Attaching Things on WCD

[Chris is me]

Originally Posted by JorgeReyes

By this do you mean calculating the correct gear ratios to use?

There's a range of ratios where you're geared too fast to push continuously anyways but too slow to really get an acceleration benefit. It's around the 9-12 FPS range depending on who you ask. I would recommend experimenting with 4 vs 6 CIMS on your own in the off season to get a better grasp on how they help. 4 CIMS is perfectly adequate for a lot of robots; the jump from 4 to 6 CIMS is a lot less dramatic than the jump from 2 to 4.

If your budget is constraining your design I would probably stick with 4 unless you want a drive in the 5-9 FPS range (you'll be able to push for longer stretches of time before tripping a breaker), or you are going 13+ FPS at full weight (acceleration is noticeably better). Even in these ranges it's not mandatory to go with extra motors.

[Chris is me]

Here's one example of how we attach super structures to our base frame. In 2012 we did something similar to what DampRobot posted, and in 2013 our superstructure consisted of four 1/4" plates serving as a gearbox and arm mount. Not a great example for general purpose robot building.



The white superstructure frame was welded at the ends to 1x1 1/8" wall box tubing. The tubing had a 3/16 rivet hole pattern in it, as we try to do for all of our parts, so we drilled out 4 holes per side for #10 clearance. The tops were drilled out for tool clearance. The corresponding holes on the chassis were drilled out and 10-32 rivnuts were installed. Rivnuts are awesome as they let you bolt things together to material of any thickness without having to hold a nut anywhere. To install or remove the entire superstructure we just screw 8 bolts in or out.

There are a number of ways to attach things to the frame without welding them to another member. Flat gusset can do a lot if you pay attention to how these gussets are loaded. The VersaChassis 90 degree bent gusset is also a big help.

[JorgeReyes]

I think I might completely change my chassis design to something like this because I was also looking at the chassis from 1241 and it looks extremely light and really easy to attach anything.

[JorgeReyes]

Weren't there only 6 cims allowed? It looks like you have 8 in the picture

[z_beeblebrox]

Originally Posted by JorgeReyes

Weren't there only 6 cims allowed? It looks like you have 8 in the picture

Two of the pictured motors are miniCIMs.

[Dunngeon]

This is a picture of our chassis, it isn't a WCD, but it is simpler for us to manufacture requiring only 2x1 al tubing, steel pop rivets, and a standardized aluminum gusset.

On the sides, you can see the gusset plates and the bolts that supported the intake structure. Our biggest problem with the chassis was the lack of super structure, made mounting many things harder, so depending on the game next year we will be adding a superstructure to assist in mounting game components.
Those plates to mount the intake structure were later modified, because they became increasingly cockeyed as the bolts deformed the aluminum tubing wall. Our solution was to create a wider contact structure and use 8 pop rivets to distribute the load better.


Also note it was in fact a 6wd later in the season, just not in this picture

http://team955.com/gallery3/var/albu...G?m=1394932723

Originally Posted by JorgeReyes

By this do you mean calculating the correct gear ratios to use?

Not just gear ratios. You need to know how you'll be using your robot. If it's acceleration Have a target distance you want to travel to (ie. 13.5 ft), and optimize the acceleration to reach that distance in as little time possible. Remember, the higher the gear reduction, the more the effects of acceleration are noticeable, but that also means you'll be moving slower. You need to find the happy medium where your speed is high enough to utilize the high acceleration to get to your target distance as quickly as possible, but your speed has to be low enough to make the acceleration effects noticeable. You can play around with numbers in the JVN design calc and JVN's acceleration spreadsheet, and see what works out best. A good thing to do is set a speed for a 4 CIM drive, then change the number of motors to 6 CIMs, and see how the time to distance changes. If it's significant, it's worth using. If it's insignificant, the extra motors aren't helping. Of course results may vary, but from some quick testing I did over the past year I've found that for most usable speeds in FRC at common target distances, the average amount of time shaved off from travel time between point A and point B is no more than a quarter second at most, which I'd consider not being worth it. That being said, this was assuming a total robot weight of ~110 lbs. Lighter robots will feel the effects of acceleration a lot more.

Another thing to optimize for in 6 CIM drives is torque, especially if using a PTO. Torque is straightforward - the greater the reduction, the more torque there is. However it is still something to optimize for if you are using PTO.

Like I said - it is all dependent on how your team sees things and the data you get from testing and doing your own math. My experiences over the past year have made me come to believe that unless you are using a PTO, a 4 CIM transmission will get the job done just as well, without the danger of tripping the main breaker and without the extra 5.64 lbs of the two other CIMs. While these are my experiences, the best way to make a decision is to go out and test things out for yourself.