Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Robot Showcase (http://www.chiefdelphi.com/forums/forumdisplay.php?f=58)
-   -   Be afraid... Be VERY Afraid (http://www.chiefdelphi.com/forums/showthread.php?t=109671)

Madison 26-11-2012 14:08

Re: Be afraid... Be VERY Afraid
 
These threads are interesting -- and we don't see many of them around anymore, unfortunately -- because they document someone's effort to make a better mousetrap. What becomes clear as the discussion progresses is that there are still lots of folks that still have a lot to learn about the forces at work and considerations we face while making decisions about how a mechanism will work. I might've appreciated this a bit more if there were not 6+ pages of attention-seeking.

As shown, this almost absolutely will not work. It appears to offer no advantage over swerve drive but has several disadvantages that swerve does not suffer. I'd like to see some math justifying ANY of the claims y'all have made re: friction, pneumatic cylinder size, etc.

Ryan Dognaux 26-11-2012 14:41

Re: Be afraid... Be VERY Afraid
 
Quote:

Originally Posted by bhumudar (Post 1196104)
Lets suppose you see an incoming robot front the right of our robot, we would turn the wheels so that they face front to back (perpendicular to the incoming force).

Maybe I'm not thinking clearly, but why would you ever want to do this with the drive module you're intending to build? If anything with this drive you would want to avoid robot to robot contact if at all possible. I just see this probably losing any pushing matches with a standard 4 CIM 6WD setup, but would love to be proven wrong.

sanddrag 26-11-2012 14:41

Re: Be afraid... Be VERY Afraid
 
I'm curious what particular product or material he intends to use for the ball.

You might be interested in this: http://www.autoblog.com/2012/09/19/s...eering-studen/

BJC 26-11-2012 22:05

Re: Be afraid... Be VERY Afraid
 
Hey there,

I see that this idea/module was pretty much completely picked apart and shown to be a horrible idea so I won't bother with telling you any of that. Hopefully, you’re not too disappointed that the module/idea that you probably spent a long time designing isn't as great as you thought it was.

I just wanted to take a second to encourage you to continue thinking, inventing, and innovating. The people on this forum just want your team to be successful. Posting your designs on the forum is a gutsy move as any new design will be dismembered by many smart people. It is also one of the best things you can do as you get free constructive criticism from a bunch of smart people that have collective experience beyond what you could ever hope to gather by yourself.

Anyways, I hope you can take this away as a positive experience. If you hadn't posted on here and instead just went ahead and built your drivetrain for the 2013 season you would almost certainly be fighting your design the entire season and had a not-fun-at-all experience as a result.

Good Luck in your 2013 season and your future design endeavors.

Kind Regards, Bryan

BrendanB 26-11-2012 22:31

Re: Be afraid... Be VERY Afraid
 
Wow I was not expecting to see that when I logged in today! :ahh:

I do agree that this drive isn't the best choice for FRC application unless the game changed to limit robot to robot contact and it wasn't so complex. Every drive has its weaknesses.

I do want to congratulate you for thinking outside the box. Too many of us get caught up in trying to keep up with the elite teams by making swerves, WCDs, octocanums, etc without really trying to come up with the something new or reinvent the wheel er... ball! Great model and design! Keep it up! :)

I would encourage to keep working on this model. Maybe you could build one as an off-season/fun build project! You could even rework your wheel choice and have a sweet module for octocanum by switching one wheel for a mecanum!

To everyone else. Is the design far fetched? Maybe. Are there some problems in the design? Possibly. But let's keep this as a positive discussion. Someone put a lot of time and effort into designing and modeling something cool and by keeping positive discussions going we encourage more designs to come forth instead of lurking on members' desktops in fear that we will tear it apart in our posts. Every year I dream of 8wd WCD swerves! :rolleyes:

Nice work and good luck in your future designs!
Brendan

EricH 26-11-2012 22:52

Re: Be afraid... Be VERY Afraid
 
I would say that if the module were used in FRC competition, you'd probably run into some difficulties in steering and traction, as noted already. (Aside from the potential for terrain popping the ball out if there wasn't a retaining device--I didn't see one in the model.)

But... I would suggest building one as an offseason project if you don't do it in season. Build a T-shirt cannon (or other promo device) on top of it. Offseason demobots need a "COOL!" factor, which I think this has--especially if it works as advertised.

Garret 26-11-2012 23:51

Re: Be afraid... Be VERY Afraid
 
I spend my free time making crazy designs like this in CAD, as such I enjoy seeing these designs whenever I get the chance.

From my personal experience. It always sucks when people criticize a design a lot. I would advise you to hear it as trying to help, rather than that the design is bad or not worth pursuing. You should still listen to most of what has been said, but I think giving up on this design would be a real shame. I know of at least one instance where a company shot down an idea in a brainstorming session, and a number of years later made that one idea the most important distinguishing feature of their technology (and it still is to this day).

In my opinion FIRST would be much cooler if more teams were trying these crazy new designs, rather than just making another WCD and doing stuff purely because "its what the Einstein teams do". I hate hearing people say that a team shouldn't do something because no teams on Einstein used it. Reasons a team should use or not use those designs would be that they are what works for that team and because they would be the most effective for that challenge, not because some other really good team has done it or, the other extreme, no one else has done it.

These types of out-of-the-box designs are always fun/inspirational to design and look at. I found this design inspirational and it is a beautiful CAD model.

Cory 27-11-2012 00:04

Re: Be afraid... Be VERY Afraid
 
Quote:

Originally Posted by Garret (Post 1196416)

In my opinion FIRST would be much cooler if more teams were trying these crazy new designs, rather than just making another WCD and doing stuff purely because "its what the Einstein teams do". I hate hearing people say that a team shouldn't do something because no teams on Einstein used it. Reasons a team should use those designs would be that they are what works for that team and because they would be the most effective for that challenge, not because some other really good team has done it so let's do what they do.

