I have been working on this drive train on and off for this past year. I have gotten to the point where I think this drivetrain is ready to go into its manufacturing stage. But before I do so, I would like the expert advice of all of you.
What should the minimum distance be from the ground to the bottom of the gearbox?
Is there a better/ more efficient way of designing this battery holder? Right now, it is comprised of Sheet metal, and I would like to find an alternative method using regular 1/8" Aluminum Plate.
As of now, all the drive shafts are 2024 Aluminum. I chose this to significantly decrease weight, but I need to ensure that it does not compromise durability. (It would never be a good thing to have the drive break on the middle of the field! What are your thoughts about this? I could always revert it back to a steel alloy.)
Thanks, and I very much look forward to hearing everybody’s advice/suggestions!
In the past, my team has made our battery holder out of 90 degree angled aluminum. You can see it if you look closely at the back of the bot in this picture.
For the width of the gearbox, those standoff/spacers don’t look substantial enough. What gears are you using specifically?
Looks like you used 0.25 plate for the bearing blocks, I’m guessing similar to 33 from 2011 style. There doesn’t look like there’s enough overlap between the 2x1 and the plate. I’d avoid running the chain below the chassis if you can help it.
The battery box looks like it’s sitting on the garolite unsupported. Looks like an unbendable part as is.
At a glance, doesn’t look like there’s enough spacing for wires since you have the Vics, PD, Vics, cRio side by side.
We have access to a CNC Milling Machine, Welding equipment, and a waterjet. Gotta love sponsors!
We are solely using WCProducts’ Gears. We have a 14 T CIM gear going to a 45. From there, the Low gear is 15 T to 60 T. The High Gear is 30 T to 45 T.
Yes, the part that is visible is a 1/4" Plate. There is .25" of overlap between the 2x1 and the plate. Is that enough?
How would I get away with out running the chain below the chasis?
That is true, I will fix it in the next version of the battery holder.
I essentially copied 254 and the Holy Cow’s electronics setup. We are not going to be predrilling holes into the bellypan, so if this setup does not work, we can find another way of mounting.
I would definitely love to use 7075 Aluminum drive shafts. Do you know where I can get 1/2" 7075 Aluminum Hex?
Worst case scenario, I could always have our machinists use a 4th-axis on the Mill to get that 1/2" Hex. (It’s just always nicer if we can buy it premilled in the Hex configuration)
I’ve been talking with Adam Heard, and he said that it’d be best to leave the bellypan as is for now until we have had some first-hand experience with professional wiring. Our team does not yet the minimum spacing between many of the electronic components. A fully pocketed Aluminum Bellypan is definitely something we will incorporate into a future revision of the drive train.
A sponsor isn’t an unlimited resource. We’ve run both garolite and waterjetted pans on multiple robots. I honestly couldn’t say one performs better (the waterjet is stronger, and is certainly cooler. Can’t say that extra strength is truly necessary tough). I know for sure that the waterjetted pan is 10x the resource and cost investment at least.
So a waterjetted bellypan is a huge resource dump for minimal competitive gain, and if that same amount of sponsor time was invested in making parts for other system it would be far more useful (a waterjetted bellypan easily equates to the gearbox for every system on the robot in most cases, plus several sets of spares).
To elaborate more on the differences. G-10/fr4 is what we use, which is primo grade garolite. Mcmaster has it in 1/16" which is what we routinely order. It’s hard to directly compare a solid composite pan to a pocketed aluminum pan in strength, but the more I think the more I’m pretty sure they are comparable in strength. We’ve run both on multiple projects, and the garolite is actually stiffer with less sag/deflection under load.
The 7075 from OnlineMetals, I found usually comes out to be around 0.500 - 0.503. Usually a pretty good fit on a broached part.
Sorry, I meant the bellypan.
Adam, we used our bellypan to fixture the chassis before welding, what’s your experience with using garolite to do the same thing? (Or do you do something different altogether?)
The overlap on the bearing plates, I’d look at doing more, going to more of rectangle shape if you have to instead of the teardrop shape. Although, doesn’t hurt to try it as is.
