We have a very young and inexperienced team this year. Not only with robotics, but with fabrication too. Does anyone know about 80/20 Inc? Would this be a good way to go for prototyping and possibly even for the robot? If so, does anyone have a good supplies list that I would need to buy so I don’t buy the wrong things or over/under buy?
Thanks! FRC 2240
80/20 is great and fairly versatile (we used it our first year). You can get 1" stock and tap the ends for nice joints when making square corners, and the sliders allow you to quickly move things around to find the right place, without needing to drill holes.
All that said, 80/20 is heavy. I mean really heavy. None of the robots we’ve made since our rookie year could have been made with 80/20 while still being under weight. Last year, I saw team with a bunch of 80/20 on their robot come in 30+ lbs over the weight limit at a pre-ship scrimmage… I believe they made the limit just by replacing that with simple 1" square stock.
You picked the right time to start asking questions… you still have 2.5 months to help your team gain experience! Find your local discount metal supplier and get a bunch of square/rectangular tubing ans sheet metal fairly cheap, then have your team start making things. Build a box. build a book shelf. Get them building stuff, and they’ll have a much easier time of it when the build season starts!
Plenty of teams successfully utilized 80/20 for framing and structural elements this past year and others…341 comes to mind. Contact them for advice?
Our team has used 1" 80/20 every year we have been in existence (3 years). The ease of tapping and attaching items is really nice for us. However, I can see the argument that it is too heavy. Our robot (during build season) in 2011 was 40 lbs overweight at one point. Oops! :yikes: Last year, we did a much better job of weighing as we went and making a more efficient design.
I have also seen some teams use quick frame
http://www.8020.net/Quick-Frame-2.asp
which I would imagine would be lighter. Not sure about it’s durability in the robotics environment though.
We will be a 2013 4th year FRC team and have used both Bosch-Rexroth and 80/20 to a steadily increasing amount over all three prior years.
Last year however, we opted for using the KOP 6-wheel tank drive “kitbot on steroids” design with KOP framing for the wheels & gearboxes. This was to ensure that we had a drivable robot at the earliest possible time. The KOP channel is not very sturdy however, and with a 4-CIM drive train this KOP frame needs some stiffening. The 1" 80/20 extrusion fits nicely inside the C-channel to strategically stiffen it, and we then transitioned going upward to 80/20 frame above the drivetrain frame level for the rest of the robot structure.
Be sure to get a lot of the predrilled corner gusset plates for the high stress joints. The price is a bit stiff on the 80/20 line, especially the fasteners.
Get as many of the drop-in T-nuts that spin 90 degrees after insertion, as you can afford. These are available, for 1" (M8 slot width) in three thread tappings - 1/4-20, 10-32, and 6mm.
They are worth their extra cost to be insertable into any slot location no matter what else is already assembled into that slot. The nuts that roll into the slot are horrible and overpriced. most of the other 80/20 nuts are also a PITA to work with, so just try and standardize on the drop-in & twist T-nuts.
Buy a bunch of zinc plated chrome-moly B7 threaded rod and cut it to length when you require custom, longer length, bolt-thru-hole-in-frame assemblies.
Then by just adding a a flanged nut with serrated locking face at the threaded rod end opposite from the rod’s slot end T-nut, a strong clamping action can be achieved for low cost. Over the years we accumulate more recycled pieces and it is great to use these for quick prototyping of ideas.
-Dick Ledford
Actually, 341 used the bosch extrusion.
I have been using 8020 since my freshman year, and it’s great in some aspects, and not so great in others.
Pros: Quick. The time between design and implementation/first prototype is generally pretty fast. Things can slide around if the dimension isn’t right the first time, and all in all, it usually means less strain on in house resources.
Cons: Weight (Ought to be obvious).
Flexibility - I’ve found that, at times, 8020 doesn’t give me the flexibility I need in design. Or rather, I can get the design I want, however, it yields a result that’s clunky, uses way too many gussets, and isn’t the most ideal path. In these scenarios, you want to be able to switch over to sheet metal/wood/polycard (the shooter hood from this past year comes to mind).
For a young, inexperienced team as you have described, I’d recommend 8020, but use it cautiously, as teams find that they have room for error, they can get sloppy, which is not good. Additionally, teams lose track of their weight and they end up going pounds over.
