The Frame thread
 

So, every year, my team makes the same exact frame, just modified for the years game. It weighs too much, and doesn't do a particularly good job of anything. Since I will have a leadership position on the team next year, I've made it one of my major goals to change this. One of my friends came up with this quote:

"The frame should be build to hold the robot; the robot should not be built to put in the frame"

This leaves me with a dilemma: We need the frame done early (that has been one of our biggest shortcomings in years past, not getting the frame welded until the end of week 2). On the other hand, I agree with my friend, which means we can't begin designing the frame until the design is done at the end of week 1.

According to Inventor, the frame weighs a whooping 35 lbs, not counting the supershifters, or the treads. Add those in, and your at 50 or so.

The problem is, we have used the same frame/style of frame for so many years now, our team has pretty much forgotten how to do anything different. So, I thought to myself, this is FIRST, teams like to share information, right? And people like to show off their designs, right? People like to critique designs, right?

I would be very grateful if other teams could show me pictures, or CAD of how they do their frames/drive trains. In return, I will show you and tell you a bit about ours!

We build a lower frame out of 1x1 aluminium T-slot. Very strong, very durable. Very heavy. We sandwich treads between two pieces of those T-slots. The treads are made from fan belts turned inside out and driven with a sproket from the supershifters (last year we used toughboxes). We build an upper frame out of 1/16 aluminum to support various mechanisms. The whole assembly is welded together, and we use T-slot sliders to put many things into the system.


Thanks in advance for other teams' help!

Re: The Frame thread
 

When I look at a FIRST challenge with whatever team I am involved with, I like to have my options open, which is why I am a big fan of a super simple drive type that can be built off of. The T-slots and the extra attachment points seem to make the design you built very modular though, which I like.

Let me get to the point, T-slot is heavy. Switching to a gusseted or welded tube frame will get you to a happy place. If you have sheetmetal options, you can go with that. I also became a fan of the kit frame working with 2702, we built a very solid, and reasonably light drive with it alone (much lighter than what you have quoted). I am sure you can find plenty of pics of the structural tubing design, whether it is a WCD Derivative or something else, they all seem to work pretty well.

I shall provide my take on the tubing frame as well, I have posted this before, but I challenged myself to make a warp resistant (keep welds away from the drive components) gusseted frame that was relatively light. Heavy 254 influence below. This has never been built, but it sums up a lot of peoples ideas.



Probably the big thing I tried to improve with this frame was robot torsional stiffness. By that I mean the frames ability to keep the two drive component rails parallel. The stringers between the drive component rails really help if you don't put a top box on the frame like you have on your design. That is the advantage you have at the expense of weight, nice torsional stiffness.

Look into Team 221 LLC's bearing block tensioning system as well as 254's. They are both good ways of going about things if you choose to go live axel.

Re: The Frame thread
 

Week 2 is reasonable in terms of timeline for welding. Seriously, that's not an issue.

50 lbs of frame might be a bit much, even including drivetrain.

330's traditional frame is welded 1/8" wall aluminum tubing, typically something like a 1"x1.5" or 1"x2" (though we have been known to use 1"x1" with drivetrain stuff mounted on blocks). Sorry, I don't know how much it weighs offhand, but at a guess around 10-20# depending on the year. Quite a lot of teams do something similar; others use the KOP frame for its adaptability and strength.

You could also try modding the West Coast Drive for tank treads. THAT would be interesting--and not all that difficult, I don't think.

Re: The Frame thread
 

The one thing that pops out to me is all the 80-20 in there. Dropping that 80-20 to 1/8" wall aluminum alone would almost definitely shave quite a bit off your weight. We've always used 1" C with corner brackets for our chassis.

Re: The Frame thread
 

Here are three diffrent chassis i've made over the year ... 1st one is our new chassis that we will most likely make. 2nd is the type of chassis that we use to use, and probably wont again for a while. and the last one is the same type that 1114 used this year (minus demesions and some fixing up on it).

Re: The Frame thread
 

Having a moving robot for the first time in week 3 is very good. We haven't had one moving in the last two years until sometime at the beggining of week 4. Your current timeline seems fine to me. Except: Your frame may a bit too complex if it takes a week to cut and weld parts. Spending a few days to achieve a simpler solution can save you days in the future.

