16-05-2015 01:05

asid61

Re: pic: Offseason Gearbox

Looks very clean.
I like the use of thunderhex and circlips to retain shafts and use round bearings, but for the amount of manufacturing this requires you may want to use 1/2" bearings and shafts with ends rounded. It's a pretty simple lathe job that IME is comparable to an e-clip slot.
You could probably get away with only 4 standoffs, but for mounting purposes you could keep it like it is I guess.

I noticed on the plates your radii are pretty variable, with a 0.04 radius in one of the plates. Depending on the machine, this could make it take considerably longer to machine than using a single larger radius throughout, as a larger endmill can take off more material in general in a certain amount of time.

Overally a pretty sweet gearbox.
It looks like it fits very well into the back of your WCD.

16-05-2015 01:15

R.C.

Re: pic: Offseason Gearbox

The better way to handle this is to use 3/8" Hex shaft and flip the bearing so the flanges are on the inside that way the bearing is geometrically retained.

Also if possible round down the shafts if possible so you can use a cheaper bearing etc...

16-05-2015 01:20

mman1506

Re: pic: Offseason Gearbox

I think your being a little picky. CAM software will try to cut the tightest radius it can with tool specified. You should be able to cut 90% of the lightening holes with a .25 end-mill and then finish things up with a 1/8 end-mill. All in all it should take 8-15 minutes a plate with decent end-mills.

16-05-2015 01:27

asid61

Re: pic: Offseason Gearbox

Wait really? I was told I had to put all the radii in manually even for a CNC.
I've been lied to... lol.

16-05-2015 02:05

sanddrag

Re: pic: Offseason Gearbox

You generally want to design the pocket radii some amount larger than the radius of the tool you intend to machine it with. This is to prevent the tool from encountering a sudden and peaking chip load when it gets into the tight corner. It needs some room to still move in the corner, so it isn't contacting all around at the same time.

16-05-2015 02:06

Cory

Re: pic: Offseason Gearbox

If you don't enter the radii you will end up with sharp corners in your toolpath or with finishing passes that take out huge chunks in the corner, not the (for example) .015" you may be running around the rest of the chain. You'll get a radius there, but it will be harder on your tool than desirable and will leave a poor surface finish.

It's always best to make your radii slightly larger than the cutter you intend to use, so that you can roll the cutter through the corners instead of coming in and making a sharp line entry/exit. We usually use .010-.015 more on the radius to help with this.

16-05-2015 02:20

asid61

Re: pic: Offseason Gearbox

Can CAM do this for you or do you have to put in all the radii manually?
What size endmill would you normally use as a minimum for something like this?

16-05-2015 02:43

MStump

Re: pic: Offseason Gearbox

I agree, using 3/8 with rounded ends is pretty nice, simple, and light. You can also use the smaller 7/8 OD bearings.

All in all it looks really clean and nice. I came up with a similar design at the beginning of the season to mount our gearbox at the back of the frame but it did not end up looking nearly this nice haha.

My suggestions for minor improvement:
1. Use the 3/8 shaft mentioned previously for the first stage reduction.
2. Take advantage of the upper bolt on the lower cim for a standoff instead of having another hole solely for a standoff.
3. I wasn't able to download the model but make sure you have straight access to all your bolts on your CIMs. It'll make maintenance a lot easier but I'm sure you already thought of that.
4. Add additional mounting holes and bolts so that it isn't completely resting on the output shaft. (There might be something I'm missing but I can't see where else it would directly mount to the frame.
5. Also just an idea to consider, having only one plate that is repeated on both sides of the gearbox saves some time cadding, machine programming, and machining. However, it also adds a little bit of weight. So if you have the time making two plates is probably better, but if you are in a rush it may be better to forgo the second plate.

16-05-2015 03:03

Cory

Re: pic: Offseason Gearbox

You should just use the fillet tool in solidworks/inventor/whatever and add .010+ to the nominal radius when you create the fillets.

I dislike using anything less than 1/4" diameter in 1/4" plate. with 1/4" cutter diameter you can go full depth (with a reasonably rigid machine and good chip evacuation). 3/16" and smaller won't allow you to do that.

16-05-2015 21:38

Dylan179

Re: pic: Offseason Gearbox

Is the bering on the output shaft pushed out some?

17-05-2015 01:34

rponmalai

Re: pic: Offseason Gearbox

Yeah the flange is actually on the inside of the tubing and that gap is for the thickness of the tubing.

17-05-2015 02:10

asid61

Re: pic: Offseason Gearbox

How will you get it into the tubing?

17-05-2015 02:12

Joey Milia

Re: pic: Offseason Gearbox

And once you do how will you get it to stay in the wall of the tubing?

