Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Extra Discussion (http://www.chiefdelphi.com/forums/forumdisplay.php?f=68)
-   -   paper: 2014 HOT Tech Notes (http://www.chiefdelphi.com/forums/showthread.php?t=127827)

Adam Freeman 01-08-2014 12:51

Re: paper: 2014 HOT Tech Notes
 
Quote:

Originally Posted by Andrew Schreiber (Post 1395018)
Was this a VP belt that snapped?

Yes, it was a VP belt.

We ran at least 6 other 9mm VP belts/robot that survived an entire season (2 on intake, and 4 on outer drive wheels).

I'm pretty confident that something else happened to the one belt that snapped. We had just replaced the wheels that afternoon. Right after that, they called me and said the belt broke. I think they tried to roll the belt on to the pulley and cut it, instead of assembling it correctly with the belt on the pulley.

JohnFogarty 01-08-2014 13:04

Re: paper: 2014 HOT Tech Notes
 
I've been analysing a lot of drive train design this summer and I've been meaning to ask since watching this year's Einstein matches. Where It seemed your drive had the worst time in escaping t-bone pins. Do you think there is a simple method to improving an 8WD to escape t-bone pins so that the power and robustness of the drive can be maintained.

The drop-down wheels that have begun popping up on various team's drives offers an interesting solution, but I'd just like to hear what you think.

waialua359 01-08-2014 15:23

Re: paper: 2014 HOT Tech Notes
 
Quote:

Originally Posted by JohnFogarty (Post 1395032)
Do you think there is a simple method to improving an 8WD to escape t-bone pins so that the power and robustness of the drive can be maintained.

The drop-down wheels that have begun popping up on various team's drives offers an interesting solution, but I'd just like to hear what you think.

We had an 8WD setup as well and T-bone pins were a problem at times. Besides the drop-down wheels solution, others employed the "hexagon" shaped frame perimeter. We plan to look at the drop down wheels as our 1st option for next season, if necessary.

Adam Freeman 01-08-2014 16:13

Re: paper: 2014 HOT Tech Notes
 
Quote:

Originally Posted by JohnFogarty (Post 1395032)
I've been analysing a lot of drive train design this summer and I've been meaning to ask since watching this year's Einstein matches. Where It seemed your drive had the worst time in escaping t-bone pins. Do you think there is a simple method to improving an 8WD to escape t-bone pins so that the power and robustness of the drive can be maintained.

The drop-down wheels that have begun popping up on various team's drives offers an interesting solution, but I'd just like to hear what you think.

I agree with your assesment... I think we were extremely bad at getting out of t-bone pins.

I'm not sure there is a simple solution to the problem. I don't think the issue was only b/c of the 8WD drivetrain setup. I think that the combination of the 8WD, long chassis, rectangular chassis shape, and driver ability all resulted in us getting stuck in a lot of t-bones.

Drop down omnis are becoming more and more common. We entertained adding breifly them during the season (after MSC), but never really pursued it to far. It would have require quite a bit of designing and changes that would have only resulted in marginal gains at Champs.

We actually started running different combination of omni wheels in the offseason that helped a little bit, but we still had issues with pins.

As Glen mentioned, we will probably evaluate different chassis shapes to help lower the chances of a full on side pin. I think it will be a point of focus for our chassis design next season.

I don't think we are ready to add in drop down wheels to our drive system. Teams like 469 and 254 did pretty well without them on a standard 6/8WD setup. I'd like to see more of a square chassis setup next year and some better driving to avoid defense. We haven't entertained switching to 6WD much in the past couple years, but it might be something to look into next year.

AdamHeard 01-08-2014 16:21

Re: paper: 2014 HOT Tech Notes
 
Quote:

Originally Posted by Adam Freeman (Post 1395047)
I agree with your assesment... I think we were extremely bad at getting out of t-bone pins.

I'm not sure there is a simple solution to the problem. I don't think the issue was only b/c of the 8WD drivetrain setup. I think that the combination of the 8WD, long chassis, rectangular chassis shape, and driver ability all resulted in us getting stuck in a lot of t-bones.

Drop down omnis are becoming more and more common. We entertained adding breifly them during the season (after MSC), but never really pursued it to far. It would have require quite a bit of designing and changes that would have only resulted in marginal gains at Champs.

We actually started running different combination of omni wheels in the offseason that helped a little bit, but we still had issues with pins.

As Glen mentioned, we will probably evaluate different chassis shapes to help lower the chances of a full on side pin. I think it will be a point of focus for our chassis design next season.

