pic: Madtown Robotics Drive Proto

[Cory]

Originally Posted by rc_cola1323

Sorry, I didn't mean steel. I was thinking about the aluminum gears.

You're making aluminum gears? I would think very hard about that before committing to a design that revolves around them. Travis has posted on here how there's no way we would have attempted the gearbox we did in 2007 if it hadn't been able to take care of his and Kirk's senior design project as well. There's hundreds of hours of machining alone put into them. The resources are really better spent elsewhere.

Regardless of the material, you need to do the math to see if 3/16" face width will work for you. It probably will for the initial reduction, but my gut instinct is there's no way it will hold up on the output.

[R.C.]

Cory I was thinking about this, what if we used a 3/16 face on the first reduction. Then moved up to 3/8 face for the rest, shouldn't that be fine. We are not going to go with an aluminum gearbox yet. In the offseason we will be attempting and testing one out, but for the season it will be steel. I just pointed out that we could anodize if we needed to harden up the aluminum gears(if we used it). I have already checked out your aluminum gearboxes and Cory, if we wanted to make an aluminum gearbox, what steps do you recommend or what is your opinion about creating an aluminum gearbox. We have the time and resources to make one over winter break. But not during the season it we don't make a proto, thanks Cory.

[AdamHeard]

Originally Posted by rc_cola1323

Cory I was thinking about this, what if we used a 3/16 face on the first reduction. Then moved up to 3/8 face for the rest, shouldn't that be fine. We are not going to go with an aluminum gearbox yet. In the offseason we will be attempting and testing one out, but for the season it will be steel. I just pointed out that we could anodize if we needed to harden up the aluminum gears(if we used it). I have already checked out your aluminum gearboxes and Cory, if we wanted to make an aluminum gearbox, what steps do you recommend or what is your opinion about creating an aluminum gearbox. We have the time and resources to make one over winter break. But not during the season it we don't make a proto, thanks Cory.

They spent hundreds of hours. Hundreds. If you worked nonstop for a week, that's 168 hours.

You know what your best bet is if you want a nice, light, simple to make gearbox? Make custom sideplates for the AM supershifters, pocket them like you did, and use standoffs rather than the extrusion casing, like you did. It'll be real fast to make, and you *know* it will work and be very reliable. It may not technically be the lightest gearbox out there, but it will be done fast and done right. You could probably drop the 4.6 lb super shifter to at least 3.5, probably closer to 3 if you lighten the gears (but do not, do not reduce the face width unless you really know what you're doing). A 3-3.5 pound, 2-speed, bulletproof gearbox is still pretty light overall.

This is our exact plan on 973. We technically have the resources to make lighter gearboxes (from Steel gears, we won't dabble in aluminum without some serious effort and planning), but it's not worth the additional time and risk in our opinion.

[R.C.]

Originally Posted by AdamHeard

They spent hundreds of hours. Hundreds. If you worked nonstop for a week, that's 168 hours.

You know what your best bet is if you want a nice, light, simple to make gearbox? Make custom sideplates for the AM supershifters, pocket them like you did, and use standoffs rather than the extrusion casing, like you did. It'll be real fast to make, and you *know* it will work and be very reliable. It may not technically be the lightest gearbox out there, but it will be done fast and done right. You could probably drop the 4.6 lb super shifter to at least 3.5, probably closer to 3 if you lighten the gears (but do not, do not reduce the face width unless you really know what you're doing). A 3-3.5 pound, 2-speed, bulletproof gearbox is still pretty light overall.

This is our exact plan on 973. We technically have the resources to make lighter gearboxes (from Steel gears, we won't dabble in aluminum without some serious effort and planning), but it's not worth the additional time and risk in our opinion.

I totally agree, but we will be facing down steel gears b/c

a. we have the resources
b. engineers are gonna help us with details
c. a great learning experience for the team

Oooh, CAD! Shiny! If criticism is welcome, then I offer up the following. I ensure you that my intentions are not to denigrate your design, but instead to offer some criticism. PM me if anything here doesn't make sense, I'd love to explain it to you over PM, without the pressure of an open channel.

First off: Cheeseholing your drive frame. Why? The weight saving there is minimal, and the time it would take to machine those parts could be better spent doing something with more impact, such as more time in design to make sure your robot will reach maximum effectiveness during competition. Instead of spending time hobbing the structure of your robot away for those precious few pounds, I suggest spending the time doing some FEA on your chassis to make sure it can withstand the forces of the game, while still using the minimal materials possible. I'm sure you can chop the weight down without needing to spend an enormous amount of time machining key structural components.

Second: Gearbox mounting. It makes me nervous having only two bolts holding something that critical to an already weakened member. If you move the gearbox down, so more of the sideplate face is flush against the chassis member. This allows the bolts to take less of the load when the gearbox is under stress.

Third: Gearbox sideplates. Yes, milling out enormous pockets on the gearbox plates look cool. Yes, it provides bragging points to other teams. Is it necessary? Not if the design is done well enough. If your goal is to lose weight, look at designing your side plates around a different material. I highly recommend Delrin or UHMW. You will (most likely) save machining time as well as cost and weight.

Fourth: Gears. Aluminum gears, unless they're pre-fabricated, are not worth the work. Your team MIGHT save some weight by going with this option, but is the insane cost and manufacturing time worth the weight? Instead of going with exotic materials, it might be a better design decision to go with something strong and proven.

