22-09-2013 13:20

Akash Rastogi

Re: pic: Spectrum Fall 2013 Drive V2

Digging the new iteration. Reminds me of 971's work + the VEX chassis.

Nice job! Can't wait to see the finished product.

22-09-2013 16:40

Seth Mallory

Re: pic: Spectrum Fall 2013 Drive V2

Looks great. I hope to see the final robot sometime.

22-09-2013 17:11

BBray_T1296

Re: pic: Spectrum Fall 2013 Drive V2

I'm sure I'll see it. We see you guys everywhere we go

22-09-2013 17:15

DampRobot

Re: pic: Spectrum Fall 2013 Drive V2

Ooh, very nice looking! I'm still not totally sold on gears, but that looks like a really nice system...

Can you go into more detail about the "extra features" you've integrated into your frame? I have't seen many frames with a integrated Ethernet jack or pneumatic system, for example.

22-09-2013 21:59

AllenGregoryIV

Re: pic: Spectrum Fall 2013 Drive V2

I sure can, this an accumulation of a lot of random features I've seen on other robots when inspecting them or have been posted on CD. The panel mount Voltmeter and ethernet port should make practice a bit easier. The pneumatics stuff is just to give a convenient place to access it. All of this will be below the tops of the bumpers so it should be pretty well protected. Here are the links to all the items.

2 wire Volt Meter
Ethernet Keystone Coupler
Panel Mount for 1/8 NPT Pneumatics
Small Pressure Gauges
Miniature Pneumatic Valve

I'm also looking at bumper mounting systems. The latch systems used by 254, 971, and other seem really expensive at around $10 each. Has anyone had experience with McMaster quarter turn fasteners?

Also here are the CAD files for the sheet metal in both Solidworks 2013 and STEP. Dropbox Link

22-09-2013 23:11

AdamHeard

Re: pic: Spectrum Fall 2013 Drive V2

I'll warn you, we've purchased their "miniature dump valve" and it really isn't what we'd call miniature.

22-09-2013 23:18

AllenGregoryIV

Re: pic: Spectrum Fall 2013 Drive V2

Thanks for the heads up, Adam. Have you found a smaller one? It's listed as 1.6875" long.

22-09-2013 23:42

AdamHeard

Re: pic: Spectrum Fall 2013 Drive V2

Not yet. Not sure I expect to, it was just thought their use of miniature to describe something that's more or less normal size (to FRC) didn't work for me.

However, we did find lighter ones!
4379K41
and
4379K61

23-09-2013 01:48

DampRobot

Re: pic: Spectrum Fall 2013 Drive V2

We used these for our offseason build, and they've worked wonderfully. I highly recommend them.

23-09-2013 02:14

AllenGregoryIV

Re: pic: Spectrum Fall 2013 Drive V2

How many did you use for a full bumper set, 8?

23-09-2013 02:22

DampRobot

Re: pic: Spectrum Fall 2013 Drive V2

For one color 2, so for a full set of two colors 4. We have two C shaped bumpers.

23-09-2013 02:33

AllenGregoryIV

Re: pic: Spectrum Fall 2013 Drive V2

Each C shaped bumper only had 1 clip or 2? Did you test that against other robots, that seems like it doesn't have a lot of holding force. Was anything else holding them in place? Do you have any pictures?

23-09-2013 03:12

DampRobot

Re: pic: Spectrum Fall 2013 Drive V2

Each only had 1 clip. The bumpers were supported in the middle by two L brackets mounted to the frame of a WCD style frame. The bumpers sit with one leg of each L against the middle of the C. One of these L brackets had the clip. There were also two L brackets on the bumper that supported the top and bottom of the c against the frame, and flanges on the bumper that locate it side to side by sliding over the L brackets mounted on the frame. When I get a chance, I'll try to upload a photo.

We've already played an offseason with the bumpers, including some elimination matches where we were pretty heavily defended. The bumpers were solid, the only damage to them was that one of the 0s on one of our bumpers peeled a little.

The important thing to remember is that the clips shouldn't really take all of the force that's on the bumper. Other features on the bumper can do this. All the clips do is retain the bumper on the frame in a way that can easily be removed.

23-09-2013 03:36

AllenGregoryIV

Re: pic: Spectrum Fall 2013 Drive V2

Interesting, we were planning to do a one piece bumper that drops over the top of the robot. In that configuration I don't see an easy way to limit to the vertical forces that will act on the latch during impact. We had issues with the double C design because they are weaker where the two C bumpers meet. Our frame bent in when we hit the pyramid once.

