Weight Loss Strategies

[PaulYavarow]

How do other teams keep your robots under the weight limit? No matter what we do, we here at 1768 always end up just on the edge of the weight limit. Is there something we're missing here? We just drill holes into things and make as much we can from ABS to cut weight from the robot, but that can only go so far.

[Alan Anderson]

"An ounce of prevention is worth a pound of cure." Manage weight from the beginning so that you do not become overweight. Keep a weight budget and have the discipline not to exceed it.

It's good advice. It's not always easy to follow.

[protoserge]

Requirements, construction methods, material selections, weight allocations per component/subsystem, a lot of CADD modeling with weights included in the materials, the list goes on.


Plywood is a great building material, especially 1/2" Baltic Birch.

0.063" and 0.090" sheet metal with 1/8" steel rivets will do a lot of the super structure for many robot designs at minimal weight (<10 pounds on many occasions).

Set weight allocations for drive, chassis, command and control (sensors), electrical and wiring, mechanical arm and end effectors/shooters, etc. One thing you can do to help for the future is to track your robot weight every year by weighing before you build, or taking apart old robots. Learn how much weight 1/4-20 fasteners add and realize that a #6 or #8 screw can handle the load (yes, you can hang a robot in shear with one #6 screw, but don't try this ).

Another suggestion is to track your weight on a weekly basis to see if your targets/guesses are correct or on track. This allows you time to go back and redesign or make compromises in the trade space.

I hope that helps!

[tim-tim]

Originally Posted by PaulYavarow

How do other teams keep your robots under the weight limit? No matter what we do, we here at 1768 always end up just on the edge of the weight limit. Is there something we're missing here? We just drill holes into things and make as much we can from ABS to cut weight from the robot, but that can only go so far.

Paul,

As mentioned, doing this from the beginning is key; but not always easy. Take some time beofre the season starts to develop a weight strategy. We typically design to 110lbs which leaves 10lbs for error/expansion/unforeseen issues/etc.

Definitely start weighing stuff from the beginning. As you mock up mechanisms for the game get weights for items that will be used on the real robot (transmissions, motors, electronics, etc.). Then develop an [educated] estimate of what the final design will weigh. Keep track of the estimated weights, for comparison purposes to the actual weights when built.

This takes a lot of discipline though. Weekly weigh ins are good, but take it up a notch. Note what was included in the weigh in, and compare to what the estimated weight for what was weighed. This wil not only give you a current weight, but indicate whether you are over, under, on weight 'budget'.

Something else to consider is the design intent of each part. Not all parts need to be 1/8" with 1/4"-20 bolts (just an example). Look at the purpose of the part and evaluate what strength it will actually need. For example, we are using 0.032" (1/32") aluminum in some areas on our robot this year that have low amounts of stress. This will come with experience.

A short list of things to do is know what weight you can't get rid of; meaning know the weight of the drive motors, cRIO (and other electronics), chassis, wheels, etc. We had a barebones driving platform weighing in around 35-40lbs this year (sorry I don't have the exact figure). This means we only had ~80lbs to 'play the game'. Realize that this 40lbs for drive is very hard to lighten due to the amount of required components; so your weight reduction will likely have to come elsewhere.

Some things to do to lighten the robot:

• Remove any unnecessary material - remove corners that aren't supporting weight/components
• Replace 1/4"-20 steel nuts, bolts, and washers with aluminum ones in non-critical applications - aluminum is about 1/3 the wight of steel
• Assemble with rivets where possible - remove bolts and add additional holes for 1/8", 5/32", or 3/16" rivets. Be sure to get the proper grip length. Again, use aluminum in most applications.
• Iterate design - develop a lighter mechanism that has no loss in performance, in fact there should be an increase in performance, that can be added during competition

If you can provide additional info, I would be glad to offer some suggestions for this year (if needed).

[thefro526]

In my experience, the big secret to keeping weight in check is to start losing it in the design phase. After an experience in 2011, where the team I was with at the time needed to 'find' 3lbs or so before our second event, I've been actively trying to design everything I'm involved in to be as light and efficient as possible from the beginning.

Some of my favorite techniques are:

- Use 10-32 SHCS/BHCS and 1/8" Rivets as standard hardware instead of 1/4-20. Most of the time, especially in FRC, 1/4-20 is used in situations where it's way overkill. Swapping out a 1/4-20 Bolt for a 10-32 nets ~20% weight savings per fastener, which adds up over time.

- Use 1/16" Wall Extrusions when possible instead of the standard 1/8". There are numerous instances where I see 1/8" Wall Extrusions used because they're common and 'easy to find' - but aren't really needed. Swapping this out to the equivalent extrusion in 1/16" Wall thickness nets a ~50% savings. (* try to avoid thin 6063 extrusions though, sometimes they end up being a nightmare)

- Use .060 (nominal) Wall 6061 Tubing instead of PVC/ABS/Etc for rollers. This can really add up on a system with a lot of rollers.

- Swap out larger gears in COTS transmissions for their aluminum equivalent. West Coast Products makes 7075 versions of many of the AM gears, swapping out two or three gears can add up to a pound of weight savings really, really quickly.

- Avoid thick polycarbonate and other plastics. Many people think that because plastics are relatively light, they can get away with using thicker sheets which often leads to 'hidden weight'.