I might be designing a little to "conservatively," but I've yet to have a gearbox failure by working with these principles. As a great mentor taught my old team, the most important question is not what can we build with this, but Why are we building this with that?

I really like the design you have going for the chassis, assuming you make a few tweaks. You've got a cool and innovative "crumple point" in the form of those rounded bumper plates. You're also going to get a lot of strength out of the triangles you've created in the frame. However, you might gain even more if you design in a chassis member across the front of the frame, and tweak the superstructure to suit.

Good work on the CAD, it's a valuable skill. Never stop designing things!

[R.C.]

Craig, Thanks a lot. That advice helped a lot. We did run FEA on the frame when I finished the design. I had more pocketed out way too much, including the top. The engineer told me that was way too much and it crumbled under the FEA tests. So that is why we have the bracing setup it like that and you can see that the top of the tubing is not pocketed out.

The gearbox won't be mounted like that, last year we mounted the gearbox with more support. I was hoping to add more support when I included the mast of the forklift or Arm. I had a question about the plates, if i use delrin. How thick should it be and do i leave everything solid except the holes for the bearings? (I will be moving the bolts down)

My purpose was to lose weight, I will redesign them after i get feedback on the thickness of those plates.

The gears will be prefabricated, http://www.chiefdelphi.com/forums/sh...threadid=70259. This sponsor would do it for us. But we are thinking that it would be better to use steel and use an index to make some holes.

Craig thanks for all the suggestions. I am going to be redesigning to your suggestions. If you have any more suggestions, please post it. The reason for shaving so much weight is. Every pound matters to our team, we always add stuff and every year and bearly make the weight limit. But that was because our stuff was way too solid.

Originally Posted by rc_cola1323

Craig, Thanks a lot. That advice helped a lot. We did run FEA on the frame when I finished the design. I had more pocketed out way too much, including the top. The engineer told me that was way too much and it crumbled under the FEA tests. So that is why we have the bracing setup it like that and you can see that the top of the tubing is not pocketed out.

Cool. It's good to see that you've been doing the right calculations. Many teams skip over this and end up with some nicely bent frames that would look quite nice in a modern art museum, but won't function for robotics. My motto has always been that my bot should be able to be dropped wheels down off a roof, and still be able to drive itself into the shop to have the flimsy manipulator fixed from the damage. Driving is everything in FIRST.

Originally Posted by rc_cola1323

The gearbox won't be mounted like that, last year we mounted the gearbox with more support. I was hoping to add more support when I included the mast of the forklift or Arm. I had a question about the plates, if i use delrin. How thick should it be and do i leave everything solid except the holes for the bearings? (I will be moving the bolts down)

My purpose was to lose weight, I will redesign them after i get feedback on the thickness of those plates.

Delrin is awesome. We used the 1/4" thick from McMaster (I don't have the part number on me, let me get back to you on that), and didn't have any problems with strength. The material is so light, cheeseholing the plates isn't even necessary. Just make sure you put washers on EVERY bolt and nut that goes into the stuff. It's also nice to machine, as CNC's cut the stuff like butter. If you think Aluminum is nice to cut, wait till you see delrin get cut. It has a tendency to load the smaller bits, but you can work around that.

Originally Posted by rc_cola1323

The gears will be prefabricated, http://www.chiefdelphi.com/forums/sh...threadid=70259. This sponsor would do it for us. But we are thinking that it would be better to use steel and use an index to make some holes.

Craig thanks for all the suggestions. I am going to be redesigning to your suggestions. If you have any more suggestions, please post it. The reason for shaving so much weight is. Every pound matters to our team, we always add stuff and every year and bearly make the weight limit. But that was because our stuff was way too solid.

Ah, I didn't make the connection between threads.. Well, if the gears are built for you, then go for it.

[roboticWanderor]

Hey there!
Not sure what is gonna go on top of the robot and whether it will provide any extra support, but there does not seem to be much keeping the two halves of the drive train parallel. putting a triangular support on the upper back corners, or bridging the gap about halfway up the chassis would do a lot to keep the two halves stuck together.
If you did that, you could reduce the beefy super structure you have and save more weight than cheese-holes ever could.
Someone else already mentioned my other concerns, good luck!

PS: nice to see you back Craig!

[R.C.]

Originally Posted by roboticWanderor

Hey there!
Not sure what is gonna go on top of the robot and whether it will provide any extra support, but there does not seem to be much keeping the two halves of the drive train parallel. putting a triangular support on the upper back corners, or bridging the gap about halfway up the chassis would do a lot to keep the two halves stuck together.
If you did that, you could reduce the beefy super structure you have and save more weight than cheese-holes ever could.
Someone else already mentioned my other concerns, good luck!

Yeah those were going (bridging the gap) to be added as soon as the rest of the bot was decided upon. We will probably do three to four beams across and one across the top of it.

[AdamHeard]

You may have the resources, but keep in mind there is a lot more to a robot than a base.

If you can make the base that performs the same for 40 units of weight, and 10 units of work as one that is 30 units of weight and 100 units of work, the first is probably a better option. It looks like with this design concept, you can get a competitively light and strong base in short amount of time, don't try to go to far and shoot yourself in the foot. Use that nice start to get a decent base done for sure, which will free up machining resources and allow assembly and test to begin sooner.