23-09-2013 03:46

DampRobot

Re: pic: Spectrum Fall 2013 Drive V2

Put a small but long L bracket on each side of the bumper that keeps the bumper from getting pushed down. The latch should be more than enough to keep the bumper from going up. (I'd be surprised to see much up forces on a low-to-the-ground bumper in a match.) I believe 971 did their bumpers this way in 2012.

23-09-2013 04:52

AllenGregoryIV

Re: pic: Spectrum Fall 2013 Drive V2

971 also used them in 2013 but they have 8 around the perimeter of their robot.

After doing more research I am looking into using twist latches that are used on road/flight cases. They're normally rated for a lot of force and you can get them for under $3 each.

I also think it might be possible to build the keeper directly into the frame, instead of having to rivet/bolt it on.

23-09-2013 16:10

kevincrispie

Re: pic: Spectrum Fall 2013 Drive V2

Allen,

I'm not sure what your thoughts are at this point, but I can help to shed some light into how 971 did bumpers this past year. Below is a picture of our 2013 bumpers.



We actually did not have 8 latches all the way around, as you can see in the picture. Some of the latches intersected with some of the robot components, so we wouldn't have been able to access them. Also, the bumpers and latches don't always go together smoothly, and pieces can warp over time, so by the time we actually competed, only around 3-4 of the latches were actually hooked on. The rest were tied away so they wouldn't jostle around during the match.

We bought the latches from McMaster, but we ended up having the hooks made by a sheetmetal sponsor out of stainless steel, 0.090" thick iirc. We then riveted these onto the frame. We had issues with the hooks bending in 2012. The stainless steel hooks this year did not have the same problem.

This system held up reasonably well throughout the course of the season. Obviously we would have liked to attach the bumpers at more points, but it doesn't always end up that way. We ran into similar issues in 2012.

I hope this helps. If you have any questions don't hesitate to ask me or another 971 member.

Kevin

23-09-2013 16:12

Travis Covington

Re: pic: Spectrum Fall 2013 Drive V2

There are a bunch of options. Cheaper ones of the same thing are here: http://www.mcmaster.com/#1889a37/=omwtea

23-09-2013 16:24

Blackphantom91

Re: pic: Spectrum Fall 2013 Drive V2

This cad model is great! I personally really like the Ethernet jack port in the side for easy quick disconnect. What type of design models did you base this off of it? seems very robust and nice. Keep up the good work!

23-09-2013 20:38

Pault

Re: pic: Spectrum Fall 2013 Drive V2

Have you ever tried that method before. Because my team tried that and 2012 and it was extremely difficult to slide it over the superstructure on the robot (of course you don't have any superstructure, but your robot next year probably will). Instead, we ended up having it come up from beneath the robot. (Yes, you have to lift the robot to do this. It takes us a long time to change bumpers). However, it is very difficult to do this with 4in wheels because of the minimum bumper height. The benefit to it is that a single piece bumper adds a ton of rigidity to the chassis (this year we were literally able to get away with a plywood board as a chassis and use the bumpers to reinforce it).

23-09-2013 21:21

AllenGregoryIV

Re: pic: Spectrum Fall 2013 Drive V2

Thanks everyone for the tips about the bumpers. We're going to experiment with the twist latches and I'll let everyone know how it goes.

What do you mean by design models? Are you referring to the ethernet port specifically or robots that helped inspire this design? Also, thanks for the compliment.

We haven't done it ourselves yet, but we have seen a ton of other robots that have. If everything goes to plan this robot will have a super structure. It's going to be the base for the prototype robot for our mock game we developed last spring, PIPEFALL.

23-09-2013 23:05

BBray_T1296

Re: pic: Spectrum Fall 2013 Drive V2

Our team uses flip-switch bumpers. While we usually (and unfortunately) seem to always make our bumpers a right pain in the neck to attach and remove, We only have to do so once during inspection. Basically, our bumpers are two sided, and we can literally change colors in <5 seconds with a simple flip. If I knew how o post a picture of this, I would

24-09-2013 04:38

AllenGregoryIV

Re: pic: Spectrum Fall 2013 Drive V2

I have experience with reversible bumpers and I wasn't a big fan. I was with the Discobots in 2011 when we did flip bumpers. They were great for week 1 but by the middle of our second event they started fraying, the velcro started breaking away from its stitching, and they just looked awful. Maybe the parent we had do the stitching didn't do it right or something but I've been hesitant ever sense. I'm also not really a fan of the seam running down the middle of the bumpers. If our design ever required us to leave our bumpers on, it's something I would consider. I also liked 33's skirt system this year but that would require a pretty good seamstress as well.

For completeness to this thread anyone looking to do reversible bumpers. This is the whitepaper by 2809 that is most often linked as instructions.

