04-07-2016 11:20

Jay O'Donnell

Re: pic: Offseason Shooter CAD Project

Looks pretty good. I see the cylinders connected to the back plate, I'm assuming that's for pushing the ball into the shooter wheels? Also what's holding the ball in place in the vertical dimension?

04-07-2016 12:07

Sperkowsky

Re: pic: Offseason Shooter CAD Project

Yea that is what the cylinders do.

As for the ball getting held in the vertical dimension. Currently the way it works is the ball compresses slightly against 2 of the hex shafts. It works in theory but in practice I'm not sure. Luckily those spacers are just 1/2 hex shafts so I could add rollers with 2 bearings and a tube on each side which should be more then enough to keep the ball in place. Worst comes to worse I could make resting points for the ball out of tubing and lexan sheets but I don't think that's necessary.

04-07-2016 12:42

S.P.A.M.er 17

Re: pic: Offseason Shooter CAD Project

Is the pulley on the side for vertical aiming? Also, How thick are your blue plates, and what tap size did you use to secure the pulley plates? Do you see any potential problems with drilled out holes that are almost as big as the thickness? I've never done something like that, so I'm just curious.

04-07-2016 12:57

Ether

Re: pic: Offseason Shooter CAD Project

What's your wheel diameter and expected operating wheel speed?

04-07-2016 13:08

Sperkowsky

Re: pic: Offseason Shooter CAD Project

Yes, the pulleys on the sides are for tilting the shooter.

The plates are 1/4" thick. I am going to tap for 10-24 bolts which require a hole of .1495. A 10-24 screw is .190" in diameter. So, it does leave a decent amount of space left. Thinking of the failure mode I have a hard time seeing them fail but again I can not be 100% sure unless I tested it.

04-07-2016 13:25

Sperkowsky

Re: pic: Offseason Shooter CAD Project

Math is not my strong suit so, forgive me if any of these numbers are incorrect.

The wheels are Colsons with a Diameter of 4"
Being its a 775pro which has a Free Speed of 18730 RPM the free speed after reduction (Technically a 2.08:1) is around 9005 RPM.

Typing the numbers into JVN's Calculator it says in reality it will be about 8990.4 RPM.

04-07-2016 14:18

Sperkowsky

Re: pic: Offseason Shooter CAD Project

I added some 2x1 supports and modeled the shafts.

This GIF shows the motion path very well.

04-07-2016 14:58

Ether

Re: pic: Offseason Shooter CAD Project

...

04-07-2016 17:15

Cothron Theiss

Re: pic: Offseason Shooter CAD Project

Any chance you could get longer belts and mount the motors on the other side of the pivot point? They'd act as a counterweight for the whole carriage and make everything a bit more balanced.

04-07-2016 17:35

Sperkowsky

Re: pic: Offseason Shooter CAD Project

Yea, I totally could do that. Thanks for the idea.

I probably would have done that to start with but, this whole thing actually started as this.
http://i.imgur.com/EFDEC8s.jpg?1

I was just trying to learn how to model belts. And.... I got a little carried away.

Now... It has become this.
https://gyazo.com/b5cd294f84ccf905d3028937b3520f59

04-07-2016 20:02

EmileH

Re: pic: Offseason Shooter CAD Project

Any plans for an intake? Or will your shooter be the intake?

04-07-2016 20:11

Cothron Theiss

Re: pic: Offseason Shooter CAD Project

No problem. With the 775's, you can certainly get away with leaving the motors on the far end, but if you were running mini CIM's or had some type of gearbox as well, moving that to the back end becomes infinitely helpful.

Are you using the Belt/Chain Tool in SolidWorks 2016?

04-07-2016 20:22

Sperkowsky

Re: pic: Offseason Shooter CAD Project

I used the Belt/Chain Tool to actually model the belt but, I first did a sketch of everything.

I don't have a picture of the one from this particular project but, here is a picture of the sketch of the drive rail I am currently working on.
https://gyazo.com/aaef1a4585d9dbe3b34d1960f3728893

The thing to look at is the Path Length measurements. To get it I just put an arc around the pitch diameter of my pulley circle on opposing sides of both pulleys I am trying to link. I then drew lines to connect the 2 arcs and selected all 4 of the sketch marks and assigned a path length measurement which can be found under the drop down of smart dimension. From there I can just assign a belt length that I can find from a vendor and I know for sure the spacing will work.

You may know all of that but, I figured Id tell you what I did anyway.

04-07-2016 20:36

Sperkowsky

Re: pic: Offseason Shooter CAD Project

There are plans for an intake on this not likely to be real robot. I am probably going to do a small beaver tail that is pneumatically actuated. If that appears to be heavy or not worth the effort I will do something similar to what 5172 did.