As for parts, here’s what was a part of our most recent order:
1010-97 - Part number for a 97" piece of 8020. You want to get quite a few of these. A thorough robot might end up using hundreds of inches of 8020.
4132 - These are gussets that allow you to connect inner corners. So if you have two pieces of metal meeting up at a T, then you would use these pieces to connect them. These things tend to be semi-expensive, but you end up using them quite a bit.
4042 - These are little block gussets that can connect pieces that meet at a corner. Expensive ($10/per) little things, but a robot may not use more than 10-15 of them.
3382 - T-Nut inserts are used in the profile to screw things down.
Screws - You can purchase these from anywhere. I’d recommend a local, cheap, bulk source.
Aside from this stuff, you’ll have to get what you need. There’s some pretty need stuff in the catalog. There are more gussets and connections that you may find useful come the design phase.
Hope this helps,
- Sunny G.
This will be FRC Team 2733’s 5th year, and I think that they have finally decided on a building method that works for them. By using the box extrusion similar to Quickframe and other such products, and gusset plates made from .050" aluminum sheet metal, they are able to create forms with amazingly complex shapes with nothing but a band saw, drill press and popriveter. One of the goals of FIRST is to expose students to new ideas and methods. In my opinion, 80/20 is a great prototyping tool, but at competition leads to a lot of trouble with things getting loose and requiring upkeep it also is not a new concept for anyone who has ever used an erector set. For a rookie team, it would be simple to use a KOP chassis, and then use 80/20 to build the rest, at the end however, converting the 80/20 bars with 1/16th wall box extrusions will save tons of weight. This is the trick to using 80/20, using as little as possible on the final robot.
For prototyping, Esto Connectors work really well, are reasonably priced (they also sell the square tubing on the website) and I’ve always gotten quick turnaround from them. We’ve used it for some framing on competition robots but generally it turns out to be too flexible for anything under load (especially dynamic loading). But still, the joints can be built up with the plastic connectors and then replaced or reinforced with shear plates once the design is firmed up. The tubing is very lightweight.
I’ve never used 8020 on a competition robot (always been a weight concern) I have used it professionally though.
I think it really does get you “80% of what you want, for 20% of the effort.”
As a side note: we’ve never had an issue with the kit-bot frame never being stiff enough or strong enough. Though we always mount a board in the center of the frame (plywood or composite) to mount electronics and stiffen the frame. Welding the kit frame drops about 3lbs in fasteners and brackets.
Regarding the weight issue, be careful how you replace 80/20. The common (and lightest) 1" square profile (1010), is .5097lb/ft. Compare to different 6061-T6 Al profiles.
lb/ft…Extrusion Profile…Length req. for -30lbs
0.513…1" (1/8") square tube…N/A
0.5097…80/20 1010 Profile…N/A
0.410…1" (1/8") channel…301ft
0.285…1" (1/16") square tube…134ft
0.280…1" (1/8") angle…,…131ft
Values vary slightly, primarily depending on the exact extrusion spec (AA, AS, etc).
Depending on your application, you’ll likely be able to get away with lighter-weight extrusions in many situations (U-channel is usually* a reasonable structural substitute), but it’s not as big a difference as many think.
*As in not always.
We take a similar approach with the esto connectors, using them in minimal load situations and for prototyping. The ability to put together a reasonably square structure in such a short time is handy. They will also powdercoat the tubing they sell if you order specific amount so it can also be handy for making displays and other structures, which is what the product is actually intended to be used for.
Edit:
We used 80/20 during our first two years (2007 and 2008) but found that keeping everything tight and in place was a challenge so we swapped to ESTO. We still have a decent amount in our shop but it only seems to be used for a linear motion prototype anymore
We actually use Bosch-Rexroth 20mm extrusion for structure on our robot (this year we used both 20x20mm and 20x60mm). The 20x20 is .269 lbs/ft and the 20x60 is .605.
The strength/weight ratio for T-slot extrusion is similar but slightly inferior (within 15% IIRC) to square tubing, but what you get in return is an easy way to tap the ends of the profile (we generally use a 1/4-20) and the ability to bolt crosswise through it without having to worry about “caving in” the sides of the tube. If you have limited machining resources, it’s a good way to rapidly build structure. It is also a great way to let the students get their “hands dirty”. But, as noted, expect to pay a slight penalty in both weight and materials cost. If you are using 1" or 30mm profiles, the weight adds up REALLY quickly. Our use of the 20mm profile helps to alleviate this, but even then there are times where we opt for other materials/profiles to get better stiffness or strength for the weight (our 2011 arm comes to mind).
We generally do not use T-nuts or specialized connecting hardware (the exception is for mounting things like optical sensors that will bear no load) because they drive up weight/cost and slide all over the place. Instead we drill through-holes and/or tap the ends of the profile to build solid connections.
In an ideal world, we would rapidly prototype using extrusion and then go for more streamlined square/rectangular tube + gusset plate or even sheet metal for final mechanisms. In the past (including 2012) we often run out of time before we get to this level of optimization (and have also had widely varying levels of machine shop access from season to season).
At one of the tournaments we attended this past year, we saw a team using Micro Rax (www.microrax.com) for parts that did not need the full strenght of the 80/20 extrusions.
I don’t think we have ever built a robot without at least a little 80/20 in it.
This year it was a couple cross braces and we also used it in an extended pickup we added after our first regional. The slots and their adjustability are great for adding devices post-build.
It is great for prototyping and we use it every year for that purpose. We have been using some of the same sticks and hardware for over 10 years. It’s re-usability is one of it’s big benefits.
967 uses 20x20mm and 20x40mm Bosch extrusion, and like 341 we often tap the ends for 1/4-20 for making corner connections. We like the 5 mm block nuts and small corner gussets. We sometimes use 10-32 screws in the M5 block nuts, because they fit nicely, and we have lots of 10-32 hardware. For the last three years, we’ve used a kit frame for the drive with Bosch structure coming up from there. In 2009 we also made the drive base out of Bosch, but that consumes a decent amount of precious time compared to tossing a kit frame together.
The sliding block nuts are pretty nice sometimes. For example, last year our shooter used a pair of 20x20 extrusion rails as fixed ball guides. Before we built it and played around with it, we didn’t know the exact angle or distance from the wheel that we’d want, so it was nice to work with an easily adjustable mechanism.
http://www.flickr.com/photos/lmrobotics/6965154821/in/photostream
T slots and block nuts can only handle so much force, and if you exceed that the T slot edges bend outward, allowing the nut to come free. So when something is going to get loaded up, you have to drill it out and bolt it through the extrusion.
We used to build frames out of 30x30mm and 30x60mm extrusion, but we switched away from that because it’s heavy. 20x20 mm is a nice FRC size.
It would be cool to prototype with Bosch and then transfer the dimensions we come up with to a lighter, tighter sheet aluminum design, our team has a ways to go before we’ll be able to manage that sort of design cycle within a 6 week build season.
2363 has a presentation we give on how we use 8020 in our frames in a weight efficient manner. I’ve tried to upload it as a white paper, but this doesn’t seem to work. The file may be too large. You can see the presentation here.
Our 2011 robot:
https://lh3.googleusercontent.com/-Z0gW5DsGP3A/TksVfPy7-fI/AAAAAAAAAIg/3feBiYo5JUY/s576/100_2916.JPG
Our 2012 robot:
We love it.
It’s easy to work with, easy to change around, doesn’t bend or deform (that I know of) and we seem to have a lot of it lying around.
http://farm8.staticflickr.com/7197/6980571552_acd4d4f76e_z.jpg http://farm6.staticflickr.com/5178/5463062502_6b6fd809ba_z.jpg
We even used it for our pit:
It’s kind of on the heavy side, but we feel it’s worth the extra pounds.
Hey, us too!
Thank you everyone for your comments, suggestions and ideas!!! I will get some 80/20 and look at some of the other materials you all have suggested.
This will hopefully get us through the first year. Right now we have 18 interested students and only 5 have done FIRST before for which 2 are programmers, and in the “building team” 2 have 1 year experience and 1 has 3 yrs exp. I hope to have everyone prototyping immediately and try to get some depth with knowledge and fabrication experience as soon as possible. I’m hoping for “Fail sooner to succeed sooner”… a slogan used by IDEO
If anyone else has ideas on how you used it that would benefit this young team, please let me know.
Thanks again for your help, the FIRST Community is the best ever!