-1/4" plywood 8WD with center wheels dropped 3/16"
-2" wide 4" Vex Pro traction wheels with blue nitrile tread
-Center wheels are gear driven; outside wheels are connected with belts
-Designed for low profile (
I like the placement of the motors and the integrated gearbox. It looks clean.
Can I ask why this is specifically a plywood drivetrain? It looks to me like a sheet metal drivetrain. What makes it plywood?
And what design decisions were made knowing it would be made of plywood instead of some metal?
Some teams either don’t have access to metal machining resources, or due to school/other restrictions, can’t work with welding/CNC. I remember seeing a team at champs last year whose bot was made entirely of wood for that reason.
Anyway, good looking drivetrain! I like the positioning of the wheels, but how are you going to drive the front/back wheels? Is there a sprocket on the other side of the wheel that is hidden in the render?
It looks like for each corner wheel there’s a vexpro pulley towards the frame perimeter on the shafts for the wheels set further inside the frame, and a pully towards the center of the frame on the shafts for the wheels further towards the edge of the frame perimeter. It also seems like there is a gap in the plywood for the belts to run through that is hard to see in this rendering due to perspective.
Though that does bring up another thought. Is there any quick and easy way of adjusting the tension of the belts, such as a vexpro versablock and cam setup?
Also, another thing I noticed, was that the lightening hole behind the inset wheels happens to line up just right so that debris could be thrown by the wheels right up into your drive-train gearboxes and potentially your electronics.
If you design with the correct center-center distances there is no need for tensioners. The kitbot is a great example of this.
We ran belts this year (from VexPro) in this way and did not have any issues related to over/under tensioning
You have the perfect place to mount fans above the cims, while I do not think they are very useful during competition, if you were to make a practice bot that would be running for extended periods they could be very valuable. Just my 2 cents.
(It looks gorgeous by the way, great job!)
Some more specs not in the original post:
-Estimated weight of 34.8lbs (Wood frame weighs 8.8lbs)
-Geared 5.82:1 for an adjusted top speed of 12.9 ft/s
-Wheels, belts and pulleys accessible from bottom of robot
-Includes mounts for encoders, battery and main breaker
Here’s another view with the top plate removed so it’s easier to see the structure and power transmission components:
http://www.bitbuckets.org/wp-content/uploads/2014/07/8WD_v2_no_lid-800x484.jpg
Our team has a sheet metal sponsor, but they have about a week turnaround time. We also have a laser cutter in-house which cuts plywood so if we use wood, we can get our parts much more quickly.
IMHO, the main design differences between plywood and sheet metal are the material thickness, the method of attachment and the lack of bends in wood.
-The drivetrain is made out of .25" plywood; if it were aluminum, I’d use .060-.090".
-Instead of being riveted like sheet metal, the drivetrain will be assembled with finger joints (http://www.instructables.com/id/How-to-Build-your-Everything-Really-Really-Fast/step2/Magical-Finger-Joints-Joining-Plates-at-Right-Angl/) and glued.
-You can’t bend wood like sheet metal so every section at a different angle is a separate piece. The wood frame has 28 parts; with bent sheet metal, far fewer would be needed.
Otherwise, I think designing for plywood and sheet metal is quite similar, since both are very strong in the plane of the material but vulnerable to cracking or bending if force is applied from other angles.
Wooly is correct in his description of the powertrain components.
There is not a tensioner for the belts; we had good experiences with center-center distance belts on the AM14U last year and would be happy to use them again.
Good catch about the lightening hole; I’ll remove that in the next iteration.
Good idea. I don’t have CAD of a fan to add now, but if we built something like this, I would figure out mounting for a fan.
The next thing I’d like to look into is repackaging a ball shifter into a lower and thinner enclosure to add a lower gear to this drivetrain. I’m also worried about the loads on the CIM pinions in the current gearbox.
Is .25" plywood really strong enough for an FRC drivetrain? Nobody in this thread or the other one mentioned strength issues, so I assume that it’s good, but my experience seems to contradict this.
I’ve attached a picture of our .75" plywood bumpers from when we rebuilt them after MAR champs. As you can see, there’s a large crack down through them. Why do you think our bumpers broke and your frame will hold up?
It’s not about the strength of the material as how it’s used. It looks to me like there’s a lot of your bumper which lies above any backing support, which means that the plywood only has its own thickness for strength. In comparison, look how every panel of his frame is supported at the top and bottom, with many ribs which support it continuously. Plywood (and most sheet materials) are much, much stronger when loaded in ways that compress or stretch pieces rather than bend them.
I for one would love to see a sheet-metal version of this design too. It’d be interesting to compare the two. I really like the layout of the whole thing.
I’d love to try this but am completely unfamiliar with the SolidWorks sheet metal tools. Any resources you would recommend for learning about sheet metal design?
971 did a similar design in sheet metal this year, with 6WD
https://picasaweb.google.com/117769834305511597729/2014CADScreenshots
https://picasaweb.google.com/117769834305511597729/2014FRCTeam971SpartansFabAndAssembly
The tutorials aren’t bad. We start by having our students do the sheet tutorial. From there, we hand them a physical part from a previous year, and have them draw it up.