The robot getting built is entirely dependent on whether We can raise somewhere around 12 - 15 thousand before September. But, if we do get the money we will definitely build this for a local off-season or two. I am aiming to make everything as cheap as possible. Doing stuff like reusing wheels and pulleys for the drive train. So, maybe even if we do not raise that much money I can get our team to do a few extra fundraisers early in the year.

I am going to get to work finishing the CAD and then raising some money .

04-07-2016 22:29

Cothron Theiss

Re: pic: Offseason Shooter CAD Project

This is probably more helpful for your drive rails than this shooter, but you can also set make the belt feature itself be a driving distance, meaning that you can leave something like sliding bearing blocks undefined in your assembly, and let the exact length of the belt feature definite the placement of your front and back axles/wheels/bearing blocks. But if you want to build your shooter model so that it's easy to change in the future, you can use this same belt feature and have the placement of your bearing holes and lightening patterns driven by the length of your belt.

Just some interesting things i found when working with the new and improved belt feature in SW 2016.

05-07-2016 08:27

JesseK

Re: pic: Offseason Shooter CAD Project

What holds the ball away from the wheels while the shooter wheels spin up?

edit - I see your comment about the hex shafts, but what about their placement gives you reasonable confidence they alone are enough?

05-07-2016 08:54

Ether

Re: pic: Offseason Shooter CAD Project

I was asking about the expected operating speed, not the free speed... or are you saying you expect to operate the shooter with 100% voltage (with no headroom for closed-loop control) ?

05-07-2016 10:05

Sperkowsky

Re: pic: Offseason Shooter CAD Project

Well, we probably will have some sort of closed loop control but I am not to sure at the moment. This year we ran our shooter motors at 80% so I am going to guess if we build this shooter we will do the same. So, I am going to guess the shooter motors would run at about 7000 RPM.

05-07-2016 10:07

Sperkowsky

Re: pic: Offseason Shooter CAD Project

In all reality it does not completely. I am not worried about the ball hitting the wheels as I can always ramp up the shooter motors in reverse for a bit longer but I do think it will hold the ball adequately enough. I am not adverse however to be a better securing system I am just trying to figure out the least obtrusive way to implement it without adding more unneeded complexity or changing the form factor dramatically.

05-07-2016 10:38

Chris is me

Re: pic: Offseason Shooter CAD Project

Good start! Designing things in the off season is the best way to get experience designing mechanisms for the competition season. The keys to getting better at manipulator design are lots and lots of practice, thorough prototyping, taking advice / ideas from other designs, and design flexibility. Keep these in mind whenever you draw something up! In reality no mechanism just goes from CAD to robot without prototyping or testing.

Here's some miscellaneous and random advice about the whole thing.

First things first, making an offseason robot shouldn't cost $12,000. If you reuse electronics and a drive base it can cost under $1,000. Unless you are trying to raise that money for other goals, and not all of it is for the robot, I would try and be extra conscious of your budget for an offseason robot.

I would tailor your design to what can be easily manufactured by your team. I'm guessing with all of the 1/4" plate, you have access to a waterjet or CNC mill that can handle all of those features? If you don't have these, you'll probably want to tweak the design to work with aluminum tubing, VersaChassis gussets, etc. Lots of great shooters were designed this way, and they can end up being quite rigid if designed right - rigidity is pretty much the most important thing for these kind of shooters.

If you are milling, be sure all of your pockets and contours are designed to be easily machined with appropriate size tooling. This means interior fillets larger than the radius of your cutting tool, etc. Try to avoid small pockets that cut relatively little weight in exchange for machining complexity, like the X shaped pockets in the back corners of the shooter. If you are waterjetting, you don't have to do any of these things, but fillets are still nice to reduce stress and to look cool and stuff.

For your shooter wheels - I would look into increasing the moment of inertia of the shooter shaft assembly. This is typically done by adding weight to the shooter shafts, as far away from the center of rotation as you can get away with. This extra inertia will dramatically help with range and repeatability in exchange for a bit more spin up time.

You are definitely going to want more positive retention of the ball than you currently have. Relying on the shooter wheels to pulse the ball backwards before spinning up to shoot is just tricky, and the more the ball can jostle loose, the less accurate your shooter will be. You also risk dropping the ball as the shooter pivots. Using cylinders to push the ball into the shooter is a good concept, but I would find a way to make a simpler piece or assembly to do that - cutting one huge plate like that is a little wasteful of resources and material.

Finally, you are not going to be able to get away with tapping 10-24 or 10-32 screws into the side of 1/4" plate. There's just not enough material for that to work. I would find another way to bolt things to the side of your material. Look at the VersaChassis gussets, simple L brackets, or capturing 2x1 tubing between your plates, etc. for an alternate solution.

05-07-2016 10:44

Ether

Re: pic: Offseason Shooter CAD Project

Does anyone have a link to a trustworthy number for the rotational moment of inertia of the CIM rotor?

