pic: Chassis Idea

[AdamHeard]

Originally Posted by Formerly Famous

loading photo ...

I'd be very weary of the weight you posted for a couple of reasons; You don't have everything modeled (chain, etc..), and even for what you have modeled, 30 lbs seems really light.

1/8" wall 1x1 is around .5 lbs a foot, and based on your 27x37, that puts you at around 17-18 lbs just for the 1x1. Add in two CIMs at 2.75 lbs each and the Gearboxes at 3.5 each and you see it starting to add up; this isn't even counting your mounting plates or wheels.

I'm not criticizing your design, it looks like a decent start to a decent drive. I just want people to realize that the weight a CAD program is the weight exactly as the model is modeled; So, if the model isn't accurate to real life in terms of included properties, applied materials, and custom mass properties (Setting the weight of parts like CIMs), the number won't be accurate either.

For the drive itself, I'd recommend you switch to a good deal of 1/16" wall, switch to dead axles and prioritize a light & robust tensioning system.

[Rion Atkinson]

Originally Posted by AdamHeard

I'd be very weary of the weight you posted for a couple of reasons; You don't have everything modeled (chain, etc..), and even for what you have modeled, 30 lbs seems really light.

1/8" wall 1x1 is around .5 lbs a foot, and based on your 27x37, that puts you at around 17-18 lbs just for the 1x1. Add in two CIMs at 2.75 lbs each and the Gearboxes at 3.5 each and you see it starting to add up; this isn't even counting your mounting plates or wheels.

I'm not criticizing your design, it looks like a decent start to a decent drive. I just want people to realize that the weight a CAD program is the weight exactly as the model is modeled; So, if the model isn't accurate to real life in terms of included properties, applied materials, and custom mass properties (Setting the weight of parts like CIMs), the number won't be accurate either.

For the drive itself, I'd recommend you switch to a good deal of 1/16" wall, switch to dead axles and prioritize a light & robust tensioning system.

I myself have been questioning the weight. I don't know exactly how much it will weigh. And honestly. There is no easy way to find out. (Without building it) The main thing that has me questioning is that is says this weighs 43.54 pounds...

The reason I went with live axles is simply because I have designed a live axle wheel. I will be making a dead axle of the same wheel and making a dead axle version of this.

Thank you for the notes. I will be taking all of them into account as I go over this design again, and again, and again.

-Rion

[artdutra04]

Originally Posted by Formerly Famous

I myself have been questioning the weight. I don't know exactly how much it will weigh. And honestly. There is no easy way to find out. (Without building it) The main thing that has me questioning is that is says this weighs 43.54 pounds...

The reason I went with live axles is simply because I have designed a live axle wheel. I will be making a dead axle of the same wheel and making a dead axle version of this.

Thank you for the notes. I will be taking all of them into account as I go over this design again, and again, and again.

-Rion

The AndyMark shifter in your CAD rendering looks just like the imported STEP file, which doesn't really have any mechanical properties. Open each individual part of the imported STEP assembly (they were all converted to .sldprt files), set the correct material, then manually set the CIM motor weight to 2.75 lbs.

[JVN]

Originally Posted by Jared341

Dead axles are useful because:

• You are minimizing the rotational mass of the wheel assembly and therefore maximizing efficiency since the shaft itself doesn't need to spin. Unless you have a truly massive axle or bad bearings/alignment, however, we're probably not talking a huge difference.
• In the cantilevered case, you are minimizing torsional loads on ball bearings with dead axles (since the bearings are still centered on the axis of rotation no matter what the deflection of the shaft).
• The closer together two bearings are, the easier it is to have them stay aligned. In a dead axle setup with two bearings in a single wheel, alignment is trivial. Live axle systems that support both ends of the shaft can get out of alignment more easily.

Did I miss anything?

We also use dead axles as structural members of our chassis (a round axle with tapped ends = chassis standoff.)

If you get creative with your axle placement and the structural design of your chassis this can eliminate the need for other structure and result in a lighter overall system.

-John

[Rion Atkinson]

Originally Posted by artdutra04

The AndyMark shifter in your CAD rendering looks just like the imported STEP file, which doesn't really have any mechanical properties. Open each individual part of the imported STEP assembly (they were all converted to .sldprt files), set the correct material, then manually set the CIM motor weight to 2.75 lbs.