24-09-2013 09:59

Blackphantom91

Re: pic: Spectrum Fall 2013 Drive V2

well main part is the ethernet port as an IT brain I think it is ingenious. Also yes what had inspired to you make a drive train like this. I saw v1 also and was curious.

24-09-2013 10:29

AllenGregoryIV

Re: pic: Spectrum Fall 2013 Drive V2

Directly it was all from Madison and the people that posted in these threads.
FRC488 -- Fall Project
FRC488 - Fall Project, Rev. A

I've liked the idea of a gear drive, ever since looking at 25's robots from the behind the design books from 06 and 07. The bellypan setup is inspired by 968, 1538, and the other west coast teams that have been doing it for years. 971 and 148/JVN/VEX both contributed to the sheet metal inspiration.

We also talked as a team about the things we liked and didn't like in our chassis from last year.

24-09-2013 10:35

Blackphantom91

Re: pic: Spectrum Fall 2013 Drive V2

Nice! I really like it keep up the good work.

25-09-2013 19:37

AllenGregoryIV

Re: pic: Spectrum Fall 2013 Drive V2

We worked on the twist latch system today at our meeting. The basic prototype worked way better than I ever expected. You can see the pictures below.

Twist Latch Bumper Prototype

We made a very simple mock board to act as the bumper and didn't modify the VEXpro chassis in any way (The 2013 KOP frame would need slots cut into it). I'm pretty sure this is a setup that most teams could use.

We will probably modify our 2013 robot to use this system for our last off season in November. That we can test it against other robots and make sure it can stand up to abuse.

These are the latches I purchased for $2.45 each.

26-09-2013 23:24

AustinSchuh

Re: pic: Spectrum Fall 2013 Drive V2

Not a bad start.

I keep seeing people here make the same mistake, and it bothers me enough that I'm going to say something.

Think of sheet metal like paper. Go ahead and grab a sheet and play with it. Bends add lots of strength. Wherever you break the bend, you make the part significantly weaker.

The first thing that I see when I see the side rail parts is that you broke the bend both around the transmission and on the outer edge all over the place. This significantly weakens the frame. I'm going to make a wild guess and say over 2x weaker when in compression. (Props for anyone who does FEA to get a better estimate.) Do yourself a favor and figure out how to not brake the bend. Space the wheels in away from the edge, or whatever it takes to make that happen. That will take your frame to the next level and make it bullet proof and incredibly stiff, which is exactly what you want.

If you don't believe me, print your robot out on paper at partial scale, cut it out, and assemble it. Try twisting and bending it, and see what happens.

26-09-2013 23:59

T^2

Re: pic: Spectrum Fall 2013 Drive V2

Weird, people keep telling me to think of paper as sheet metal.

26-09-2013 23:59

AllenGregoryIV

Re: pic: Spectrum Fall 2013 Drive V2

Thank you for the feedback, Austin. I agree with you and this is the type of feedback I hope to get by posting these, I'm still learning. However, I don't think it would be able to be assembled if the belly pan had a full flange along the outer edge of the wheel wells.

27-09-2013 22:06

AustinSchuh

Re: pic: Spectrum Fall 2013 Drive V2

Hi Allen,

If I'm understanding you correctly, I think it is still possible to assemble with a full flange along the belly pan. Take the clamshell, angle it at ~45 degrees, and then slide the inner edge into the frame rail. Then, rotate it to the correct angle and slide it in until it is in place.

I would recommend moving the flange from the bellypan to the clamshell. That is what we do and works marvelously. It makes it so that the brake doesn't need to be the correct length for the front and back of the bot.

Also, consider adding some ribs inside your clamshell. That will add quite a bit of strength. Take a look at how we do our frame. There should be pictures available somewhere online. I expect to only make small modifications to it next year, assuming that the game cooperates.

What machines do you have available to cut out those parts? That will require a lot of laser time if the bellypan is being laser cut, and is a pain to punch if it is going to be punched due to the triangles. Some small changes should make it a lot easier on your sponsors.

Consider integrating your battery box into the frame. There should be some pictures of how 971 does it somewhere, or you can read my suggestion in Madison's thread.

Austin

28-09-2013 00:48

AllenGregoryIV

Re: pic: Spectrum Fall 2013 Drive V2

Since it's all dead axles does the clamshell really need ribs? Don't the bolts and spacer act as ribs? Also how would you go about getting wheels in and out of the robot in your configuration? The current system has them drop out the bottom very easily. Also when I said it couldn't be assembled I was talking more about getting the wheel and gear assemblies into place not just attaching the sheet metal.

We do it all on a laser cutter, but it's run after work hours so they aren't that worried about laser time. They don't have a turret punch so it doesn't really help to make them circles, the laser is doing all the work either way.