17-05-2015 20:25

rponmalai

Re: pic: Offseason Gearbox

http://imgur.com/a/n6cZ8
One of the things I was playing around with in this design was running the timing belts through the tubing of the drive base. I have an access hole for putting the belts through that can easily fit the bearing. The entire shaft would be held together with snap rings. The grooves for those snap rings are on either end of the output shaft.

17-05-2015 20:36

Joey Milia

Re: pic: Offseason Gearbox

That answers the bearing questions. What's the belt set up you were thinking about? Pulley tooth count and belt size?

17-05-2015 21:19

rponmalai

Re: pic: Offseason Gearbox

I have a Vex 18T pulley for 5mm HTD timing belts. The pulley thickness is for a 15mm belt.

17-05-2015 21:47

R.C.

Re: pic: Offseason Gearbox

I would recommend at least 24t or larger if your using HTD.

17-05-2015 22:52

Chris is me

Re: pic: Offseason Gearbox

You really can't get away with less than 24T 15mm wide on a drive belt. 24T belt-in-tube generally works best in a 3" tall drive tube, though it could probably be made to work in a 2.5" tall tube as well. So going belt in tube generally costs you a little bit of weight.

When doing a gearbox in the back like this rather than the middle, keep in mind that belt connecting the back wheel to the middle wheel will always see the full load of the drive base. So it's extra important to use a properly sized pulley.

17-05-2015 23:07

rponmalai

Re: pic: Offseason Gearbox

Am I correct in assuming that you need a larger pulley to keep the belt from slipping off of the pulley?
The more I hear about this I am starting to lean towards chain drive to run through the tubes. My only draw back is keeping the chain tensioned. What would be the best way to keep the chain tight while it is in the tube?

17-05-2015 23:25

Abhishek R

Re: pic: Offseason Gearbox

If it's nice and tight properly upon installation, you shouldn't need any form of tensioning for the chains.

17-05-2015 23:56

Knufire

Re: pic: Offseason Gearbox

FRC runs belt some belt very close to it's max load. Both a wider belt and a bigger pulley reduce load on the belt.

If you have the proper center-to-center distance on the chain, it should stay tensioned. 5188 did the calculated center-to-center + 0.018" this past year, and the chain stayed perfectly tensioned all season. This was based on Paul Copioli's recommendation and further coborrated by the testing done in this thread. However, with the chain in tube, remember that a tigher chain is more difficult to install, and there isn't anywhere for the chain to go, so the added distance may not be necessary.

18-05-2015 00:17

Chris is me

Re: pic: Offseason Gearbox

Slipping off, no, ratcheting or failing, yes.

That said, you never have to tension a properly dead spaced belt in tube drive. It has its pluses. It's just not

Maybe in certain applications depending on load, chain length, etc, but this just isn't a blanket statement you can make. Both chain stretch and sprocket wear are very real phenomena that shouldn't be ignored. The 118 solution does an unconventional but effective job at containing this chain stretch to prevent failure, but in most cases you're going to want a tensioner. Optionally, you could replace the chain once it's stretched to the point of giving you problems.

For a very extreme example, our chain elevator this year had a nice 13 foot long loop that was dead spaced, and it stretched noticeably over the course of a few hours of low to no load testing. It is a function of sprocket diameter, chain length, and several other factors. At the size of sprockets used in west coast drives, it's a factor.

The other important caveat here is that your chain and sprocket should be constrained by the tube itself for this to work (very close to inner walls of tubing). One of the reasons chain in tube seems to work so well is that there is nowhere for the chain to go when it does stretch, preventing the jumping problems found with looser chains.

18-05-2015 00:42

MichaelBick

Re: pic: Offseason Gearbox

We found out this year that Coast Aluminum in California actually has 2" x 2" x 1/16" tube (its not in their catalog) which is really useful material. One of the things I have been considering is doing a belt in tube drive with this stock, which compared to the 2" x 1" x 1/8" tube we use right now in our WCD we'd actually be taking a weight savings.

18-05-2015 01:03

Abhishek R

Re: pic: Offseason Gearbox

Yeah, that's pretty much a requirement too, thanks for clarifying. There's literally nowhere for the chain to go, so as long as it doesn't break you're good.

18-05-2015 19:38

rponmalai

Re: pic: Offseason Gearbox

Is 1/16th strong enough for a drivebase?

18-05-2015 22:09

MichaelBick

Re: pic: Offseason Gearbox

I know people do it all the time with sheet construction, but they may also be running taller drive plates. Sheet drives however also have to trade off material strength because 5052 is weaker than 6061. Its going to be something we have to test out