Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Robot Showcase (http://www.chiefdelphi.com/forums/forumdisplay.php?f=58)
-   -   Baxter Bomb Squad 2014 ~ Nighthawk (http://www.chiefdelphi.com/forums/showthread.php?t=127052)

Canon reeves 23-02-2014 21:56

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
wow, now this will make Arkansas regional even more interesting! Can't wait to watch yall drive circles around us and most of the other robots again!:p It really looks great, yall did a really good job this season. How did I know that yall would do a catapult? Gut feeling I guess?

jwfoss 24-02-2014 08:53

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
This Bomb Squad machine looks to be one of your most competitive yet. Thanks again for the inspiration on the intake, FRC558 is running a polycarbonate armed, motor in roller, intake as well this year.

Rob Stehlik 24-02-2014 09:12

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
Quote:

Originally Posted by JTN (Post 1348299)
The reason I said the original design of the cams earlier in the post was because we added one slight notch on each side that allowed the catapult to rest slightly when cocked to lessen the stress on the van door motors. We actually learned that it was possible to remove those two motors entirely (the 775s were doing all the work anyway) and the small notches prevent any possibility of back-driving.
-John Taylor Novak

Nice to see another cam based catapult! We came up with a different solution to back driving that you may be interested in: The final 30 degrees of our cam profile is circular, so there is no reaction torque from the catapult springs. Once the cam follower gets to the circular part of the cam, there is no chance of back driving.



As a side note, we built a small cam driven catapult last year for a non FRC project, and experimented with a notch in the cam to prevent back driving. It worked, but we found there to be a large current spike when the motor was trying to get the follower out of the notch.

Jefferson 24-02-2014 09:29

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
Quote:

Originally Posted by Rob Stehlik (Post 1348843)
Nice to see another cam based catapult! We came up with a different solution to back driving that you may be interested in: The final 30 degrees of our cam profile is circular, so there is no reaction torque from the catapult springs. Once the cam follower gets to the circular part of the cam, there is no chance of back driving.

As a side note, we built a small cam driven catapult last year for a non FRC project, and experimented with a notch in the cam to prevent back driving. It worked, but we found there to be a large current spike when the motor was trying to get the follower out of the notch.

The reliability and repeatability of the cam driven catapult was the deciding factor for us.

We had a similar profile you describe, just not as long on the flat. Hitting that small window right before the roll over was tough. We also like the idea of being mechanically locked in there.

We do get a pretty significant current spike coming out of the notch. I watched our practice bit spike over 60A the other night. It took us several rounds to get the slope flat enough come back out to reliably fire.

Btw, y'all's robot looks very nice again this year. Good luck out there.

Jefferson 24-02-2014 10:11

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
Quote:

Originally Posted by Andrew Schreiber (Post 1348605)
Nice to see this thing finally public. Hopefully I'll be able to stick my head inside and talk swerve with you guys at CMP this year.

Please do. Nothing I like more than to geek out over swerve drive for a while.
:D

Andrew Schreiber 24-02-2014 10:19

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
Quote:

Originally Posted by Jefferson (Post 1348866)
Please do. Nothing I like more than to geek out over swerve drive for a while.
:D

JT was mentioning you guys had some issues with Command Based robot stuff last year, care to elaborate?

And, I'll absolutely come over and geek out. :)

Jimmy Nichols 24-02-2014 11:42

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
Quote:

Originally Posted by Jefferson (Post 1348355)
There is a 550 inside the beater bar. So all we have running up the arms are the motor leads.

We also have Polycarb arms and inserted the motors inside the acquisition tube. I was wondering if anyone else would do this. See ya at QCR!

Jefferson 24-02-2014 12:34

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
Quote:

Originally Posted by Andrew Schreiber (Post 1348875)
JT was mentioning you guys had some issues with Command Based robot stuff last year, care to elaborate?

And, I'll absolutely come over and geek out. :)

I wouldn't say we had issues, but we definitely didn't use commands this year. They overcomplicated the small tasks and made the more complicated tasks very difficult. Especially where we needed more time resolution than whatever cycle the commands were on. The compile time for that many objects was hard to stomach also.

Our 2013 code is up here if you want to see how we inplemented commands:
https://github.com/FRCTeam16/TMW2013

I guess we are using commands one place this year, dashboard buttons. We have two or three things we need to do infrequently enough that don't need to be accidentally pressed on the joystick. Commands make that very easy.

Jefferson 24-02-2014 12:47

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
Quote:

Originally Posted by Jimmy Nichols (Post 1348912)
We also have Polycarb arms and inserted the motors inside the acquisition tube. I was wondering if anyone else would do this. See ya at QCR!

Yeah, the Lexan arms may look floppy and less "industrial" than rigid ones, but there's good reason we've had essentially this in 2008, 2010, 2012, and 2014. I'm beginning to see a pattern...

Andrew Schreiber 24-02-2014 13:19

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
Quote:

Originally Posted by Jefferson (Post 1348930)
I wouldn't say we had issues, but we definitely didn't use commands this year. They overcomplicated the small tasks and made the more complicated tasks very difficult. Especially where we needed more time resolution than whatever cycle the commands were on. The compile time for that many objects was hard to stomach also.

Our 2013 code is up here if you want to see how we implemented commands:
https://github.com/FRCTeam16/TMW2013

I guess we are using commands one place this year, dashboard buttons. We have two or three things we need to do infrequently enough that don't need to be accidentally pressed on the joystick. Commands make that very easy.


Interestingly, I felt similarly when I tried to use CPP for commands last year. Since switching to Java I found Command base was fairly simple and it meant a good chunk of our code was much cleaner than comparable other methods. I know it has allowed us to iterate our code far faster. I'll have to dig into this code and take a look at your new code at CMP this year.

Of course, my day job is primarily ruby/javascript code so that could explain why CPP felt less intuitive.

Bryce2471 24-02-2014 14:47

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
Could you provide some details about your swerve drive design? Such as wheel size, steering gearing, drive gearing, slip rings, or coaxial.
Our team did independent swerve drive for the first time this year.
Thanks in advance.

Meredith Novak 24-02-2014 16:53

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
Quote:

Originally Posted by Andrew Schreiber (Post 1348605)
Nice to see this thing finally public. Hopefully I'll be able to stick my head inside and talk swerve with you guys at CMP this year.

Just make sure the catapult is disabled before you stick your head in...that thing really scares me!

Andrew Schreiber 24-02-2014 16:55

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
Quote:

Originally Posted by Meredith Novak (Post 1349105)
Just make sure the catapult is disabled before you stick your head in...that thing really scares me!

I'll avoid standing on it while it's loaded too. ;)

Meredith Novak 24-02-2014 16:56

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
Quote:

Originally Posted by Andrew Schreiber (Post 1349107)
I'll avoid standing on it while it's loaded too. ;)

I have pictures. JT better stay clear of safety judges.

JTN 24-02-2014 17:08

Re: Baxter Bomb Squad 2014 ~ Nighthawk
 
Quote:

Originally Posted by Bryce2471 (Post 1349030)
Could you provide some details about your swerve drive design? Such as wheel size, steering gearing, drive gearing, slip rings, or coaxial.
Our team did independent swerve drive for the first time this year.
Thanks in advance.

Our swerve drive uses slip rings mounted at the top and we use 6" AndyMark performance wheels. I have some CAD models of our 2013 drive train if you'd like to see. Our output gear ratio is about 9.5:1 on the drive and for steering we use AM 9015s geared down using Versaplanetaries from VexPro (I'll have to figure out the gearing of that when I go to the shop later today). Here are some pictures of the model for a more visual approach. If you have any other questions, please don't hesitate to ask. Just send me a PM asking for the gear ratios for the steering and I'll try to get them to you as quickly as possible.



-John Taylor Novak


All times are GMT -5. The time now is 16:49.

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