There's nothing wrong with investigating "crazy new designs" from a learning perspective, the reality is just that in real life when you get a job you don't innovate for the sake of innovating, you only do it when existing solutions do not provide the solution to your problem.

Innovative is also not synonymous with good, or indicative of sound engineering. Both are important.

rcmolloy 27-11-2012 00:13

Re: Be afraid... Be VERY Afraid
 
Quote:

Originally Posted by Cory (Post 1196422)
There's nothing wrong with investigating "crazy new designs" from a learning perspective, the reality is just that in real life when you get a job you don't innovate for the sake of innovating, you only do it when existing solutions do not provide the solution to your problem.

Innovative is also not synonymous with good, or indicative of sound engineering. Both are important.

This has everything wrapped up in a nutshell regarding this entire thread.

Also to the original poster of the other quote, these teams aren't doing crazy designs because of their limitations and just the amount of work and effort required to realistically do some of them. Even though the crazy and novel ideas are very cool, they usually aren't the best practical ideas that you would like to go with.

Garret 27-11-2012 00:15

Re: Be afraid... Be VERY Afraid
 
Quote:

There's nothing wrong with investigating "crazy new designs" from a learning perspective, the reality is just that in real life when you get a job you don't innovate for the sake of innovating, you only do it when existing solutions do not provide the solution to your problem.

Innovative is also not synonymous with good, or indicative of sound engineering. Both are important.
I agree with you completely, I was more saying that people should use (or not use) designs for the right reasons.

Madison 27-11-2012 00:20

Re: Be afraid... Be VERY Afraid
 
Okay, I have a bit more time now that I'm not at work and can list some of the questions I have about how this would work and what the benefits are.

Years ago, I was able to receive a lot of fantastic feedback from engineers on this site, so there's definitely merit to posting ideas and soliciting feedback. That should continue, but a bit of humility goes a long way.

Questions/Concerns --

0. What problems were you trying to solve when making this design? What were the criteria you used to guide your decision making process and how did each of this model's elements satisfy that criteria?

1. Do you intend for the sphere to slip sideways against either driving roller? If not, no part of the sphere that isn't touched by a roller will ever contact the ground, so why use a sphere at all? If yes, the maximum possible force of friction between the sphere and the driving rollers must be less than the maximum force of friction between the roller and the carpet; this sacrifices available pushing force, putting your robot at a disadvantage in pushing matches.

2. The sphere is driven by a roller using friction. The available force of friction will be determined by the pneumatic cylinder pushing the roller against the sphere. You can use leverage here to make a smaller cylinder provide force that nears or exceeds your practical maximum possible friction force as determined by the coefficient of friction between your roller and sphere/sphere and carpet, but in all cases, the maximum friction force the robot can exert on the ground will be dictated by the sphere/carpet interaction.

3. When a roller is being pressed onto the surface of the sphere, because there is currently nothing shown to retain the ball from moving away from the drive rollers and ball casters, your mechanism will push the robot away from the sphere, unseating it from within the four ball casters. You will now be riding only on the thin interface between your drive roller and the sphere.
Driving forward or backward will cause the sphere to shift in the opposite direction. Any sideways motion will cause the sphere to slide sideways off the driven roller and, for lack of a better explanation, wedge itself into place between the drive rollers and two of the ball casters. You may not be able to recover from this condition except by going to a middle position that removes both rollers. The condition will repeat anytime you try to drive, however. You absolutely have to capture the sphere so that it cannot move vertically independently of the rest of the module for this to have ANY shot at working at all.

There are a lot of problems here and I think that, if you worked through them, you'd end up at a swerve module.

Akash Rastogi 28-11-2012 08:22

Re: Be afraid... Be VERY Afraid
 
Quote:

Originally Posted by Siri (Post 1196123)
What's the full weight of the unit (with swerve)? And what leads you to believe the ball system will be lighter than using wheel?

Going to add to this: Why not try to further the efficiency of existing designs? If you have the capabilities to create a carbon fiber sphere correctly and within budget, why not try to mold your own wheels or other parts out of carbon fiber? (doing either one properly is pretty difficult from what I've learned in materials classes... http://helix.gatech.edu/classes/ME41...inalpaper.html)

cmrnpizzo14 28-11-2012 21:02

Re: Be afraid... Be VERY Afraid
 
First off, awesome concept. I really can't believe that I just spent 30 minutes reading through all of this, but it is a fantastic concept. Your presentation of it leaves much to be desired ;) but it's forgiven (imho).

Secondly, I have a question about the maintenance of it. If you have the balls held in place by something other than the weight of the robot (which I heavily advise!) replacing a ball will be difficult. Hopefully you will never need to actually replace one, but just in case....

Also, the actual balls you will use will possibly be carbon fiber? Carbon fiber is cool, that is not deniable. However, how will you make them to resist impact. I am no expert on carbon fiber by any stretch of the imagination, but I do know that it is fragile. In addition, your wheels are what takes the most weight on the robot. I understand that a sphere is ideal for force distribution, but I still would be worried about damage to your balls.

Lastly, if you get this to work in any way at all, please put up some sort of video of it driving. This would be amazing to see work, even if it was not in a competition. Please keep us updated!

RedLeader342 15-12-2012 09:53

Re: Be afraid... Be VERY Afraid
 
it looks like the small one could be a build up wheel to the big one which then shoots a foam basketball.

Andrew Lawrence 15-12-2012 11:54

Re: Be afraid... Be VERY Afraid
 
Read through the thread, Redleader. The final design was released a long yime ago.


All times are GMT -5. The time now is 23:32.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi