Team 1511 Prototype Drivebase CAD

[Vespasian]

Hello All,

Team 1511 has been working on a summer drivebase project to help improve our competitive edge in upcoming years. These pictures show the current progress of the project, which is mostly in CAD at the moment. I have added captions just to clarify what each view entails.

We are using VexPro 3-CIM Ballshifters as our gearboxes. There are two CIMs on each side along with a Banebots RS-775 in a CIM-ile gearbox. The four pneumatic cylinders on the corners cause the bellypan assembly to drop down and greatly increase the amount of frictional force needed to move the base.


Isometric view of the whole drivebase from the front right corner.


Right side assembly. The outer wheels are using dead axles, the output shaft and center wheel are using live hex axles. The nylon cams are for chain tensioning.


Shows the top of the transmissions, the breaker, and the cylinder mounting plate.


A view from the bottom, shows the cool belly-pan lattice and rivet holes.


The separate bellypan assembly, electronics included free of charge.

Thank you for any feedback on the design work, your insight is greatly appreciated.

Sincerely,
Elliot Dowd

[Madison]

Do I understand this correctly? You're going to drop your electronics into the floor to resist being pushed?

[Electronica1]

I am just wondering why are you using flat metal plates rather than putting flanges to make it rigid? You use flanges in other parts but still use flat metal plates along the length of your frame.

Also, make sure you don't damage the field when you drop your "super brake".

[KrazyCarl92]

Is there a compelling reason to use a live axle on that center wheel? Dead axles can better help strengthen the sheet metal frame if used properly. Common reasons to use live axles on a center wheel in an application similar to this is to direct drive the center wheel from the gearbox or to run chain for a wheel on the opposite side of a frame member from the wheel. Neither of these criteria are met, so using a live axle on the center axle in this particular design is not taking advantage of the benefits of using live axles.

If the corner wheels are on dead axles, why are there bearings in the model? The axles do not rotate in a dead axle setup, so the axles should be in some way constrained to the frame. Most commonly bolts are used as dead axles and tightened into the frame.

Is there some kind of traction material that would be used on the bottom of that belly pan to keep it from damaging the carpet or provide extra traction? Depending on your material choice, a metal belly pan could violate the FRC rules (of course dependent on the rules for the upcoming season, if you plan to incorporate aspects of this design).

With the available motors from the past 2 seasons, it's difficult to justify an RS550 in a drive train application. CIMs and MiniCIMs have much more thermal mass and do not rely on the high speed of the motor to cool with an internal fan. This means that they are better suited for the constant use and sporadically near stall conditions that are seen in the drive train. Although RS550s have similar power to these other motors, they are better suited to most manipulator conditions where they are not subject to constant and heavy loads. With 6 CIMs and 4 MiniCIMs at your disposal, there are still many powerful motors left over even with 6 CIMs in your drive train. I'd expect at least the very least 4 CIMs and 2 MiniCIMs to be allowed by the rules, so even with all of those in the drive train you would still have a sizable selection of motors to use on manipulators.

EDIT: Why use the optional third stage of gearing on the ball shifter gearbox if the reduction could be achieved by sprocket and chain from your gearbox to your wheels? This could save weight with 1 less stage of gearing, less chain, and smaller sprockets with the same force on the chain.

[Tom Line]

Drive bases with omniwheels at each corner like that can be very hard to control at times. They tend to be difficult to drive straight.

What is the final speed of the drivetrain? That's a pretty large gear on the transmission output shaft.

We've had very little luck tensioning chains using the manner you've shown. The friction from the single fastener in the end of the cam shafts simply isn't enough to keep the cam stationary and they will end up spinning. Drivetrains generate a surprising amount of force on the chains.

It appears that the only thing holding the chassis 'square' are the 2 bolts at each junction between the side modules and the front crossbar. This will not be strong enough, and you will want gusseting. You will probably also want support across the middle of your chassis - that's why so many teams use a solid belly pan. It stiffens up the platform significantly.

[nicholsjj]

It's an interesting concept, but like Madison I would recommend not dropping your really expensive electronics on the ground that hard. Also be aware that you will need to prove that your robot's belly pan is fully in compliance with current rule r6, wherever it get put in next year's rules, meaning that you will need to add some type of rubber tread to the bottom of it. My suggestion would be to instead of dropping your entire belly pan that instead, design four posts that do the same function, or look at 118's and 1114's drive train's drop down omni-wheels as inspiration instead. I also know that 1730 did something similar as the two previously mentioned, but instead they used a ball caster like these http://www.mcmaster.com/#standard-ca...nsfers/=t2it6g.

[orangemoore]

I'm not sure if this was mentioned yet. But you are going to need to redesign the electronics board if you want to use this in the future. We would be using the RoboRio and not the Crio

[ThunderousPrime]

Originally Posted by orangemoore

I'm not sure if this was mentioned yet. But you are going to need to redesign the electronics board if you want to use this in the future. We would be using the RoboRio and not the Crio

Yup we are aware of this. Our idea behind this project is to get the experience we need going into next build season to build a similar drive train for next year. Additionally since we are a beta team we will be gaining experience with the RobotRio during the summer so it shouldn't be too difficult to change the electrical board next year.

[ThunderousPrime]

Originally Posted by Tom Line

Drive bases with omniwheels at each corner like that can be very hard to control at times. They tend to be difficult to drive straight.

In regards to omni wheels we have used them three times out of the four past years and our driver is more comfortable with them.

[pmangels17]

I like the concept, but I do have a few suggestions. You might be able to save some footprint on the battery mount by changing from the folded sheet metal to a few pieces of aluminum 1x2 angle (or something similar cut out of some 1x2 rectangular tube) or something similar. Also, I'd be very wary of having the battery on the drop down bellypan, that seems a bit sketchy. As previously mentioned you're probably better off designing a few drop down brake pads and making sure that they comply with the rules (though the 2015 rules haven't actually been announced yet). Not only would you not have to drop the battery, the bellypan would go a long way towards stiffening your frame. Best of luck with the beta testing, and nice job on the design head start, you'll definitely have something you really like by build season.

[Vespasian]

Originally Posted by nicholsjj

It's an interesting concept, but like Madison I would recommend not dropping your really expensive electronics on the ground that hard. Also be aware that you will need to prove that your robot's belly pan is fully in compliance with current rule r6, wherever it get put in next year's rules, meaning that you will need to add some type of rubber tread to the bottom of it. My suggestion would be to instead of dropping your entire belly pan that instead, design four posts that do the same function, or look at 118's and 1114's drive train's drop down omni-wheels as inspiration instead. I also know that 1730 did something similar as the two previously mentioned, but instead they used a ball caster like these http://www.mcmaster.com/#standard-ca...nsfers/=t2it6g.

One thing I did want to clarify is that this project is largely experimental. Some of the ideas (belly pan drop) may seem crazy or unnecessary, but that is why we are attempting them now. I really like the points that have been brought up, as I am new to drivebase design and really had no idea what I was doing until I started working on this project (and I still have a LOT to learn).

That being said, we plan on putting some blue nitrile tread on the bottom of the belly pan. This will help make the base more difficult to move while preventing carpet damage.

[Vespasian]

Originally Posted by pmangels17

I like the concept, but I do have a few suggestions. You might be able to save some footprint on the battery mount by changing from the folded sheet metal to a few pieces of aluminum 1x2 angle (or something similar cut out of some 1x2 rectangular tube) or something similar. Also, I'd be very wary of having the battery on the drop down bellypan, that seems a bit sketchy.

The battery holder was just a fun idea I wanted to try out, but it is very excessive, and may not even be realistic to make. I have another student working on a better system that involves a quick release pin to replace the current model. Basically it looked pretty in CAD, and that is why we have others to serve as my reality check

I like the description "a bit sketchy," in referring to the battery dropping, and it may even be an understatement. That is why we are trying it now, not during build season. The idea came from a mentor who recently left for California, and we decided to go with it at least for this project. Hopefully it will make lifting the wheels off the ground a possibility.

[Vespasian]

Originally Posted by KrazyCarl92

Is there a compelling reason to use a live axle on that center wheel? Dead axles can better help strengthen the sheet metal frame if used properly. Common reasons to use live axles on a center wheel in an application similar to this is to direct drive the center wheel from the gearbox or to run chain for a wheel on the opposite side of a frame member from the wheel. Neither of these criteria are met, so using a live axle on the center axle in this particular design is not taking advantage of the benefits of using live axles.