Sheet design is a different animal. I enjoy it because it is very challenging and rewarding. The best recommendation that I can give is to start with something simple, and work your way up from there. Maybe try drawing a square sheet frame first, before getting into all the complicated angles that are part of an octagonal frame. I recommend that people think of sheet metal like paper. All the strength is in the bends.
Post your design when you have something drawn up, and get feedback on how to improve it. You can run FEA on it if you are adventurous.
One week turnaround is actually very good. We built our entire robot last year with 2 to 3 week turn on all of our sheet parts. You are lucky to have a sponsor that treats you so well, take care of them.
I would argue that particularly for a drive train, the 1wk turn on sheet parts is more than sufficient. Our approach with the drive train is to send out the parts for fab week 1 while we are continuing to design the super structure. By the time we have the superstructure designed and sent out for fab, the drive train parts are coming back. Having sponsors help us with fabrication lets us focus on design and speeds up the process.
As you should be, the pinion loading in this design is the biggest weak point.
If you want to continue with this kind of layout I recommend doing some calculation on the pinion loading. There are different ways of calculating the load on the pinion, the most common is the Lewis formula. The Lewis formula models the gear tooth as a cantilevered beam with the load applied to the tip. To use this formula you need to know the tangential load on the tooth, the diametric pitch, the face width, and the Lewis form factor. The Lewis form factor is based on the number of teeth on the gear, the pressure angle and depth of the tooth. A table of Lewis form factors can easily be found with a google search. If you want to preform this calculation on paper use the fowling formula:
bending stress= (WtPd)/(FY)
where Wt is the tangential load,
Pd is the diametric pitch
F is the facewidth
Y is the Lewis form factor.
If you don’t like paper and pencils, lucky for you there are many online calculators, this is one of many: http://www.engineersedge.com/calculators/gear-tooth-strength-calculator.htm
I should note that many of the online calculators, including the one I linked to, solve for the max load a tooth can take while remaining under a specified amount of stress. The formula used for this calculation is the following:
Wt=(SYF)/Dp
S is the maximum bending stress the tooth will undergo. It’s commonly recommended to make S 1/3 of the materials ultimate strength.
These equations are intended for gears operating in idea conditions, meaning they are properly lubricated and not experiencing shock. If your gears are running in un-ideal conditions (this is the case for most FRC gearsets) S should be lowered.
These formulas will help give you a pretty good idea whether your gear will fail , but Lewis formulas are only the tip of the iceberg when it comes to gear strength calcs. The Lewis formula does not account for rpm, other geometric properties of the gear, and surface wear.
If you want to know more equations I recommend you look up the following:
Barth velocity factor: This is a modified version of the Lewis formula that accounts for the gears velocity.
AGMA bending stress: This is pretty much a much more advanced version of the Lewis formula with more factors. This formula is probably more accurate than needed for your application, but if you find this topic interesting you might enjoy looking it up.
Hertzian contact pressure: This calculation models two cylinders in edge contact and will tell you the max normal pressure at the line of contact. This calculation is very important as it will tell you whether or not the surface of the tooth will wear. If you use the Lewis formula and it indicates failure you don’t need to calculate contact pressure. But if the Lewis formula indicates that the max bending stress is in range you might want to calculate the contact pressure to determine if the teeth will significantly wear.
-Adrian
Edit: I should mention that while the internet is a decent resource for finding and understanding these formulas a machine design book will give you much more information. It wouldn’t be a bad idea to talk to one of your ME mentors on the subject, if they can’t help you with the formulas they probably have a textbook that can.
One thing to keep in mind is that the wheels don’t have positive traction. So to calculate the force in the pinion gears you need to figure out the loading the wheels will apply before slipping. Adding up the stall torque of the motors and factoring in the gear reduction to the force applied on the pinion tooth is the front half of it. Traction and slippage is the back half. The pinions do take more loading than the usual frc gearbox.
I like the cim layout and have been playing with it around a year. I haven’t been satisfied with the packaging to move forward with testing. Its neat to see how others approach similar problems with their designs.
2337 the enginerds have done a similar 8wd sheet metal layout with an integrated two cims and two speed gear box the past two years. I really like what they did. They have their 2013 cad online, you should check it out for design solutions implemented if your interested in this 8wd layout.
My concern would be the impact resistance of the 1/4 inch plywood. Metal bends before cracking better than wood in impact situations.
I can’t see exactly what you’re doing with your pinions, but 254’s been running double loaded pinions in all their 3 CIM gearboxes.
Theoretically you can triple load a pinion gear as the drivetrain rocks on one set of wheels or the other as a worse case scenario. If both sides of the gearbox are loaded evenly, its more of a double loaded worse case scenario.
The worst case load scenario would be during a high speed direction change in high gear. The pinion could see upwards of 6x stall torque.
-Adrian
Wow, what responses for a summer post.
I was only thinking what wonderful designs two heads can come up with…