For this next part, don't think I am saying any of this is wrong, but they are questions you need to answer:
1. Why shift?
2. Why treads?
3. Why T-slot?
4. Why multiple heights of tubing structure?
5. Live vs Dead axle

These are all questions you have to answer for your team. There is no universal best option. The best way to design and build a drivetrain is the way that works for your team.

Re: The Frame thread
 

8020 is incredibly heavy. Even 1/8" thick 1x1 aluminum tubing. For over 5 years, our team has used 1/16" thick, and we've only broken one segment (unguarded, whole section was pushed inward about 3 inches).

If you use 1/16" you can use these Brunner Connectors. They are light and excellent. Just attach a small aluminum gusset with sheetmetal screws to the tubes and you're done.

Re: The Frame thread
 

I'm curious what exact belts and pulleys you're using. Looks like it would be for a large supercharger.

Re: The Frame thread
 

Also, if you want, I can send the cad files for a very simple drivetrain and chassis that we put together.

Here is a photo:

Re: The Frame thread
 

The simplest, most reliable, easiest to fix and cheapest to build drivetrain we ever built was also our best performer.

It took seven years to get to this point... we used composites, the kitbot (which was pretty good actually), TIG welded aluminum tubing (which was better)...

And my favorite was this one, seen here airborne and with its twin.

Jason

Re: The Frame thread
 

I've always wanted to do this. It goes really well with our Frog Green project, but we just haven't had the time or resources to build something like this during the offseason.

Re: The Frame thread
 

http://www.simbotics.org/media/videos/presentations

If nothing else, you can build a super easy drivetrain like the one above, that is also "better than 90-95% of teams". Certainly better than the tank drive and very simple.

Re: The Frame thread
 

why is one of the transmissions rotated 90 degrees, would it be in the way of something or is it just something you haven't noticed yet ?

Re: The Frame thread
 

Kit Chassis and transmissions.

We usually play around with wheel configuration and drive ratios, but our record with the kit chassis speaks for itself.

2 Quality Awards
5 Finalist Awards
4 Championships
1 Trip to Einstein

Re: The Frame thread
 

If it is the same type of tranny as the AM supershifters then by rotating one 90 degrees puts the output shafts at the same level.

Re: The Frame thread
 

the things you should think about while building/designing a frame is

- torque (push through robots playing defence)
- speed (get around the field quickly
- Light frame (don't waist all your wieght on the frame)
- Strong frame (haveing to fix somthing on it or bending it back in place, can cost you the regional if you can't do it fast enough)
- Easy to repair ( ^ )
- easy access points to put frame together ( including wireing, attaching parts, fixing. You don't want to spend a lot of your days during build season building the chassis
- Easy to make ( ^ )
- simple to design the rest of the robot mechanisms on top of it (useing only a few new holes and bolts to attach the mechanism ontop of the chassis, or inside the chassis)

These are things that i have learned over the years that must be thought of while designing a chassis.

Re: The Frame thread
 

Not if they are using Gen2 Andy mark transmissions, the output shaft is not in the middle, but lowered. making the output shaft moved to the side and up higher. unless he is using a different transmission though.

Re: The Frame thread
 

He's using the AM Supershifter, which doesn't have a centered output shaft.

This is an amazing video and should be required viewing for all rookie teams.

Re: The Frame thread
 

I see now, i've never really seen these or even looked at them to use on a chassis. Does anyone know if they are better than Gen 2's?? :confused:

Re: The Frame thread
 

The major flaw I saw at champ's with design that Alex posted is with the side rails -- the ones that are supposed to back the bumpers. 973, 254, 968, and 1868 all used the same exact drive train with minor differences. Their bumper-backing side rails were made out of 1/2" box tubing.

On 3 out of the 4, that 1/2" box tubing was warped or bowed inward by 1/8" to 1/4" -- somewhat negating its effectiveness and being technically illegal to the most pedantic of bumper rule interpretations. To be honest, 99% of 2-level welded frames I saw are in violation, even ours. It's simply too difficult to get that thing 100% straight if it's not perfectly flat on a welding table

1" box tubing could possibly make the 2x1" vertical frame members torsion/warp rather than simply the side rail, which would then stress the side rails that hold the wheels. I'd consult with a welder before giving that concern any weight though. Perhaps Cory will chime in if he sees this.

