Getting Started with Sprockets and Chain

[romanddimov]

Hi Guys!

For some reason I can't post through my own account, so I'm posting through my friend's.

My name is Somer Kwon and I am a hardware member of Team 5700, the SOTA Cyberdragons. During this year’s off-season we wanted to expand our knowledge of robotics. We decided to break into 3 groups and begin growing our knowledge in 3 areas:: gear boxes, chain and sprocket, and flywheel and other shooters. My group is working on learning the applications of chain and sprockets.


As a proof of concept, my team plans to build an 18” arm that can lift a 5 lb weight at the end at a controllable speed We would like to use a sprocket and chain system from a gearbox to actuate the arm. We’re looking at 971’s 2016 intake CAD as an example.


The basic power and shop tools we have access to include: a lathe and mill (w/o DRO), a vertical bandsaw, chopsaw, bench grinder, and maybe a water jet (at TechShop).


So far, we have decided to use a RS775 18V motor through a 180:1 reduction. This would provide plenty of torque and a 1 rps angular velocity, which seems to be slow enough, at 20 amps. This reduction seems really huge and we’re not sure if we did the correct calculations. Also, how would we be able to maintain a 20 A current? Since motor controllers control motor speed by changing average voltage with PWM, is it possible to change the input voltage to reduce the speed of the arm and maintain enough torque, therefore reducing the needed reduction?


On the mechanical side of things, we need help with choosing what sprockets, gear boxes, and chain we should use for this project. How do you choose between 25 and 35 chain? What are the best products to use, and where should we purchase this hardware. Tensioning, adjusting chain length, sprocket mounting, greasing, chain and gearbox efficiency, etc., are all new concepts we have little experience with. If you have any tips, resources, videos, or comments about how to work with and think about chain and sprockets, that would be extremely helpful for us.


Thank you for your help!


Somer Kwon
Mech En | SOTA Cyberdragons Robotics, Team 5700
sotacyberdragons@gmail.com

[Owen Busler]

This past season my team had a shooter arm rotate about a shaft powered by a chain and sprocket. The first step in a successful design is selecting the correct motor and reduction. The RS775 is a great motor but it doesn't do the best when stalled as its fan needs to be spinning for it to stay cool. You can check out the amazing resources at Vex Pro (http://motors.vex.com/) to see the very elaborate motor testing they have performed. Towards the bottom of this page (http://motors.vex.com/bb-775) you can see the locked rotor stall test. The graph shows that over just 20-30 seconds a stalled 775 will burn out and provide no torque. Taking that into consideration, I would recommend a bag motor (http://www.vexrobotics.com/217-3351.html) Although it has a lot less power, it is okay with being stalled for a little longer (http://motors.vex.com/bag-motor)

That being said, there are some amazing resources out there to help you design a chain driven arm. The first and a tool I use very often is JVN's design calculator. Don't forget you can get a reduction in your gearbox and in the sprocket.

https://www.chiefdelphi.com/media/papers/3188

Using this very simple spreadsheet you can mess around with different motors and reductions until you get the speed and power draw you are looking for. I suggest you use the "Rotary Mechanism" tab for your current project.

As for what gearbox to use, there are quite a few choices but my team and many others have successfully used vex pro VersaPlanetary gearboxes (http://www.vexrobotics.com/vexpro/mo...planetary.html) they are very flexible and can be used in many different gear ratios.

Finally, once you have your reduction picked out you will need to calculate the spacing between the two sprockets. This site (http://www.botlanta.org/converters/d.../sprocket.html) has worked well for us. I would use your mill with a DRO to machine a perfect center to center between the sprockets. If you choose not to do this, you can always use a chain tensioner of some sorts, there are many different designs out there.

[Cothron Theiss]

Since you're powering an arm, I'd check out this thread about passive assistance.

https://www.chiefdelphi.com/forums/s...&highlight=jvn

There's some good advice in there on how to reduce the amount of power you need running through your system, which is always a good thing to consider.

[DonRotolo]

#25 chain is plenty strong for moat any FRC application, but due to the relatively small tooth size on the sprockets, the chain tension is very important. It weighs less than #25. Also due to the smaller tooth size, you can get more teeth on a sprocket of a given diameter.

#35 chain is considerably more forgiving on chain tension, but it weighs more than #25. It is also a bit easier to work with since things are not as small. It is stronger, but you generally do not need such strength.

Why are tooth counts on a sprocket important? The ratio of speed for sprockets goes by tooth count. For example, a 15 tooth sprocket turning a 30 tooth sprocket (via chain...) has a 2:1 ratio of shaft speeds. A 72-tooth sprocket is fairly large in #35, more reasonable in #25. McMaster.com can show you the range of sprocket sizes available.