It might be instructive to do some trade-off computations comparing the increase in rotor angular momentum with increasing rotor speed versus the drop-off of CIM torque with increasing rotor speed.

05-07-2016 12:16

Sperkowsky

Re: pic: Offseason Shooter CAD Project

First of all thanks. I will say getting better at CAD is something I have put off for about a year. I always saw CAD as something necessary to gain measurements and translate ideas to other people but until recently I did not realize its a great tool to come up with designs as well.

With the off-season robot I have been trying to make it as cheap as possible. I reused our wheels, belts, hubs, and motors. I also used some of the spare material we have leftover in our shop. If we were to build the robot we could probably do it for under $500. But, there are priorities like funding for a second regional, funding for both a normal robot and possibly a practice robot along with hopefully funding for our first champs trip.

We have a water jet sponsor and cnc mill sponsor. The cnc mill sponsor is new so currently I designed everything as if it was going to be water jet cut. There are 0.1 fillets but that is not big enough for milling. The cnc mill sponsor is kind of unique in that its not a normal service. They use large HAAS machines but restrict the size of parts due to the way they operate. Being we are sponsored by them and not a normal customer I am not sure whether we could get an exception. That's something we can hopefully work out. We also just bought a Shapeoko 3 XL so we possibly could cut the plates with that although we do not plan to do that as its really not optimal.

I was talking to some people about adding large AM Steel gears and I think I will proceed in doing it. It would also be cool to test the difference in shot arc, speed, and distance with and without the added weight.

I agree with all these comments about ball retention I am seriously doubting the compression is good enough. Currently I think I am going to add arced lexan strips on both sides. We have a lot of tooling to help with that sort of thing like specialized heaters and brakes so it is a nice and easy job. Plus its a good learning experience for incoming freshman. In terms of wasting material I agree. But, I do like that assembly a lot so what I did was change the material to Lexan. Its easy enough to cut on our router and we have a ton of it.

Lastly thanks for the last thing. I was not sure whether there was enough material left over. The L bracket idea works good for us currently. Quick question. We are currently trying to get a welding sponsor. Would welding these work well?

Thanks for the help.

05-07-2016 14:25

Cothron Theiss

Re: pic: Offseason Shooter CAD Project

I'll go ahead and throw in my two cents about how to attach those side plates. I would not weld those on, for a couple of reasons.
First off, what material are these plates being made out of? My first guess would be 6061-T6 aluminum, but it could be several different alloys depending on your water-jet sponsor's resources (and generosity) and your needs. Depending on which material you choose, you may not be able to weld it whatsoever. Secondly, you'd probably have to remove all of the pocketing around the area you'd want to weld just to ensure there is no weakening of the plate itself due to the weld. Also, having to send off this assembly to another sponsor would most likely increase your turn-around time quite a bit, for only a few small welds.

My suggestion would be to cut out these side plates and then bend them into a U-bracket of sorts (you mentioned you had brakes). Again, you have to consider your material considerations for the plates you're bending, but that constraint only applies to the side plates themselves. Also, if you design the side plates so they mount using the holes you're already using for the standoffs, you can get away with only changing the design of the side plates.

However, if you go down this path, you have to have pretty accurate bends. If you incorporate some guiding grooves into the design of the side plates and you have members who are pretty careful, you can get away with a hand brake, but it is another thing to consider.

05-07-2016 14:45

asid61

Re: pic: Offseason Shooter CAD Project

I 100% agree with everything you said except for this last part. Tapping 10-32 into 1/4" plate has been done by my team and others many times (particularly last year) and is definitely strong enough to take loads. 10-24 might be worse considering the lower TPI, but 10-32 is definitely ok. Considering the number of screws in his design I don't think it's necessary to complicate it by adding different fastening methods.
I do agree that a lot of this can become tubing and that will simplify a lot of things (including being able to remove the sideplates or mount them easier).

05-07-2016 14:57

Chris is me

Re: pic: Offseason Shooter CAD Project

You can totally tap 10-24 or 10-32 into 1/4" plate just fine - I was talking about tapping into the SIDE of the plate. The .190ish hole in the .25" ish material is really cutting it close, and a #8 screw at the very least would be more comfortable.

05-07-2016 15:02

asid61

Re: pic: Offseason Shooter CAD Project

Good point, 100% on that.
I thought he would first screw it on, then weld it to save some strength.

05-07-2016 20:34

Sperkowsky

I worked on a better way to mount the tilt plates and went with the l bracket method. I left out fasteners due to the fact that this will be cut by hand most likely by a brand new member. I also gave up on Lightening holes as there's not much surface area left but the good news is that means if we want it can be done with our manual mill. I am going to work on finishing this robot up in the next two ish weeks so I'll probably have a few more threads like this. The help is really too good to pass up and it's offseason so everyone is winning. Thanks

http://i.imgur.com/yLaPG0F