It's easy to prevent warping during welding on a 2D frame (just the 2x1 rails). Yet things can get very tricky when moving to a level that isn't clamped directly to the welding table. It's one of the many reasons why WCD has been a very successful drive train for so many years -- its function:simplicity "ratio" is unsurpassed, thus it mitigates many of the issues that more complex frames have. Compliance with bumper rules in recent years has been the root cause of the extra necessary complexity of side rails.

Re: The Frame thread
 

Don't get me wrong, we didn't move until week 6. Akoscielski3, none of your images aren't there.

We do have sheet metal capabilities, so I would be interested if any of the team IFI teams could show us how they do their frames, or just any sheet-metal teams.

I wonder if anyone has done carbon fiber, or other "space age" material frames...

Re: The Frame thread
 

We have two sheet metal machine shop sponsors, and have used sheet metal for our base chassis/drivetrain in 2007, 2009, 2010, and 2011. 2007 and 2011 were 6WD, 2009 was 4WD, and 2010 was independent swerve. Here's some photos of our 2011 drive train:







This drive train was an absolute tank. Everything shown in the top CAD model (chassis, gearboxes, wheels, chain, etc) came out to 37.4 lbs. Each of our 2-speed gearboxes (custom ones made from a la carte AndyMark gears) weighed less than a stock ToughBox, and were geared for speeds of 6 ft/sec and 16 ft/sec at nominal load. The center shaft is direct-driven from the gearbox output, with the end two wheels being powered by 25p roller chain. All shafts are 1/2" 12L14 steel hex, with all shafts being live axle (it's the simplest way to power Colson wheels). Except for the custom sheet metal and COTS AndyMark parts, all other fabricated parts could be made on a manual lathe.

Even without a belly pan, this chassis was extremely rigid and had no noticeable flex. Once assembled during the build season and run through its initial break in period, we haven't touched it. Nothing broke, nothing needed adjusting, nothing needed replacement.

Edit: If you're interested, I've posted the CAD model of our entire 2011 robot to CD-Media.

PROTIP: If you can't get powder-coating, use enamel spray paint. If you properly clean all parts ahead of time (e.g. use acetone) and let the paint cure for at least 24 hours, it's nearly as durable as powder-coating and much cheaper.

Re: The Frame thread
 

I can't figure out how to get them up ... can you help me :)

thx

Re: The Frame thread
 

You are linking to photos on your own hard drive, only you can see those on your own computer. For everyone to see them you need to upload them to an image hosting website (like imgur).

Re: The Frame thread
 

We actually didn't have the same style bumper support, we just had some "cheater plates" (as we called them) that came out and touched the bumper (1x~3.5"x.125" plates riveted on) every 8" or so. The actual load of the bumper was really only supported by the two bumper mounts that were 1" diameter aluminum round sitting flush against our siderails.

Re: The Frame thread
 

K thx i think i fixed them now! Thanks ... now i know lol :)

Re: The Frame thread
 

Take a look at our galleries of previous builds. We use sheet metal and rivets for almost all the parts on the robots we build.

laser cut .090 5052 alum
student designed using SolidWorks



https://picasaweb.google.com/971.FIRST/2011Build#

Re: The Frame thread
 

I don't have access to a CAD file right now, but i have a picture of the drivetrain:

I apologize for the massive size, but i couldnt figure out how to make it smaller (thread for original photo and topic: http://www.chiefdelphi.com/media/photos/36226)
This year we made a laser cut Sheet Metal Drivetrain done by Rowe Trucking. Its thickness is 1/8'', and is extremely strong. However, we could cut the thickness down to help weight wise.
All in all, i liked the drivetrain, but i really like simbotic's "kitbot on steroids".
P.S. A box frame isnt half bad either.
Hope this helps,
Good Luck,
-Duke

Re: The Frame thread
 

here is another link to our 2011 build season. Has some great pictures of our frame.

https://picasaweb.google.com/1045493...on?feat=email#

http://tinyurl.com/3vsbadq

Re: The Frame thread
 

Though not a frame, we were fortunate enough to have a sponsor donate carbon fiber drive shafts for our support structure and arm.

http://www.flickr.com/photos/robosta...7626347607815/

Re: The Frame thread
 

Team 2502 uses mostly 80-20 to construct our robot, while the sentiment that T-slot is heavy is correct that doesn't make it a bad building material necessarily. Our robot this year ended up at about 110 pounds with an 8 foot lift



Also included on the robot were a pair of RS775's and about 15 feet of chain. Tslot can be doable at light weights, you just have to keep in mind it is very strong. I (150 lbs) have stood on a 4 or 5" piece of cantilevered tslot bolted in the end by an end fastener and 1/4-20 screw without causing any damage. I am not familiar with square tubing but I would expect it to have different results.