We’ve never tried it. The G-10/fr4 is “fire retardent” so it might handle heat at a reasonable distance, but certainly shouldn’t get too close to the weld.
We’ve never had it on during welding, when we welded our chassis we fixtured with other means. Since then we’ve just not welded the chassis. If we did again, we’d probably fixture with disposable gussets.
I have heard a lot about using the bellypan to fixture the chasis before welding. Can you please explain more about how you perform this task?
(I am still relatively inexperienced when it comes to the complete fabrication of such a drive train)
I’d prefer the teardrop shape, as it provides a nice and easy point of contact for the CAM. However, I will look into making the circular part wider to allow for more contact with the 2x1 frame.
I would recommend using steal axles. Some may disagree but I have had personnel experience where an axle has completely bent because we got hit on the side of our robot. Axles aren’t always the easiest thing to replace, and in a competition environment you won’t want to have to change them, especially during the Elimination rounds.
How large are the slots in the frame for the bearing blocks to slide in?
In 2011, we used 1/2" slots with 1/2" keyed axles and 1/2" OD spacers between the bearing blocks. Once everything was tensioned, we ran it in and the axles opened their own holes in the slots and everything was great.
We had no issues with 7075 aluminum axles, but did preventatively replace two because they keyways were opening (after two tournaments and significant pre-ship practice and development). We also ran 7075 axles this year, including many 1/2" dead axles and two 1/2" keyed primary drive axles (which drove the one live-axled wheel directly, and all of the others via chain).
They look pretty much identical to ours, which have never had issues.
2024 is definitely not good enough for your axles. Go to 7075, or stick with steel.
the bellypan is used for fixturing by riveting it to all of the frame members, which holds them in place as the frame is welded. This results in a better aligned, more square frame than if everything has to be tacked together. It also saves hours of your welder’s time. We could never go back to the way we were doing it before at this point.
How is the belly pan secured to the frame? It appears to be sitting in between the drive chain loop. Most drivetrains run the belly pan riveted or screwed to the very bottom of the drivetrain, to the 2x1’s. However, in your case, it does look like you need to change the wheel sprockets in order to switch to a bottom-mounted belly pan.
The output side of the Victors on the right are being blocked by the cRIO. If you’re going to keep it that way, invest in some 90-degree terminal connectors?
Also, the placement of the power switch in relation to the PDB’s battery terminals and the area for fastening the Anderson connector on your battery box is a little tight and normal 6AWG wire may not be able to loop around the area that well. However, you should be able to pull it off with some welding cable.
Spike and Air Tank for the Compressor
Finally, your battery box doesn’t necessarily have to be that high. It also looks a little tight. I would say do a little prototyping. Only the corners of the box are “extremely” important, the extra material on the “faces” of the battery box may or may not be necessary.
Also try and not to run the chain underneath the belly pan. You can space up the 1x1 so its closer to the middle of the 2x1 by using 1/2" length spacers with a .375" OD and .196" ID.
How do you attach your bumpers from front to back?
The battery box could get centered, If your waterjetting. You can make a ring similar to what 973/1323 did in 2011. Its pretty nice. We did the same thing for 2012, I forget what 973 did.
Putting the compressor between the gearboxes is semi awful, as when taking the gearboxes out its gonna be a pain.
Why leave the entire front of the DT open? Is this a proto WCD? If so why not fill it up entirely?
We happened to approach the battery box in a similar fashion for 2012. It was just a ring setup but recessed about 3 inches below the bottom of the frame. The battery rested on a 1/16 plate mounted by standoffs.
Try to keep the wheels as close to the perimeter. If the game involves a ramp of some kind the wheels will come off the ground when the front or back tube hits the ramp. You might try running the front/back tube at a 45 deg angle.
If you can bend flanges into your G10 belly plate it would strengthen the pan considerably and give you a flange to rivet the plate into the side tubes.
And remember to keep enough ground clearance on the outside wheels even with the chassis tipped to the front or back. We forgot about that this past year and couldn’t cross the bump because of it.
