I searched litle bit in chiefdelphi but I can’t find the exact same post.
We are planing to use cancoder for our elevator in offseason but we dont have any encoder to use. We also buy some off these encoder but which encoder is best for Elevator. We might to buy REV Bore Encoder or another magnetic encoder.
I would either use
•A through bore encoder, I’d reccomemened either the rev through bore or the helium canandcoder with the through bore kit
•The integrated encoders of the motor
There are 9 to 1 ratio on elevator.
The integrated encoders could be give wrong data or maybe not but there are 2 stage gearbox right there
We’ve happened to use that specific Rev THrough Bore Encoder on major components of our robot the past 2 years and have nothing but good things to say about it.
I’ve never had an issue where it disconnects. seems to be very accurate. Not a terrible price point either.
I’d use that and move on to the next problem.
I’m also just a big fan of that hex mounting on the shaft. Not that its the only encoder that would mount like that. Just makes it nice and simple. You’ll never have to worry about it not being 1 to 1 with the shaft movement.
We are thinking that option as well, but we have about $100, and this encoder costs $48, so I’m not sure if we want to spend half of our budget on it. Additionally, with the added shipping and customs fees, we might encounter some difficulties.
With the budget you have available, have you considered using a potentiometer and rope?
Nope we didn’t think that.
ah yeah. thats tough. Can’t really speak much to it. I’m sure there might be cheaper encoders elsewhere.
you also might be able to find another team with this used. not really sure.
$100 unfortunately is a pretty tight budget. If your saying its for specifically your ‘sensor’ budget I’d still say do it haha. That should really be all you need for the elevator. 0 it low and know the max amount of rotations it can go up to.
Depending on how you drive this thing though… and if your using a motor with an integrated encoder like a Falcon500 or a Neo you might not need that encoder.
I don’t know your full system though so I can’t really speak to how it works
Firstly our whole budget is $100 not the sensor budget. We are using two Neos with 9.1 ratio with 2 stage.
The best encoder is:
The mag encoder you show in your model won’t work as shown because you can’t mount it to a bearing. It doesn’t mean that you can’t devise a mount that straddles the bearing. There might even be a mount you can buy. You would also need to bore an axial hole in your hex shaft to mount the magnet. Easy if you have a lathe. Not so easy if you don’t.
The Rev Through-bore encoder was designed for an application just like this. it’s available, at least in the USA.
You can also use encoders built in to your motor or maybe even your gearbox (e.g Versaplanetary with encoder slice). In this application, using the built-in encoder on a NEO or Falcon motor is maybe the best option if you are using a motor like that.
Remember that with any relative encoder application, you need some way to know the initial position of the elevator. It might be sufficient to ensure that the elevator is fully retracted when powering up the robot, but some teams like to add one or more limit switches to the elevator to get direct feedback on elevator absolute position.
you could drive purely off that then if you want to.
Your software would know from a zero point where the integrated neo encoder is, I’m assuming at the start of a match your robot would always start with the elevator down anyways. From that point you know how many ticks up is to the top of your elevator.
Is this is the case don’t even bother with the rev encoder. I think technically it might be a little easier to understand since the shaft mounted encoder would be a little more 1:1, but its certainly doable without!
Just realizing I basically just said the same as @paulonis …
The mag encder is just a example in photo.
When we use neo encoder in regionals with 65.1 ratio it didn’t give correct data for our system.
We might use some limit switches for absoulete position for elevator.
You really should just use the built in encoder on the neo. it’s far easier and has less failure points and assembles faster. it also allows you to use the onboard pid control of the spark max.
what was the mechanism you had issue with a 65-1. with an elevator we have always used integrated encoders on brushless motors since they became available and prefer them.
The encoders in your NEOs are going to be very reliable. Brushless motors rely on the electronic shaft position data to operate. If the encoder is not working, the motor is not running.
The gearbox behind the motor is not a problem as long as you aren’t breaking and skipping teeth in the gearbox. There is a direct relationship between the motor shaft and the gearbox output shaft. You’d be relying on the same kind of direct relationship between the hex shaft rotations and the elevator height. There is a much greater chance of skipping teeth on the chain or belt drive of your elevator than there is of skipping gear teeth in the gearbox.
Wouldn’t a 9 to 1 ratio be a problem?
Not a problem at all. If anything, it increases the resolution of your elevator height computation (not that you need better resolution). With an encoder on your hex shaft, you would be computing height based on rotations of the hex shaft. With the encoder on your motor you compute height based on rotations of the motor shaft. It’s just 9 times more rotations if you have a 9:1 gearbox.
Nah that should be fine still. That’s plenty of resolution.
Our setpoints were always changing, but the values we provided were always the same.
can you describe the mechanism or provide a picture. it may be something else in the system was causing your inaccuracies. if we can identify why you had the issue with that system it will help prevent it in the future. by default using the motor integrated encoder will not cause that.