I just noticed that when I opened it, there are no part files. I have probably done something wrong. If that is the case, how to I fix this?

Also. I have no idea what type of material to use for each part. I discovered the side plates are steel, past that I am lost. (RC, you need the GEN 1 in the CAD Library. )

If anyone could help me with this; please do not hesitate.

-Rion

[AdamHeard]

Originally Posted by Formerly Famous

I myself have been questioning the weight. I don't know exactly how much it will weigh. And honestly. There is no easy way to find out. (Without building it) The main thing that has me questioning is that is says this weighs 43.54 pounds...

The reason I went with live axles is simply because I have designed a live axle wheel. I will be making a dead axle of the same wheel and making a dead axle version of this.

Thank you for the notes. I will be taking all of them into account as I go over this design again, and again, and again.

-Rion

Okay, that is definitely not 43 lbs.

What you need to do is make sure every component is set correctly. If it's a machined part or metal piece that is a uniform material set it's material in SW, that will make apply the proper density and weight. For everything else go to tools > mass properties > check the box for assigned mass properties, and manually assign a mass. CIMs are 2.75 lbs, compressor ~5, etc... Someone posted a spreadsheet a while back that had most of the electronic's weights. You can do assigned mass properties for an assembly, and ignore the properties of it's components; for example, set the shifters to 3.4 lbs + 2.75 lbs (the weight of a CIM plus the gearbox).

As for trusting parts that have been imported or given to you from others; just don't. If a goal of yours is to use your CAD model to provide accurate weight estimates, get in the habit of making sure all parts are set correctly as they are made and/or inserted into the assembly. It is a horrible experience trying to figure out later on why the robot is 20 lbs lighter/heavier than you think it should be and you have 1000+ parts in the assembly.

[,4lex S.]

A Box Drivetrain like this generally looks quite good on its own, but when you consider a real FIRST robot, some flaws pop up. From a structural standpoint, you need to ensure that you will be able to adapt this design to manipulators.

About 50% of the time my team had to take a chunk out of the front of the chassis to facilitate claws or collectors. Just keep this possibility in mind and adapt your design for potential geometry changes.

If you can't get it welded, this design should work quite well with rivets. Triangular corner gussets might be a good idea for some of the pieces, rather than just right angle aluminium.

[Rion Atkinson]

Originally Posted by AdamHeard

Okay, that is definitely not 43 lbs.

What you need to do is make sure every component is set correctly. If it's a machined part or metal piece that is a uniform material set it's material in SW, that will make apply the proper density and weight. For everything else go to tools > mass properties > check the box for assigned mass properties, and manually assign a mass. CIMs are 2.75 lbs, compressor ~5, etc... Someone posted a spreadsheet a while back that had most of the electronic's weights. You can do assigned mass properties for an assembly, and ignore the properties of it's components; for example, set the shifters to 3.4 lbs + 2.75 lbs (the weight of a CIM plus the gearbox).

As for trusting parts that have been imported or given to you from others; just don't. If a goal of yours is to use your CAD model to provide accurate weight estimates, get in the habit of making sure all parts are set correctly as they are made and/or inserted into the assembly. It is a horrible experience trying to figure out later on why the robot is 20 lbs lighter/heavier than you think it should be and you have 1000+ parts in the assembly.

Thank you for this! I probably would have been killing myself come build season while working on CAD... Thank you very much. I'm going to get started on changing all the masses.

Originally Posted by ,4lex S.

A Box Drivetrain like this generally looks quite good on its own, but when you consider a real FIRST robot, some flaws pop up. From a structural standpoint, you need to ensure that you will be able to adapt this design to manipulators.

About 50% of the time my team had to take a chunk out of the front of the chassis to facilitate claws or collectors. Just keep this possibility in mind and adapt your design for potential geometry changes.

If you can't get it welded, this design should work quite well with rivets. Triangular corner gussets might be a good idea for some of the pieces, rather than just right angle aluminium.

