View Full Version : Gear Teeth Clearance
So my team is building a gear system and we had to make the solidworks for the individual gears ourselves. What should be a clearance between each tooth for the robot to function? The gears are small, 1.5 inch to 0.75 inch. Can we get away with 0.017" of clearance after running them for a while so they set oooor nooo...?
Pitch Diameter?
Number of teeth?
Spur Gear?
What are you using to lay out your gear geometry in solidworks?
Can you provide more detail like material, intended use, or how the gears will be made?
It sounds like you're trying to design your own gear tooth profile, which is a really bad idea. Instead, try using existing gear pitches/modules.
Spur Gear, Pitch diameter=1.5/16? 13 to 12 tooth, then 12 to 32 tooth, Aluminium (or maybe steel) CNC milled or waterjet
Intended use is our little rookie secret but it's for a major part of our robot which will see alot of motion
https://www.bostongear.com/pdf/gear_theory.pdf
This should help with some of the clearance (backlash?) questions you have, as it would have been what I would use for reference. If you have a copy of the Machinery Handbook it would also help and provide guidance. I had said Pitch Diameter, but really the Diametral Pitch you choose for the gears determines what diameter you have, as well as the amount of backlash you should expect, though the criticality of the backlash is only if you are going to move the mechanism back and forth often.
Also keep in mind manufacturing tolerances tied to the machine you are making them with. The teeth may grow or shrink due to this, either increasing or decreasing the clearance you experience.
I would also recommend not making gears out of aluminum unless you plan on anodizing them. Aluminum on aluminum tends to gall.
I'll see what we can do for steel then, thanks
I am definitely not an expert at gear design. I have designed some limited gear systems for FRC in the past, but not with custom made gears. I do have some experience with custom made sprockets, which I used the Machinery/Machinists Handbook as guidance (back in 2001 when I was still in high school). I think most CAD programs tend to have "canned" operations to make gear/sprocket profiles now.
Yeah I think so, we just used the circular pattern to line the gear teeth around the base radius then went from there
DonRotolo
27-01-2016, 19:48
I'll see what we can do for steel then, thanks
Brass is a reasonable alternative if strength is not a big issue
I would strongly recommend reconsidering this approach. Your rookie year is hard enough (even if you have some experienced mentors). Don't screw around doing something exotic you've never done before. Get COTS gears. Modify them if you really need to.
You will not be able to mill or waterjet a true involute tooth profile. If the gears in question only mesh with each other, an approximation would be fine. The best way to do this yourself is to get involute tooth cutters for the mill and a indexer so that you can cut them. This requires a series of cutters to cover all possible tooth sizes.
Second what Cory said. If you really want to stick to it though, Solidworks Toolbox should be able to generate you a gear model without having to sketch an involute profile.
We'll Show you WE'LL SHOW ALL OF YOU.
See you at worlds
With love: 5870 <3
I would strongly recommend reconsidering this approach. Your rookie year is hard enough (even if you have some experienced mentors). Don't screw around doing something exotic you've never done before. Get COTS gears. Modify them if you really need to.
You will not be able to mill or waterjet a true involute tooth profile. If the gears in question only mesh with each other, an approximation would be fine. The best way to do this yourself is to get involute tooth cutters for the mill and a indexer so that you can cut them. This requires a series of cutters to cover all possible tooth sizes.
Ditto,
Making gears is work, I mean legit work. Would recommend just buying some off the shelf.
We'll Show you WE'LL SHOW ALL OF YOU.
See you at worlds
With love: 5870 <3
Listen to the smart people above.
To get gears that are as accurate as any COTS Vex or Andymark gear you would need involute gear cutters.
CNC mills would need a tiny endmill to do gears, and waterjet is inaccurate and has an extreme taper. If you buy a COTS gear, the two pitch circles should be tangent to each other exactly. But using your method, I would want 0.004" or more of clearance depending on how you are making them.
Your gears will be less efficient and more likely to bind than any COTS gear, not to mention will take far more hours than COTS.
We'll Show you WE'LL SHOW ALL OF YOU.
See you at worlds
With love: 5870 <3
Consider what you're getting from making your own gears. I find it very hard to believe that you can't just find a gear with the right gear teeth from Vexpro, WCP, Andymark, SDP-SI, etc and modify it to fit your design. Unless you want to make gears just for its own sake, I don't see any reason to try it. It wastes a lot of time and labor resources on something that won't give you any competitive edge.
You have been warned by many others smarter than me above.:p
waterjet is inaccurate and has an extreme taper
That's not strictly true. A good waterjet will produce little taper on a 1/4" or so thick part. One with a tilting head will effectively produce zero taper and hold .001-.002" overall tolerances. You certainly can cut a gear that will run at relatively low speed, but it will only be an approximation of the involute tooth profile and it will mesh relatively poorly with a COTS spur gear. I had some very large ring gears waterjetted for a project at work and they worked well.
That being said, whatever the OP is trying to do can almost certainly be done faster, easier, and with higher quality by adapting the design to use COTS gears, even if they are steel Martin gears that require putting a custom bore in.
Monochron
28-01-2016, 03:51
We'll Show you WE'LL SHOW ALL OF YOU.
See you at worlds
With love: 5870 <3Put bluntly, there is a VERY good chance that you are wasting your time with endevour. If your team has the massive skills to design their own gears and gearboxes in your rookie year, then you can definitely turn that skill towards building innovative and robust mechanisms that blow everyone else out of the water. Seriously, consider using a couple COTS parts instead.
pmangels17
28-01-2016, 10:32
For the love of all that is good, don't waste your time making your own gears. It is your first season, and frankly, I have never heard of any team that has made their own gears and used them successfully on a high-load mechanism. This doesn't mean it can't be done, but you can save yourself a ton of time and money and just buy them online. You will be much happier.
Joe Johnson
28-01-2016, 13:58
I agree with all those who encourage the OP to find another way but if you won't be discouraged, I say more power to you. Live and learn. Either you learn or you show us and we learn.
I would ask the OP not to bet the entire season in it if you can help it. Have a back up plan, another source for the product or another option. Something to keep from having your robot be a box of rocks bot.
YMMV.
Dr. Joe J.
Chris Fultz
28-01-2016, 22:54
And if you decide to use COTS gears (lots of good sources), a good general guide for gear spacing is (1/2)PD1 + 1/2(PD2) + .003. (in inches).
PD1 is Pitch Diameter of Gear 1, PD2 is Pitch Diameter of Gear 2.
Edited - i meant .003!
346CADmen
29-01-2016, 08:06
We'll Show you WE'LL SHOW ALL OF YOU.
See you at worlds
With love: 5870 <3
I hope it works out well for you, I really do.
Perhaps you have a very unique mechanism and making gears is the only way you see things working. But look at all the sources for off the shelf parts.
I design gears for work, we manufacture custom gears. It is expensive and hard work, even with the proper machinery and tools. If you can buy gears I would think you would save time and money.
Sperkowsky
29-01-2016, 08:19
Op you are completely disregarding people like Cory from 254 advice. That's for lack of batter words very stupid. Very stupid. If you did not know 254 is one of the best teams in Frc and makes a lot of their own parts. Making your own gears is useless imo when there are so many stock options and you will probably have more issues with them then its worth. What you may or may not have realized yet is the frc community is helpful. As a rookie team you do not know what your doing.
But I can't force you to make a better decision you will have to learn the hard way.
BeardyMentor
29-01-2016, 09:10
Don't cut your own gears. There is a 99.9% chance that you are spending your time on something that could be done much more easily with COTS parts. I have all the right equipment and I still wont do it if there is ANY way I can get the pitch and tooth count I need from an off the shelf part. Revisit your design, shop around, call suppliers, do what you need to do to not have to cut your own gears.
JackFisher
29-01-2016, 19:28
All cots gears are the best way to go. But if you still insist on generating your own gears, I would use a website called rush gear that has software for generating custom gears quickly. That way you have an accurate cad file to reference, create dxf's, etc.
http://www.rushgears.com
Again, I'll re-emohasize cots gears are the way to go.
Op you are completely disregarding people like Cory from 254 advice. That's for lack of batter words very stupid. Very stupid. If you did not know 254 is one of the best teams in Frc and makes a lot of their own parts. Making your own gears is useless imo when there are so many stock options and you will probably have more issues with them then its worth. What you may or may not have realized yet is the frc community is helpful. As a rookie team you do not know what your doing.
But I can't force you to make a better decision you will have to learn the hard way.
Thank you for this and all the positive input.
Billfred
08-04-2016, 15:42
And if you decide to use COTS gears (lots of good sources), a good general guide for gear spacing is (1/2)PD1 + 1/2(PD2) + .006. (in inches).
PD1 is Pitch Diameter of Gear 1, PD2 is Pitch Diameter of Gear 2.
I know the additional .006" is some fudge factor, effectively trading gear life for less friction in operation. But the number I'd learned from oral tradition* is .003" or so. What are the benefits of adding the extra distance?
*Even before I learned this from other FIRSTers, I knew a variation on this from racing R/C vehicles, which have a similar opinion of gear life vs. speed**; there, the conventional wisdom is to run a piece of paper between your pinion and spur gears to ensure "proper" spacing--and a piece of paper is about .002" thick.
**If you haven't gathered from the "racing" part, they skew towards speed.
IronicDeadBird
08-04-2016, 15:43
We'll Show you WE'LL SHOW ALL OF YOU.
See you at worlds
With love: 5870 <3
Bring the heat mate!
Although I'd have a concern about how crucial the gear is because another advantage to COTS or close to COTS is the ability to not have to cut into your withholding allowance. While the success of the part may at this point in time be unknown, learning is always worth it.
Just don't go mental and if you don't get it right in time, consider making it an off season refinement.
Just a quick note, this was from over a month ago. We went with vex gears after being suggested
IronicDeadBird
08-04-2016, 15:57
Just a quick note, this was from over a month ago. We went with vex gears after being suggested
Care to share what the original intent was or any details?
Michael Hill
08-04-2016, 15:58
I know the additional .006" is some fudge factor, effectively trading gear life for less friction in operation. But the number I'd learned from oral tradition* is .003" or so. What are the benefits of adding the extra distance?