The list goes on and on from here, but I hope you get the idea. Also, sometimes, it's worth trying the 'lighter' version of something first. If it fails, you'll know that you can easily replace it with something stronger/heavier. If you do the opposite, sometimes it's hard to replace that part because it's been on the machine for too long and/or is too well integrated to make it easy.

[JamesCH95]

Originally Posted by thefro526

In my experience, the big secret to keeping weight in check is to start losing it in the design phase. After an experience in 2011, where the team I was with at the time needed to 'find' 3lbs or so before our second event, I've been actively trying to design everything I'm involved in to be as light and efficient as possible from the beginning.

Some of my favorite techniques are:

- Use 10-32 SHCS/BHCS and 1/8" Rivets as standard hardware instead of 1/4-20. Most of the time, especially in FRC, 1/4-20 is used in situations where it's way overkill. Swapping out a 1/4-20 Bolt for a 10-32 nets ~20% weight savings per fastener, which adds up over time.

- Use 1/16" Wall Extrusions when possible instead of the standard 1/8". There are numerous instances where I see 1/8" Wall Extrusions used because they're common and 'easy to find' - but aren't really needed. Swapping this out to the equivalent extrusion in 1/16" Wall thickness nets a ~50% savings. (* try to avoid thin 6063 extrusions though, sometimes they end up being a nightmare)

- Use .060 (nominal) Wall 6061 Tubing instead of PVC/ABS/Etc for rollers. This can really add up on a system with a lot of rollers.

- Swap out larger gears in COTS transmissions for their aluminum equivalent. West Coast Products makes 7075 versions of many of the AM gears, swapping out two or three gears can add up to a pound of weight savings really, really quickly.

- Avoid thick polycarbonate and other plastics. Many people think that because plastics are relatively light, they can get away with using thicker sheets which often leads to 'hidden weight'.

The list goes on and on from here, but I hope you get the idea. Also, sometimes, it's worth trying the 'lighter' version of something first. If it fails, you'll know that you can easily replace it with something stronger/heavier. If you do the opposite, sometimes it's hard to replace that part because it's been on the machine for too long and/or is too well integrated to make it easy.

All of this x2.

We use these techniques and am critical of every component's weight as we make them. We ask questions like: does this sheet really need to be 0.125in or can it be 0.100in? Can we use smaller or shorter bolts? Can this part perform multiple functions? Can we weld it instead of using fasteners? I.e. welding a kit frame drops 3+lbs of brackets and fasteners.

Material selection, even subtle differentiation, can be very important. For example, instead of going to HomeDepot and getting 1/8x1x1 6063 aluminum angle, order 1/16x1x1 6061 angle from McMaster. It's just as strong and half the weight. Where stiffness is important we use *gasp* steel!

Using these sorts of practices since 2010 we've been underweight by 5-15lbs without a weight budget, serious CAD'ing, or removing any mechanisms.

Edit:

FWIW we also up-size motor wire, i.e. instead of 12awg we use 10awg.

Don't forget that pounds are made of ounces, when you start dropping ounces whenever possible good things happen. To paraphrase Talladega Nights: "If you're not reducing weight, you're adding it."

[Alan Anderson]

Originally Posted by JamesCH95

Don't forget that pounds are made of ounces, when you start dropping ounces whenever possible good things happen. To paraphrase Talladega Nights: "If you're not reducing weight, you're adding it."

That reminds me of the old backpacker's adage: "Take care of the ounces, and the pounds will take care of themselves."

(Yes, I know it's a riff on the even older pence/pounds saying, but nobody uses pence as money anymore.)

[AdamHeard]

Originally Posted by thefro526

- Use 10-32 SHCS/BHCS and 1/8" Rivets as standard hardware instead of 1/4-20. Most of the time, especially in FRC, 1/4-20 is used in situations where it's way overkill. Swapping out a 1/4-20 Bolt for a 10-32 nets ~20% weight savings per fastener, which adds up over time.

A #10-32 is actually closer to 60% the weight of a 1/4-20.

[eedoga]

Our team has switched over on a number of fronts and we really haven't had a weight problem since year two when we found ourselves 20 pounds over weight 2 days before ship date. (I really hope this doesn't come off as preachy...Our team motto is "we don't lose, we learn", and we have done a whole lot of learning.)

1 - Take advantage of and use CADD. Model the heck out of everything ahead of time and over budget based on the weight estimates you get. (this will save you time and money as well.)

2 - Use the stress analysis tools to see where you can get away with using less. I regularly see teams using 4 1/4 inch nuts where two might just as well do the job, or better yet, a weld or rivet which would require no screws and nuts at all.

3 - Keep track of the weight of subsystems

4 - Use shipping weights to estimate the weight of parts before you order them. This allows you to swap out heavier parts for lighter ones.

5 - Try to have parts have duel purpose. Use a power take off system to limit the number of motors you need, or lose entire subsystems and have a more focused robot.

6 - Choose materials carefully.

These things have helped us...Hopefully they will help you too!

Edoga

[rwkling1]

One way that we keep underweight is using sheet metal. We started using it this year and we hve seen a huge weight decrease from our previous robots. Last year at competition we weighed in at 199.9 lbs... no joke, this year before we bagged it we weighed around 105. Sheet metal is great for electronics so you don't add lots of weight by using thicker metal.

[Cal578]

Originally Posted by rwkling1

... Last year at competition we weighed in at 199.9 lbs...

Did you mean 119.9?