If the corner wheels are on dead axles, why are there bearings in the model? The axles do not rotate in a dead axle setup, so the axles should be in some way constrained to the frame. Most commonly bolts are used as dead axles and tightened into the frame.

With the available motors from the past 2 seasons, it's difficult to justify an RS550 in a drive train application. CIMs and MiniCIMs have much more thermal mass and do not rely on the high speed of the motor to cool with an internal fan. This means that they are better suited for the constant use and sporadically near stall conditions that are seen in the drive train. Although RS550s have similar power to these other motors, they are better suited to most manipulator conditions where they are not subject to constant and heavy loads. With 6 CIMs and 4 MiniCIMs at your disposal, there are still many powerful motors left over even with 6 CIMs in your drive train. I'd expect at least the very least 4 CIMs and 2 MiniCIMs to be allowed by the rules, so even with all of those in the drive train you would still have a sizable selection of motors to use on manipulators.

EDIT: Why use the optional third stage of gearing on the ball shifter gearbox if the reduction could be achieved by sprocket and chain from your gearbox to your wheels? This could save weight with 1 less stage of gearing, less chain, and smaller sprockets with the same force on the chain.

These are a lot of very good points, and much of it comes down to my inexperience. We may switch the live axle in the center to dead, since there is really no logical reason for it to be live (other than it is in CAD). I'm not sure it will hurt anything, other than adding a bit of complexity...?

About the bearings on the dead axles, those were originally live as well, and I didn't think to take the bearings out. The shafts are constrained using collars, is there a problem with doing this? It would add some weight, but for the purposes of this project, is this particularly detrimental?

The motors being used (other than two CIMs on each side) are RS-775's, not 550's. They are another sort of experimental idea, in case we don't have 6 CIM's to allocate to the drivebase next year. Larry (our team leader) brought this idea up, and we are at least going to try out the CIM-ile gearbox from VexPro with a 775 in it. If we don't like the results, then we will switch it to a CIM or Mini-CIM.

The third stage reduction was originally put in to achieve the proper ratio without needing a sprocket reduction. We have generally had a 1:1 ratio between our output shaft and wheels, and tried to stick with that. But it turns out the specific reduction we needed from VexPro was out of stock until at least September, so we bought a transmission with a larger reduction and compensated with sprockets. If that makes any sense.

I wish I could find the JVN gearing spreadsheet we used, but it is only on our school network at the moment. When I get in there on Monday I will upload it to Drive to show everyone. Our high gear theoretical speed came out to be around 17 ft/sec, which is a lot faster than our normal drivebase design (12 theoretical, 13.5 actual).

[Vespasian]

Originally Posted by Tom Line

It appears that the only thing holding the chassis 'square' are the 2 bolts at each junction between the side modules and the front crossbar. This will not be strong enough, and you will want gusseting. You will probably also want support across the middle of your chassis - that's why so many teams use a solid belly pan. It stiffens up the platform significantly.

One thing I forgot to add is that the outer four plates (the big ones) are made of steel. This was a design decision that we wanted to try out. Though I think you are correct in saying that we will want gusseting.

We are also planning on having bumpers that are entirely connected (one piece), and that are reinforced with aluminum. I don't know if anyone else has had good results with bumpers that also serve to reinforce their structure, or if that is just not a good idea.

[zinthorne]

I love the design and the ideas in it. Why not go with somewhat of a WCD style drive if you are not using sprockets for a final gear ratio? You could just gear directly to the center wheel and save yourself some weight. Also on the gear boxes I would definitely have multiple replacements. This last season we had one of our vex gearboxes crack and the motor almost fall all the way out of the transmission in finals of a competition. We also had the output shafts fall out several times during matches. (This could be fixed by glueing the shaft back in, but it was a huge pain)

Dropping the belly pan is an interesting idea, but I would be very cautious in doing that. First if you lost air pressure and put it down, you would be unable to lift it again leaving you useless for the rest of the match. Second if you get in a pushing match and drop it, you will be putting your robots weight and all the force of the other robot on 4 pistons. If you were rammed at a high speed while you had it dropped, I think its possible you could break the pistons or the supports. Third I do like the idea of dropping a separate piece down to act as a brake though.