I’ve ever returned for a refund. It has a lot of room for improvement.
- The first set of magnetic discs delivered with them were too small.
- The second set of magnetic discs were also sometimes small but also had other issues (see other users comments above).
- There are little black spacers that are square-ish that need to be used. This is a problem if you happen to install the encoder over one of them without lining it up correctly. I have no idea why it isn’t a square or better yet, a circle. Failure to align followed by even a touch of tightening will crack the encoder housing.
- Lastly, they feel like a cheap product and they were made to a budget rather than being made to a purpose.
Can you use these successfully? Probably but why bother when you can use something like the CTRE SRX mag encoder and just drill a .25” hole on a shaft? If you don’t have a lathe for this operation then stop looking at encoders and get a lathe first.
For our drivetrain this year we were originally planning on putting an SRX mag encoder on one of the non-gearbox drive shafts. We realized later in the design process that we needed that space for our intake and we didn’t have space on our drive gearbox to put one one the back of the output shaft. Adding the CIMcoder was easy as an afterthought, and it worked pretty well giving it a plan B. I wouldn’t purposefully use one, but if the same situation should arise next year I would use them again.
For some teams, this isn’t an option. Even if we start looking at economical options, it’s simply too much of a safety risk to have one on-site without a trained mentor. It’s one of the few tools that 5202 had on our hypothetical ‘never-buy’ list.
We used them this year on our 3x-MiniCIM drive into WCP SS gearboxes, because they’d been on my “to-try” list for a while.
The construction is pretty cheap, but that’s because they only cost $42, don’t look a gift horse in the mouth.
We didn’t have magnet sizing issues.
The molex connector is pretty dinky, but it didn’t actually fail on us. I was suspicious of the hair-thin wires, but they didn’t fail either.
What we did have trouble with was getting enough tension/preload on the two bolts to have them stay in place and hold the motor for more than about six or eight matches at a time - even with loctite. I’m most suspicious of the plastic casing, with secondary concern being whether we actually got our bolt length correct with the added thickness.
If we use them again, I’ll be pushing for us to drill out the areas around each bolt hole and put in 1/2" aluminum plugs to get a nice surface for bolt preload. Similar to how the VersaPlanetary encoder runs aluminum tubes to get preload through the plastic-bodied encoder stage.
Side note, you’re measuring at the motor instead of the final actuation.
So, it’ll let all of your backlash from (mini)CIM to wheel go un-measured. We definitely had more backlash in the WCP SS than I was used to from previous custom setups, with a 6" wheel we were seeing almost an inch of slop, but I don’t think this really held us back in any particular way - like one of the Citrus students pointed out about pathfollowing & recalculation in the 1678 code release thread, “sometimes close is good enough”.
Any mechanical failure will also go un-measured, so if your top miniCIM with encoder stage is popping out of engagement on the cluster due to a loose bolt, everything will be very confusing… 
Between the mechanical annoyance and our coders preference we measure the final actuation wherever possible, we’re going to be going back to encoders on the final drive axles in the future…
Our Harbour Freight 7x12 with ~600W motor is primarily a safety risk to my knuckles due to the size/packaging of the control wheels (the safety interlock on the chuck guard works really well), and it has made a surprisingly valuable addition to our shop, including several appearances for custom drive gearbox shafts (with embarrassing runout, but still worked).
I’m also a trained mentor, so perhaps I don’t understand how hard getting safety right can be without that background to fall back on.
[/offtopic]
1072 tried using them in 2017 and had random issues where simply tightening the CIM onto our gearbox would result in a loss of readings. This occurred on 2/4 CIMcoders. I would not recommend them to any team looking to get into sensors.
The Vex Ballshifter + Mag Encoder or Grayhill combo is a much safer option.