*Even before I learned this from other FIRSTers, I knew a variation on this from racing R/C vehicles, which have a similar opinion of gear life vs. speed**; there, the conventional wisdom is to run a piece of paper between your pinion and spur gears to ensure "proper" spacing--and a piece of paper is about .002" thick.
**If you haven't gathered from the "racing" part, they skew towards speed.
Extra distance will buy you more tolerance in manufacturing at the cost of more backlash. That said, .003" we usually use. Vex gears (the ones we usually get for custom gearboxes) are AGMA 11 quality, which means their total composite tolerance is < 0.001" for all gear sizes (http://www.americangearinc.com/quality.html). You could get away with 0.002" spacing, but you're leaving yourself with no tolerance for the manufacture of your gearbox plate. Our gearboxes are cut by either waterjet or CNC mill, so we can hold somewhat tight tolerances. However, Even if the waterjet was slightly out of tolerance, it's still probably fine because it (should) be unlikely that both gears are > 0.0007" too large. Additionally, you'll get a bit of wear that makes things run fairly smoothly.
Care to share what the original intent was or any details?
Yeah sure! Triwheels, Its actically the only thing that makes us not KOP sincethe space where our intake was going was shortened by 4 inches due to bumper crew misreading a rule. I have numerous pictures of the wheels on my cd file and media has uploaded numerous videos on our blue alliance of the wheels actually working. Check em if you wan. They're my babies <3
TheModMaster8
09-04-2016, 12:11
Take a paper towel or toilet paper and stick it between the teeth as you mesh them then remove the paper and you have the correct distance
Take a paper towel or toilet paper and stick it between the teeth as you mesh them then remove the paper and you have the correct distance
Actually an amazing tip thanks!
Michael Hill
09-04-2016, 14:26
If you want to do it like the old school clock makers, you could use a pinion head depthing tool: http://www.clickspringprojects.com/pinion-head-depthing-tool.html (Also, I HIGHLY suggest subscribing to that guy's YouTube channel.)
Chris Fultz
09-04-2016, 15:08
I know the additional .006" is some fudge factor, effectively trading gear life for less friction in operation. But the number I'd learned from oral tradition* is .003" or so. What are the benefits of adding the extra distance?
My error in that number. i meant .003" and i corrected my earlier post.
Extra distance will buy you more tolerance in manufacturing at the cost of more backlash. That said, .003" we usually use. Vex gears (the ones we usually get for custom gearboxes) are AGMA 11 quality, which means their total composite tolerance is < 0.001" for all gear sizes (http://www.americangearinc.com/quality.html). You could get away with 0.002" spacing, but you're leaving yourself with no tolerance for the manufacture of your gearbox plate. Our gearboxes are cut by either waterjet or CNC mill, so we can hold somewhat tight tolerances. However, Even if the waterjet was slightly out of tolerance, it's still probably fine because it (should) be unlikely that both gears are > 0.0007" too large. Additionally, you'll get a bit of wear that makes things run fairly smoothly.
The Vex gears may be AGMA 11, but it looks like they make them with a huge root clearance (maybe somebody can confirm?). We've tested gears anywhere from -0.007" to +0.010" without issues, so we design with exact C-C to simplify the machining and design a bit.
Michael Hill
09-04-2016, 21:10
The Vex gears may be AGMA 11, but it looks like they make them with a huge root clearance (maybe somebody can confirm?). We've tested gears anywhere from -0.007" to +0.010" without issues, so we design with exact C-C to simplify the machining and design a bit.
I believe it's something like 0.503 hex rather than 0.500, so that buys you some tolerance depending on your hex stock. Unfortunately, nobody seems to want to publish much tolerance information other than "AGMA 11", so you can't really do any tolerance stackup analysis anyway. One can really only make reasonable assumptions.
Michael Hill
09-04-2016, 21:13
And by not publishing tolerance information, I'm talking the GD&T, where you'd actually know a tolerance for things like perpendicularity, concentricity and runout.
And by not publishing tolerance information, I'm talking the GD&T, where you'd actually know a tolerance for things like perpendicularity, concentricity and runout.
Who cares? The gears are fantastic. The best you can buy on the market. VEX doesn't need to provide that info given that most people can't even check runout/perpendicularity (and definitely can't check concentricity). It's not going to change your design.
Michael Hill
10-04-2016, 07:53
Who cares? The gears are fantastic. The best you can buy on the market. VEX doesn't need to provide that info given that most people can't even check runout/perpendicularity (and definitely can't check concentricity). It's not going to change your design.
I agree that they're great. All I'm saying is it's a lost cause to try to do real tolerance stackup analysis.
Who cares? The gears are fantastic. The best you can buy on the market. VEX doesn't need to provide that info given that most people can't even check runout/perpendicularity (and definitely can't check concentricity). It's not going to change your design.
I've run into some issues with the tolerances on the tiny hubs on them while lightening a couple times, but apart from that I agree they are the best on the narket. Nowhere else can I find aluminum gears with hex bore for those amazing prices.
vBulletin® v3.6.4, Copyright ©2000-2017, Jelsoft Enterprises Ltd.