What materials do you most commonly use on your robots?

[Mark Sheridan]

Last year we did a west coast drive style aluminum frame with sheet metal inserts. It was 1x2" aluminum box tubing 1/8" thick and 1x1" box tubing 1/16" thick. Our belly pan was 0.090" thick. Our climber was 1/8" sheet metal. I forget when steel we used in the climber but it was 1/4".

this year wear going heavier into sheet metal. We are back to having more sheet metal resources than welding. So likely most of the robot will be 6000 series aluminum at .125 , .090 and .060" thickness. to avoid welding, we will be machining and cutting gussets to hold any 1x1" box tubing we way use via rivets or screws. We had some good luck with ABS bearing blocks last year, so we will continue to use that. We makes ours thicker so they are for applications were a very thin aluminum block is not practical. We will also being making our larger gearboxes out of machine 1/4" aluminum. Our smaller boxes will be 0.090" aluminum sheet metal likely. A few other material we may use include Acetal, HDPE and PTFE. We may get into machining a few shafts with 7000 series aluminum, whatever we can dig up at the scrap yard.

[Oblarg]

449 just made the change from predominantly 80-20 frames to 2''x1'' aluminum tube (and good riddance to the 80/20, it's a terrible material that breeds bad habits and should only be used for prototyping, in my experience), assembled with match-drilled gusset plates and 1/4''-20 bolts. Easy to do and very sturdy.

4464 this past year used the standard kitbot c-base frame, with a superstructure of aluminum tube of varying sizes and riveted gusset plates, which worked great.

[bs7280]

Originally Posted by Oblarg

and good riddance to the 80/20, it's a terrible material that breed bad habits and should only be used for prototyping, in my experience

I fully agree with you there. Our past two robots (the first one doesn't count, it was our rookie year) were basically tributes to the company that made 80/20. The robots were almost entirely made out of 80/20 and were barely able to be finished within 6 weeks.

One of my goals for this year was to make a prototype 80/20 robot (to some degree) in 3 weeks, then remake everything better the second time around. I have found that testing the robot for only a day or two without any field components has left our robots with many hidden problems. With a lot of machines and more supervision, it will be a lot easier (hopefully) to make better robots in shorter time.

[RyanCahoon]

Whatever material you decide to use, make sure to experiment with it before the season starts. A robot made out of 80/20 that can play the game is better than spending 6 weeks trying to figure out a new material and not having enough time to refine your design. If you're barely finished your robot last year in the 6 weeks, make sure your team this year is able to be competitive before taking on additional challenges.

Originally Posted by MichaelBick

It all depends on your resources. 341 built one of the best robots in 2012 with 8020.

Quoted for truth. Resources means experience as well as material and capability.

[MichaelBick]

I strongly suggest everyone who thinks that 8020 is an terrible material read this: http://www.fightingpi.org/Resources/...Team%20341.pdf

As I have said before, 341's 2012 robot is one of the best, if not the best, Rebound Rumble robots. They finished the season seeding first at EVERY SINGLE EVENT they went to(including IRI and CMP), 87-9-1, division finalists, and IRI finalists. They maximized their resources, using what little resources they had only where they needed it. I love using box tubing because we have the machining resources necessary to fabricate with it and it permits us to build extremely light robots, but 8020 can be just as good if you need to build quickly with little resources.

[Oblarg]

Originally Posted by MichaelBick

I love using box tubing because we have the machining resources necessary to fabricate with it and it permits us to build extremely light robots, but 8020 can be just as good if you need to build quickly with little resources.

It is not true that you need significant machining resources to do tube framing; 449 has had a very easy time of knocking together gusset plate/tube constructions using pre-machined 80/20 gussets and match-drilling everything. This allows you to clamp everything in its correct configuration before drilling, and allows lots of slop with the actual drilling (if you're careful you can actually get away with doing pretty much the whole thing with a hand drill without any trouble, though a drill press is better).

I've found that the problem with 80/20, in addition to the absurd weight, is that it makes it very easy for people to get into the habit of making things which "sort of fit together" instead of things which actually fit together; the play inherent to the system allows you to nudge things around when you haven't really figured out where everything should be. This is very useful when you're prototyping, but it is the enemy of a good final product. You end up with things that gradually walk out of position and you get lots of silly, avoidable failure modes which shouldn't be there.

One can alleviate these issues to an extent with judicious use of locking hardware/loctite and through-drilling, but I find it more effective at that point, both practically and pedagogically, to simply manufacture the final product out of a better, lighter material. If you stick with 80/20 it takes a fair bit of wherewithal to ensure that all your important dimensions end up fixed and secured, whereas if you switch to manufacturing the final product you are essentially forced to do it right, else you can't construct the thing.

So, it's not really so much that 80/20 can't be used well on a final robot so much as there is a lot of potential for it not to be used well compared to other materials. Also, did I mention it's heavy?

[MichaelBick]

Originally Posted by Oblarg

So, it's not really so much that 80/20 can't be used well on a final robot so much as there is a lot of potential for it not to be used well compared to other materials. Also, did I mention it's heavy?

It's all about what allows you to build the quickest. Match drilling tube works well, as long as you have the resources to make gussets. However, not all teams have the resources. In this case, 8020 provides a quick and easy method to making parts, that uses resources available to the team. Furthermore, teams with limited resources cannot afford to remake parts. Instead, 8020 affords teams the ability to iterate without using up more machining resources. Lastly, even though it is heavy, 8020 allows teams to build robots effectively with a chopsaw and drill press. I'd much rather have a robot with 1 mechanism that works at 100% than 2 mechanisms that work at 50%.

8020 can be used effectively, and there is no need "nudge things around" if you "know where everything should be". Any material can be used incorrectly, but this does not make the material inherently bad.

[Oblarg]

Originally Posted by MichaelBick

8020 can be used effectively, and there is no need "nudge things around" if you "know where everything should be". Any material can be used incorrectly, but this does not make the material inherently bad.

"Bad" is not a binary value.

A greater propensity towards incorrect use is a detrimental property of a material, both from a standpoint of delivering a good product and perpetuating good engineering habits. 80/20 certainly has benefits, but it has noted drawbacks and, from my experience, it is a material I'd much rather not use in anything past a prototyping capacity.

I also think you're still overstating the cost, both in machining capability and time/effort, of building a tube frame; 449 has been machining with little more than a miter saw and a drill press for the entirety of the time I've been on the team, and we currently have no problems at all with tube frames. As mentioned, we don't usually make our own gussets (though we certainly have done, with success, on a bandsaw), but rather purchase ones from the 80/20 catalog and cut them down/redrill them as necessary.

[BurkeHalderman]

Originally Posted by Oblarg

good riddance to the 80/20, it's a terrible material that breeds bad habits and should only be used for prototyping, in my experience

Originally Posted by bs7280

I fully agree with you there. Our past two robots (the first one doesn't count, it was our rookie year) were basically tributes to the company that made 80/20. The robots were almost entirely made out of 80/20 and were barely able to be finished within 6 weeks.

80/20 is not a bad material at all. It all depends on the your teams resources and knowledge. There are actually many advantages to using 80/20. The T-slot profile allows for easy mounting and adjustment, it is designed to have the center tapped for 1/4" bolt. You don't even need to use gussets to assemble it as you can drill and tap the centers. While to use square tubing you have to either weld or use gussets.

Originally Posted by Oblarg

Also, did I mention it's heavy?

The 1" square profile, at .59lbs/ft is barely heavier than a 1" square, 1/8" wall aluminum tube at .524lbs/ft. So no it is not heavy. If you use 80/20 with an absurd amount of gussets then yes your robot is going to be heavy, if used properly it is not heavy.

80/20 extruded aluminum is a great resource to use, especially if you don't have access to anything more than basic tools. I'd recommend it to any team. As mentioned before Team 341's 2012 robot was predominantly 80/20 and was easily one of the best robots in FIRST that year. We have used an 80/20 and c-channel frame for the last three years and couldn't be happier with the ease of construction it has provided us and we've never had our frame wobble loose. In 2013 using a combination of 1" 80/20 and .120" aluminum sheet metal we had our practice bot completed by the end of week three and ranked 1st at the Queen City Regional

[MichaelBick]

Originally Posted by Oblarg

I also think you're still overstating the cost, both in machining capability and time/effort, of building a tube frame; 449 has been machining with little more than a miter saw and a drill press for the entirety of the time I've been on the team, and we currently have no problems at all with tube frames.

Again, I highly reccomend you give the article a good read. I think you are overstating the negatives of 8020. Essentially, 8020 can be used exactly like box tube, without the use of gussets. This can be a big benefit to many teams(anywhere from rookies to season veterans). Anything that "gives" you time for free is a good thing. Whether the time savings is more beneficial than the weight loss is based on the team and their resources.

Your original statement was that 8020 is a "terrible" material. My primary point is that this is obviously not the case, because a robot that was mostly built out of 8020 did so INSANELY well.

[Oblarg]

Originally Posted by MichaelBick

Your original statement was that 8020 is a "terrible" material. My primary point is that this is obviously not the case, because a robot that was mostly built out of 8020 did so INSANELY well.

The best robot I've ever worked on (449's bot for Overdrive) was made mainly out of 80/20. This does not offset the fact that by far the single largest source of mechanical failures I have witnessed on our robots, both directly and indirectly, has been 80/20 construction. I have not seen those same mistakes made with tube/gusset construction, simply because the modes of failure I've seen most often are simply not possible with tube/gusset construction.

[MichaelBick]

Originally Posted by Oblarg

I have not seen those same mistakes made with tube/gusset construction, simply because the modes of failure I've seen most often are simply not possible with tube/gusset construction.

How is it possible to make more mistakes with 8020 if you build in a very similar way, however without gussets?

[mman1506]

Any ideas for what to do with these 3/4 to 3 inch thick sheets of scrap T6 hardened 7075 aluminium we were donated?
https://fbcdn-sphotos-d-a.akamaihd.n...86326484_n.jpg

We've been thinking of making a 7075 belly pan

[Oblarg]

Originally Posted by MichaelBick

How is it possible to make more mistakes with 8020 if you build in a very similar way, however without gussets?

If you only use 80/20 in this way, you get none of the added utility when prototyping.

The problem, in my experience, has always been the transition from prototype to final mechanism with 80/20 construction. It is very easy for sloppiness from the prototype to bleed through, and it causes failures. Switching to tube/gusset construction forces you to not make these mistakes. There is utility in this.

I think we may have to accept that we're at an impasse, because we're dragging this thread off-topic and drowning out the other discussion.

[BurkeHalderman]

Originally Posted by Oblarg

This does not offset the fact that by far the single largest source of mechanical failures I have witnessed on our robots, both directly and indirectly, has been 80/20 construction. I have not seen those same mistakes made with tube/gusset construction, simply because the modes of failure I've seen most often are simply not possible with tube/gusset construction.

Can you describe these mechanical failures that were caused by using 80/20?

Originally Posted by MichaelBick

How is it possible to make more mistakes with 8020 if you build in a very similar way, however without gussets?

I'd assume it'd be due to parts shifting in the T-slot profile. However if you simply drill through the 80/20 after positioning it using the T-slot then the problem is resolved.