4488 Off-season Gearbox

[Thad House]

Total cost not including the metal for the plates, is $101.8 all from Vex. Thats enough for 2 gearboxes. The only thing that doesnt include is the sprockets, which the WCP doesnt include either so its moot.

As for machining, at our school we have 9 lathes, 3 mills, a CNC, and students who know how to use all of them. During week 1, we have kids that are much more interested in machining then designing, so we can make these during down time, when we don't have a full robot designed yet.

I'm not writing them out, and if we run into trouble that will be option 1. But I think we could do it more efficiently, cheaper and more focused to our needs by doing it manually. Plus it will keep the students occupied week 1.

Also what 4 inch wheels actually fit the 1.25 inch output shaft? I'd love to know, so we could make ours even shorter if needed.

[MichaelBick]

The gearbox is designed to fit their colson hubs. Colsons come in a lot of sizes. Here is a 5" one that as far as I know will fit their hubs: http://www.robotmarketplace.com/products/BPDWC14.html

Also if you want to use wider wheels what we did was tap the end of the shaft and put a spacer(we put hex, can be round) to get the extra length needed.

Keep in mind that the WCP product is an end all solution. It just works. If you design your own you will need to prototype(you will not get it right otherwise). Also you will have to make your own replacements if parts fail(WCP takes care of you on this). Lastly the time can just be better used elsewhere. Sure people may not like designing, but prototyping is very important. If you don't want to buy premade bearing blocks(WCP sells COTS ones too) than those need to be amchined for a WCD. Drive rails and shafts are more things that can be machined week 1.

[craigboez]

This has been a recurring theme of commenters in several threads: "Why would you build X when you can just buy it? It will save time, money, be more reliable, you can spend time doing more productive things, etc." I think we can all agree that there are very good reasons for buying stuff off the shelf, and we're all aware of the upside in doing so. Is it really necessary to first point this out before helping answer a few good questions?

The goal of all FRC teams is not to build the best robot as efficiently as possible. Many teams try to learn, do new things, and knowingly do things the hard way.

Just think if some 6th grader would have given you a calculator in 3rd grade and said "Here, use this instead of doing that long division stuff. It's much faster and you'll make fewer mistakes".

Originally Posted by craigboez

This has been a recurring theme of commenters in several threads: "Why would you build X when you can just buy it? It will save time, money, be more reliable, you can spend time doing more productive things, etc." I think we can all agree that there are very good reasons for buying stuff off the shelf, and we're all aware of the upside in doing so. Is it really necessary to first point this out before helping answer a few good questions?

The goal of all FRC teams is not to build the best robot as efficiently as possible. Many teams try to learn, do new things, and knowingly do things the hard way.

Just think if some 6th grader would have given you a calculator in 3rd grade and said "Here, use this instead of doing that long division stuff. It's much faster and you'll make fewer mistakes".

That's what the offseason is for. It's great if you've tested an idea in the offseason, but very rarely will a team be able to implement a new custom design in the season and be more successful than a COTS alternative.

tl;dr: Learn in the offseason, apply in the build season.

[Thad House]

Originally Posted by Andrew Lawrence

That's what the offseason is for. OP is suggesting using a new idea in the build season. It's great if you've tested said idea in the offseason, but very rarely will a team be able to implement a new custom design in the season and be more successful than a COTS alternative.

tl;dr: Learn in the offseason, apply in the build season.

We are actually building them right now. The plates are currently at the waterjet shop being made as I type this. We also plan on running this test robot at the offseason events this fall. We are prototyping, and if it fails we will look elsewhere. Thats why i was asking for suggestions on the actual design itself, not if we should use it.

Originally Posted by sst.thad

We are actually building them right now. The plates are currently at the waterjet shop being made as I type this. We also plan on running this test robot at the offseason events this fall. We are prototyping, and if it fails we will look elsewhere. Thats why i was asking for suggestions on the actual design itself, not if we should use it.

Oh, in that case carry on - my mistake.

However like everyone else says, you're building the WCP transmissions. It would be a lot better learning experience for your students to develop something entirely new that fits your needs. All of my team's transmissions have been COTS, and will continue to be COTS in the future, however I've definitely tried developing my own gearboxes in CAD that were unlike any I'd ever seen, and that has been the greatest learning experience to me.

[Thad House]

Yeah. We are starting with this so we know that we can actually do the design and manufacturing. If we know machining our own works, we can customize them more when we actually need them to be custom. But for now, going with what we've seen work on many robots seems like a great starting point, then add more later if needed. In addition, I started working on these gearboxes before WCP even existed, but at the time we didnt have the machining resources, so now that we do I want to see what our kids can do.

[magnets]

Here's my thought about buying vs building. Yes, buying would be simpler/easier for this gearbox, but when it's time for the team to build a gearbox with a PTO or something more complex, they'll have the experience and knowledge required to build it.

[MichaelBick]

Nobody is arguing that building is not a good experience. We're just trying to point out that COTS helps teams use less resources which means they can use their resources better which means they can build a better robot. Even teams who work extreme hours with extreme resources can benefit(hint hint 254. Their drive gearboxes used quite a bit WCP parts from what I heard and vex gears replaced the custom ones they used to make. COTS is allowing them to build better/faster.)

Today I received an email from a sponsor that happened to pertain to this. This is from a sponsor with a lot of resources machines(This picture accurately depicts their facility. This is only a small part of the entire facility). This exerpt regards the use of prefabbed parts in industry:

#1 – Something anyone in engineering needs to keep in mind at all times is a tried and true saying: “Don’t reinvent the wheel”. Even though it is easy to understand, it is very easy to fall victim to. If you are indeed going to all of the work to design something, before you start, make sure it doesn’t already exist elsewhere (this would apply to levels from a single component, all the way to the end product). Even if you end up going through with a new design, knowing what is out on the market already will help you do it better the first time around. Additionally, if it doesn’t already exist, there may be a very good reason for it.

#2 – Whatever you make, use as many off the shelf components as possible. This not only helps with development costs & lead times, it also provides a much easier ability to get repair parts down the road significantly faster & cheaper. Also, if you need to recreate something custom later on, ideally you will have the respective print for it too. Without that print, then you have to re-engineer a custom part, which may not be as successful as the initial one, and will cost even more.

Back on topic, I don't think that water jetting is a good process for a part like this. Drive gearboxes are important parts of your robot, that get a lot of use, so tolerances are important. You will not be able to achieve the required tolerances with a waterjetting machine. All of the 1/4" Al gearboxes I have seen have been done on CNC machines. Also, we use .1875" thickness outside of our bearings(.625" radius for .875" OD bearings, .75" radius for 1.125" OD bearings).

[Seth Mallory]

Buying a gearbox is usually cheaper, quicker, and performs as well or better then building your own. GRT has probably designed and built more drive gearbox's then almost any team in FIRST. We like the engineering design and build project for the students. The students have to figure what gears to use and design the shafts and the shifting. They have to design a new gear box and not just side plates to use with someone else's gearbox. Not using some else's gearbox parts meant when our custom shafts, 8 of them, broke in 2008 we almost quit running after the 3rd match. We finished the regional with our last 2 shafts on half power. Buying gearbox's is probably the smart way for competing in FIRST but we are building a really neat gearbox for this year.