I would have no qualms with supporting my weight on any of the bars on that frame.

Re: The Frame thread
 

I wouldn't. We picked up the entire 2009 robot on 2-3" cantilevers with abosolutely no visible flex. For the record: 1X1 1/8" wall 6061 Al.

Re: The Frame thread
 

We "discovered" a company ESTO that makes connectors that are primarily designed for trade show displays and such non-impact applications. Their connectors seem to hold up pretty well and make for some pretty quick prototyping and/or finished products.

Magically they seem help any project "new" students touch stay "square."

Re: The Frame thread
 

I haven't played with stress analysis on square tubing for a long time so this may be correct. I was more trying to get at the fact that while T-slot is heavy it is also very strong and has its merits (such as linear bearings)

Re: The Frame thread
 

We've run into much the same problem over the years, except not with weight, but with delivery time.

Taking a quick look at your design, it does look very bloody strong, which is good, and heavy. One thing that my team is looking very long and hard at is building a chassis out of 1/4 inch Al plate that has been machined down. A preliminary image is posted immediately below... But what we plan to do is cut lightening groves in it--- projections put it at ~20 lbs, and it incorporates our Electronics, Gearboxes, and bumpers nicely:

As you can see, it's just 1/4 inch plate on the sides, front, back, and top. Sides hold together the actual Drive Train mechanisms and two internal bars that support the Kit Metal in the middle, the top allows for the side bumpers to be mounted, and the back and front plates allow for the front and back bumpers.

Re: The Frame thread
 

the pic:

Re: The Frame thread
 

The chassis was 53 lbs with even more than what is seen there, I would estimate maybe 35 pounds for that including the linear bearings and all the hardware.

Re: The Frame thread
 

Here is Team 85's 2011 frame:



Some features:
-1/4" AL, bolt together construction
-Outer wheels articulate
-Very reliable, no issues all season long
-Requires significant lighting (not shown)
-Very stable even with the arm raised

Although the drive-train/frame was successful, we are opting for simpler designs in the future. We simply need more time to concentrate on the rest of the robot.

Re: The Frame thread
 

We found an cool way around having to deal with welding that super structure. The bumper mounts are pictured below and were relatively easy to use, although next year we may update the design a bit.



That is our drive base from this year fresh from the welder. (You can see the welds aren't even ground down.) 3"x1.5" tubing that is 1/8" thick doesn't require bearing blocks, and internal belts are easy to change with the window above the center wheel. This DT is REALLY simple and it doesn't even require CNC milling (although it is nice). Virtually no maintenance was required and our 9mm wide belts held up fine for 2 regionals.



Above is a concept drive based on our "WCD" from this year. Custom 3" wheels only require a single 12:56 reduction in the gearbox to go 12 fps. (The 56 tooth gear is ~2.8" in diameter so there is little ground clearance, but it is right behind the wheel anyways. The tubing is also only 2"x1") It weighs somewhere in the low 30s with CIMs and w/o the shown arm superstructure. I'm working on getting a similar sheet version out soon. Basically this is a lighter and simpler version of our robot this year (which weighed a whopping 98 lbs.)

Feel free to contact me for any CAD.

Re: The Frame thread
 

I like how we're trying to save weight when we came in at 95 pounds. Continuous improvement!

Re: The Frame thread
 

816 experimented with a "Drive train in a tube" style frame this year, similar to Team 221 LLC's "Rockbox". Either side of the drive was put into a 4x4x1/8" Aluminum Tube, and the two tubes were welded to some 1x1x1/8" Cross tubes to make the frame. The Bare frame weighed 15lbs without any pocketing. If we were to have done some modest pocketing we probably could've pulled out a pound or two but we didn't see the need to.

Here's a picture of the frame with all drive components and electronics installed:



In retrospect, the frame was easy to build and plenty strong for a game with bumpers and moderately light weight. If you've got welding and some basic machining abilities, it's not a bad way to build a frame, though there are better options out there.