The battery in this robot will end up lying flat, I'm also not a big fan of the battery on either bumper rail. If our design ever needs that much weight shifted to one side of the robot we would probably do it but it has never fit our design.

28-09-2013 02:15

Travis Schuh

Re: pic: Spectrum Fall 2013 Drive V2

The dead axles will help maintain the width of the siderail, but I would not consider them a replacement for the stiffness ribs add. Ribs provide a shear wall which helps resist lateral load (parallelograming) and torsional loads, something that an axle alone does not provide much support to. A larger diameter tube coupled with a compression bolt will start to adequately resist lateral loads, but I would not consider a wheel with a 0.5 axle to be a sufficient replacement for a rib.

971 is able to drop our wheels straight out of the bottom because our tensioner blocks space the wheels out past the flange that bends inward. This could be achieved in your design by adding a spacer to the shaft. While a straight up round spacer could get annoying to keep track of and get aligned when putting the wheels in, I could imagine making a simple spacer block that gets affixed to the frame sheet. This could be done quickly on a mill, or even by hand with proper match drilling techniques.

I still think it is important to be cognisant of laser time. It is still costing your sponsor proportional to how long it takes to cut your parts. If they are cutting lights out, then your parts are being cut instead of a customer's parts. If the machines are being attend to while cutting, then someone has to stay extra. Being respectful of their time and money helps maintain a good relationship. If you are interested in saving laser time, I would work on minimizing the number of unique holes. We have learned from our sponsors that the time it takes to pierce the material is significant compared to the time it takes to traverse a profile. Another factor is that it takes longer for thicker material. This has pushed us to going with thinner metal and fewer lightening holes when appropriate, a combination that works well together.

28-09-2013 03:58

BBray_T1296

Re: pic: Spectrum Fall 2013 Drive V2

To put ribs into a bend, the only way I've seen it done is with special brake tools used when you brake press the sheet metal into 90 (or 88) degree bends



These add massively increased rigidity to the structure.

28-09-2013 18:47

T^2

Re: pic: Spectrum Fall 2013 Drive V2

Why not just make some ribs separately, and then rivet them on?

28-09-2013 19:07

BBray_T1296

Re: pic: Spectrum Fall 2013 Drive V2

You certainly could, and achieve the same effect. At least with our metal sponsor, the difference between bending a part with ribs and bending one without ribs is just bolting tool "17b" to the brake press instead of "17a". Adding ribs this way is marginally lighter than separate pieces and takes up less space. For a 1/16" radius bend the ribs extend probably 3/16" up the surfaces of the metal whereas separate ribs would have to have mounting holes somewhere. Also these holes have to be cut and it will take some time to manufacture and install more pieces. Again, these differences are very subtle, so do what you'd like.

09-10-2013 09:54

nuttle

Re: pic: Spectrum Fall 2013 Drive V2

Thanks for sharing these parts -- some really nice finds, all collected in one place. You might look at Clippard GTV-2-P12 or similar for the miniature air vent... Disclaimer: I have not used one of these, nor carefully checked legality/safety, even under prior years' rules.

09-10-2013 10:12

TD78

Re: pic: Spectrum Fall 2013 Drive V2

Allen, been following along the past few weeks with your design...any updates?

09-10-2013 15:37

AllenGregoryIV

Re: pic: Spectrum Fall 2013 Drive V2

Not much, we have been doing a lot of different things (VEX competition season is starting), and working on this has been left behind a bit. We're working on retrofitting our 2013 robot with the new bumper attachment system for our last off season event and that is going very well. We should have photos and videos soon of the new system. I'm slowly adding the slots to the drive train rails and we should be almost ready to send them off to be produced.

17-10-2013 13:30

roystur44

Re: pic: Spectrum Fall 2013 Drive V2

Instead of 1 90 deg bend how about 2 45 deg bends.

You can also design in a radius bump bend in which a series of small tangent bends together form a large radius bend.

You guys ever wonder how Team 67 can get away with using the light gage round aluminum tube and why it is so strong?

25-10-2013 10:22

Ryan Dognaux

Re: pic: Spectrum Fall 2013 Drive V2

Our team is looking at making a gear drive as well and I had a few questions having never built one.

Dustin stated in an earlier post that you can use the 1/2 inch tubing that Vex Pro sells and run 1/2 inch hex broached gears on it with no issues. Can anyone else verify this? It'd be nice if we didn't have to shave a few thousandths off the tubing to have the gears spinning with little to no resistance due to friction.

Using this setup would mean running 1/4 inch bolts through the tubing. Does anyone have recommendations on the type of fastener to use & what hardness grade? I've never ran anything smaller than a 1/2 inch bolt on a drive shaft so I'm curious what others have used.

Thanks for posting these designs, I know I've learned a lot from reading through these threads.