[MichaelBick]

Originally Posted by Owen Busler

The graph shows that over just 20-30 seconds a stalled 775 will burn out and provide no torque. Taking that into consideration, I would recommend a bag motor (http://www.vexrobotics.com/217-3351.html) Although it has a lot less power, it is okay with being stalled for a little longer (http://motors.vex.com/bag-motor)

There is a bit more that goes into this decision. If you look at the graphs 775Pros don't really mind stalling @4V. If your mechanism design/gear ratios don't need more than .237N*m (AKA .71N*m * 4V/12V) of torque while stalled, you should be good.

[snorthey]

some real world advice...

use the chain that is easiest available... look to industry around you for parts...

35 is easy to find parts and links in hardware stores...

look to old farm machines for great examples of ho to convvert speed to force and back again.. look at specs for a square baler or corn picker... both are driven off 1 or at most 2 power sources (the PTO)& (second can be the wheel)
all the moving parts are driven from that one source...

BEWARE of Fingers...

[Greg Needel]

If you are looking to learn the chain basics we (REV Robotics) released a chain guide with our products for FTC this season. Since we use #25 chain in those robots also the information is directly applicable to FRC as well.


http://www.revrobotics.com/content/d...cket-Guide.pdf

[asid61]

Originally Posted by MichaelBick

There is a bit more that goes into this decision. If you look at the graphs 775Pros don't really mind stalling @4V. If your mechanism design/gear ratios don't need more than .143N*m (AKA .43N*m * 4V/12V) of torque while stalled, you should be good.

Adding onto this, because the BAG motor and 775pro motor have the same mass, they theoretically should be able to sustain the same/similar current draw for the same period of time before burning out (assuming they're solid blocks of the same material, surface area, etc. which they are not). That means a 775pro stalled at 40 amps (or about 4v) is equivalent to stalling a BAG motor at 12v (also 40a). With that in mind, I try not to use BAG motors anymore because the 775pro also has the advantage of ball bearings and the integrated fan in the back, not to mention the higher stall torque/current if I want to use it.
It's not just stalling a motor that kills it; heat kills it, which is proportional to current. A 775pro stalled at 4v is the same as stalling a BAG at 12v in that regard. You can "stall" a CIM at 0.5v, but it's not going to kill the motor any time soon.

To the OP: 775pro > RS-775 for a variety of reasons, but you can make any motor work. If you want to save money you might be able to dig up a CIM (many teams have them in excess). 180:1 may not be high enough if you want to control it accurately. As a rule of thumb, I like to make sure I'm only running at 5-6% of a motor's stall current (or about 7 amps for a 775pro) to make something very easy to control. JVN Calculator 2016 makes these calculations a breeze. Then you can use a simple PID controller to control your position. This is very conservative, but for teams just starting out with encoders and controlling things accurately I like it.
For a low load like yours on an 18" arm, I'm getting a number around 600:1 for a speed of 0.5rps to maintain 5-6% stall current, but JVN calculator will help you narrow down your options.

[Ari423]

Originally Posted by snorthey

some real world advice...

use the chain that is easiest available... look to industry around you for parts...

35 is easy to find parts and links in hardware stores...

look to old farm machines for great examples of ho to convvert speed to force and back again.. look at specs for a square baler or corn picker... both are driven off 1 or at most 2 power sources (the PTO)& (second can be the wheel)
all the moving parts are driven from that one source...

BEWARE of Fingers...

This can either be good advice or really bad advice, depending mostly on where you live. I can't speak to the availability of corn picker parts in MN or San Francisco, but I know that if you wanted to find a specific sprocket from a corn picker near Philadelphia you'd be SOL. That means that if you do find one and it breaks, it will probably take a while to get an exact replacement. If it breaks at a competition, there probably won't be any other teams there that have a spare one to lend you. And if you have to modify it in any way so it interfaces properly with FRC parts it's no longer COST so any extras would count towards your withholding.

What I would recommend is to use parts from a standard FRC supplier (VexPro has a huge supply of sprockets and chain to choose from), even if sometimes they may be slightly more expensive. With FRC suppliers you know the quality of the product, they have amazing customer service and fast shipping, and the parts are standard so it greatly increases the likelihood of bring able to find a replacement when you badly need one. This isn't to say that industry parts won't work (god knows I've stuck a few of them on robots in the past) just that in most cases there are better options.

[snorthey]

to clarify.... I did not say use machinery parts.. I stated look at examples of how it works.. often we can stan on the knowlege of others and not recreate the wheel... for that matter look at a 21 speed bike.. same concepts of how to set up gears... to get the most movement or power depending onthe need...

then go and find approved parts to do the job...


good luck... READ BEFORE your comment... not SKIM...