This is really just my first step in designing a working FRC chassis. I would except part to get chopped up. If they didn't, then the GDC decided not to be evil... (Wouldn't that be the day. )

I have seen pictures of riveted Aluminum chassis so I figured I would be pretty safe in doing so. When I get far enough into the design process I will probably create a CAD with those in it. (Yes rivets and all. I may add wires... )

-Rion

[Akash Rastogi]

Originally Posted by ,4lex S.

A Box Drivetrain like this generally looks quite good on its own, but when you consider a real FIRST robot, some flaws pop up. From a structural standpoint, you need to ensure that you will be able to adapt this design to manipulators.

About 50% of the time my team had to take a chunk out of the front of the chassis to facilitate claws or collectors. Just keep this possibility in mind and adapt your design for potential geometry changes.

If you can't get it welded, this design should work quite well with rivets. Triangular corner gussets might be a good idea for some of the pieces, rather than just right angle aluminium.

That's pretty much a given when it comes to designing anything, you need to be able to adapt it to the circumstance. I'm sure Rion knows this already. It is the same with electrical layouts (hence why its funny when people complain about electronics placement in CAD).

[Andrew Schreiber]

Originally Posted by Formerly Famous

We are hoping to be able to weld it. Money and knowledge may hinder that though. If we can not then we will be creating brackets out of 1/16" Aluminum and riveting it together.

Consider contacting a local college. You could probably convince them to let you use their welding equipment (I would imagine most colleges have access to the equipment) Barring that, contact car repair places. Should be free and someone can teach you how/ do it themselves. Make the brackets anyway in case the welds break though.

[Brandon Holley]

Originally Posted by artdutra04

Turning down the ends of shafts on a lathe takes one tool.

Drilling and tapping the ends takes three: a center drill, the drill bit, then the tap.

On a manual lathe, it's faster to turn the ends down.

# of tools is kind of an unfair comparison here. Drilling the ends of a piece of stock is probably the simplest operation you can do on a manual lathe. No zeroing (assuming your going to just do a rough estimate of depth because your tapping the hole anyway) and no measuring of the piece to ensure correct diameter.

My guess is that the lathe operations would take close to the same amount of time for both kinds of axle, including the tool change for the drill. The tapping will cause the dead axle to take a longer amount of time, but we're talking about a matter of minutes which I do not feel is enough to justify changing a design from dead to live. There are a ton of other considerations obviously, but I just wanted to point that out.

-Brando

[IndySam]

Nice looking CAD.

We have done similar box chassis in the past and depending on the game we will more than likely do one again this year.. They are strong, simple and very reliable but heavy.

You can replace the shaft supports with simple aluminum blocks and make chain tensioning and changing wheel drop easier.

You can also loose the entire upper structure or wait and make one that is more game appropriate out of lighter stock.

We don't use welding anymore. Instead we glue everything together. Never had a failure.

[R.C.]

Originally Posted by Formerly Famous

Also. I have no idea what type of material to use for each part. I discovered the side plates are steel, past that I am lost. (RC, you need the GEN 1 in the CAD Library. )

-Rion

Will do Rion, Jeff and I have been working on getting more models up. We'll hopefully have that up before season.

-RC

[Tom Line]

We have used the same method for our "dead axles" on a couple bots.

We have an inner and outer rail and simply use a shoulder bolt as the axle. 1/16 plastic tubing keeps the wheels centered between the rails.

One tool and a couple turns and the axle can be removed and the wheel drops out.

It's simple and it's worked well for us, and best of all it requires a minimum of machining since all of our machine work is off site and donated.

[Gdeaver]

Team 104 made a box frame like this the last 2 years. The frame was made out of 8020 Quick Frame 1/16" box tubing. We also used the associated plastic connectors. In a couple of hours we cut and whacked our frame together. We had no failures or structural issues. We did use polycarbonate covering and other structures to tie everything together which is critical. We used dead axles. The axle supports were made from 8020 10 S 8611 double retainer profile . The axle supports clamps onto the frame allowing the chain to be tensioned. For the axle span 3/8" grade 8 bolts work well and save weight. The bottom board also is a structural part of the frame. A piece of 1/4" Baltic birch plywood would be a good choice. To go one step further, laminate the board with a layer of 5.7 OZ carbon cloth on both sides. This type of frame is good for low resource Teams.