I recently learned about these: http://www.pneuaire.com/reca44cuin.html pneumatic air tanks that are both light, and effective. My team is big on saving weight, and would like to be able to more control what our robot weighs without drilling 30,000 holes into it.
So, what are your secret tips on saving robot weight?
Use a stick on a motor whenever possible 
Take a spreadsheet and a scale. Weigh every part, and estimate weight where you can, before the part goes on the robot. As you figure out you’re close to being overweight, figure out where you can trim weight.
I did this in 2010 and works great. You gotta make sure to get the exact weight on each part and classify large pieces of wood/ aluminum based on weight per foot or inch to avoid confusion in the middle of build season.
P.S. It is very time consuming to weigh each part.
Can’t echo this enough! Having a reasonable weight budget for subteams to stick too is key along with having an allowance for essentials like electronics and such.
Keep weight savings in mind while you are building/designing. Our team ate up weight fast on 1x1x1/8 aluminum box that could have easily been replaced by 1/16 but it was too late to change.
It is also a whole lot easier to swiss cheese while the part is not yet on the robot!
Team RUSH hasn’t used pneumatics in their robots since 2004 and this has been a tremendous weight saver in my opinion.
Initially we stopped using them because we thought the package available at the time was too heavy for the utility you achieved. We would often have one or two systems that were candidates for pneumatics…this didn’t justify adding the ~8 lbs. needed for the compressor, tanks, hose etc, etc. Of course we discussed eliminating the compressor for some designs, but in the end we just eliminated the whole system.
Originally I hated that we didn’t have access to pneumatics…each year I would lobby for it. Now that I recently started helping a new team, I find myself lobbying them to eliminate it to save complexity and weight. 
Of course a lot of teams are able to integrate all motors and a complete air system while staying in weight, but it’s a very liberating feeling heading to scales with 8-10 lbs. to spare.
Now that the systems have become more open and lighter I’m sure we’ll take a serious look at going back.
Would any team care to post a past weight budget? I’m curious to see the average weight composition of the FRC robot.
Save hella weight - buy aluminum gears
Design everything ahead of time in CAD. If you set the materials appropriately, you can get an accurate estimate of your final robot weight. The more detail stuff you model, the more accurate the weight estimate will be. Changing designs in CAD to reduce weight is much easier, cheaper, and more effective than having to drill hundreds (potentially thousands) of holes in your mechanisms after they’re already assembled.
I don’t know much about how you build your robots but making everything in 1/16 wall tubing instead of 1/8 will save a lot of weight and time(if you pocket 1/8). It is also stronger than pocketed 1/8 wall tubing. You can save a lot also if you use aluminum rivets instead of bolts whenever possible.Smaller wheels and custom gearboxes can save weight if done correctly(if you are unsuccessful in designing your own just copy our 2010/2011 gearboxes/drivetrain(link below)).
Design it to be underweight. That’s why they give us CAD!
Ike put together a nice paper on robot weight reductions. http://www.chiefdelphi.com/media/papers/2220
Start off by weighing all parts and recording that weight in an Excel Spread Sheet.
For those parts that are designed in Cad set the the material density correctly and CAD will calulate the part. Insert that number in the Excel Spread Sheet. Where possible use Aluminum. Limit the number of bolts to what you actually need to prevent rotation or shear. If you have to drill holes to mount something then use those mounting holes to mount something else if possible. That will reduce the number of bolts. If you keep track of weight only in CAD then use lump masses for those things like electrical.
The weight spread sheet can also be used for cost of your Robot because you have to account for every part on the Robot.
Fortunatily programming code does not weigh any thing! (
I have the mindset of making everything as light as possible before adding it to the robot. “Pounds are made of ounces” is one of my favorite mantras from building race cars that I have brought to robotics. That is, we try to lighten every part before it goes on the robot. The savings seem trivial at first, but they add up to be significant.
Some tricks:
-welding the kit frame saves ~3lbs of fasteners and blocks and eliminates loosening frame fasteners from your pit checklist
-plastic bearings are super-light and can replace metallic bearings in some applications
-using 25 series chain instead of 35 series chain drops a significant amount of weight, I would not consider this on the drivetrain except as a last resort though
-designing with an eye to keep wiring runs as short as possible not only reduces weight, but improves electrical system efficiency (consider making that beautiful wiring run a little more utilitarian)
-the new air compressor is around 2.6lbs lighter than the older compressor, that’s huge (FWIW I aim to use pneumatics for at least 3 small functions or 2 big functions to consider them worth the weight)
-using high-grade fasteners can drop weight if they are selected properly
-consider eliminating redundant fasteners
-consider replacing low-load fasteners with zip-ties, plastic bolts, or velcro
-speed controlling through software rather than excessive gearing where possible, one encoder weighs a lot less than another sprocket and chain set or pair of gears
There is very rarely a weight-reducing trick that saves a huge amount of weight. In my opinion good light-weighting is accomplished through a systemic discipline of cutting ounces, or fractions of ounces, wherever possible.
Using thinner materials is the most frequent weight reduction I see. Often teams will use 1/8" plate where 1/16" will do (be it plastics, aluminum, or whatever…).
This is especially true wih Body panels. I frequently will see teams with large 1/8" body panels. Often 1/16" or even 0.030" thick polycarb will work just as well. In fact the 0.030" polycarb and access to a vinyl cutter make your robot looks classy, and add protection for very little weight.
When working with sheet-metal, use “shape” to add stiffness, not thickness. Adding in flanges can often increase the stiffness of a panel with very little (+10-20%) additional weight. Adding thickness will often do very little (compared to flanges), and increases the weight proportionally. Adding thickness and then removing material via a CNC, looks cool, and functionally may work well, but does require a fair amount of work or tooling.
The pneuaire.com accumulators are a great way to reduce weight and add air volume. I saw a bunch of these on 1503 last year and got the manufacturer from them. We used several on our robot, and i had a whole bunch of spares that we donated to another team that had around 10 metal accumulators on their machine.
Be careful with aluminum gears. When done right, there are some significant weight savings opportunities, but know that is is difficult to do them right. Aluminum has several properties that are quite poor for gears. Many gear experts would recommend to NEVER use aluminum. I do think they have their place, but you should be extremely careful and really need to know what you are doing.
Beware of the attitude that “its too late to replace…”. I have heard teams say this before, only to remove functional mechanisms at competition in order to make weight.
Composites can save weight over AL. If I remember correctly, This years arm which was made of pultrutions and polypropylene weighted 65% compared to the same structure in AL. Working with composite require methods that are different than working with metal. Do the research before using them.
Lightening the steel gears, while not particularly FUN to do, isn’t very hard and will get you a decent amount of the way there with regards to weight savings.
I’ll put in a vote for composites as well. Pultruded fiberglass is very rigid, thick stuff with a low density. Fiberglass tubing is great for something like an arm.
Absolutely, I can’t believe I forgot to add “lightening” steel gears. For those that are not familiar, you can either cut pockets with a mill, or undercut the gear between the teeth and the hub on a lathe (this is what we usually do). When doing this, make sure to leave a littl material underneath the root of the gear tooth as this can be a highly stressed location on a gear. Also leave material for the hub as this is either the torque transferring area, or has the bearings on an idler gear. We can often cut about 50% of the weight out of the standard AndyMark gears.
Just be careful how you “cuck up” the gears in a lathe so you do not damage the teeth.
You can do the same thing with sprockets. Just be careful, for the same reasons. A couple of large sprockets have failed catastrophically in competitions after being lightweighted. I remember or remember hearing about two incidents from 2004–in one case, the sprocket turned into a pretzel; in the other, the hub broke free after the season, disabling that particular robot by removing one of its highest-traction wheels from the drivetrain.
One trick to saving weight is to get rid of unnecessary parts.
Another thing i advise teams NOT to do is to use plastic chain where any significant torque is involved… I’ve seen it bomb catastrophically before… remember that with chain, one weak link ruins the whole loop.
Agreed on using 1/16" where possible…
one last tip is that if you have to resort to cheese-holing, do it to low load things higher up on the robot… having a low center of gravity is key to a stable robot. every little bit helps.