I don't think we are ready to add in drop down wheels to our drive system. Teams like 469 and 254 did pretty well without them on a standard 6/8WD setup. I'd like to see more of a square chassis setup next year and some better driving to avoid defense. We haven't entertained switching to 6WD much in the past couple years, but it might be something to look into next year.

This is all conjecture, but it looks like your chassis might have the ability to twist some vertically.

I've had a running theory for a while that a chassis such as that gets shoved into the carpet a bit odd under high defensive forces, contributing to the "locking up" affect.

Any thoughts?

Andrew Schreiber 01-08-2014 16:26

Re: paper: 2014 HOT Tech Notes
 
Quote:

Originally Posted by AdamHeard (Post 1395049)
This is all conjecture, but it looks like your chassis might have the ability to twist some vertically.

I've had a running theory for a while that a chassis such as that gets shoved into the carpet a bit odd under high defensive forces, contributing to the "locking up" affect.

Any thoughts?

Could be a similar effect from what 33 noticed in 08 with the impact chassis rigidity had on their ability to turn (http://www.chiefdelphi.com/forums/sh...ad.php?t=68307) ?

AdamHeard 01-08-2014 16:28

Re: paper: 2014 HOT Tech Notes
 
Quote:

Originally Posted by Andrew Schreiber (Post 1395050)
Could be a similar effect from what 33 noticed in 08 with the impact chassis rigidity had on their ability to turn (http://www.chiefdelphi.com/forums/sh...ad.php?t=68307) ?

Exactly, 100%.

My thoughts are it allows wheels that shouldn't be touching to be forced to touch, causing a wheelbase longer than you're supposed to be on. Combined with the already huge lateral forces from the defender, you're STUCK!

If only a bunch of teams would be willing to mount gopros to the underside of their robots for a few events...

It's interesting how the drivetrain arms race has evolved.

Andrew Schreiber 01-08-2014 16:31

Re: paper: 2014 HOT Tech Notes
 
Quote:

Originally Posted by AdamHeard (Post 1395051)
Exactly, 100%.

My thoughts are it allows wheels that shouldn't be touching to be forced to touch, causing a wheelbase longer than you're supposed to be on. Combined with the already huge lateral forces from the defender, you're STUCK!

It certainly makes sense that it'd be possible. Got a rigid chassis and a "noodle" chassis handy? Seems a pretty quick test to see if it's easier to tbone a noodle chassis.


I'd also be curious to see how it deals with omni directional drivetrains (mecanums or 33/78 style omnis). Guess I should get drawing FBDs.

Andrew Lawrence 01-08-2014 17:06

Re: paper: 2014 HOT Tech Notes
 
Quote:

Originally Posted by AdamHeard (Post 1395049)
This is all conjecture, but it looks like your chassis might have the ability to twist some vertically.

I've had a running theory for a while that a chassis such as that gets shoved into the carpet a bit odd under high defensive forces, contributing to the "locking up" affect.

Any thoughts?

If I may request clarification - when you say "twist some vertically", do you mean the chassis bows upwards down the center parallel with the drive wheels due to lack of frame stiffness, and through this bowing all wheel sets are in contact with the ground?

AdamHeard 01-08-2014 17:13

Re: paper: 2014 HOT Tech Notes
 
Quote:

Originally Posted by Andrew Lawrence (Post 1395056)
If I may request clarification - when you say "twist some vertically", do you mean the chassis bows upwards down the center parallel with the drive wheels due to lack of frame stiffness, and through this bowing all wheel sets are in contact with the ground?

Corners (with respect to the other corners) move up or down as the frame twist.

As for Andrew's request, we only have rigid frames ;)

MrRiedemanJACC 01-08-2014 20:21

Re: paper: 2014 HOT Tech Notes
 
We have a flexible chassis from this year, for Andrew's testing! But we used mechanums and therefore we weren't as concerned with frame stiffness. I don't know for sure that is was a good idea though. I think the hard part was that the loading would change on each wheel therefore changing how it drove. We would keep contact with all 4 wheels though. Also in our third event we changed from mechanum to traction and kop wheels (4wd) to try to play more defense for other newer teams. Worked well, but not sure I'd do 4wd drive again. Too much power with 8 motors in the drivetrain and we tripped the main breaker in our last two matches.

It is interesting in the drivetrain wars, how much it has changed the forces the robots receive. There is much more involved than picking motors and gear ratios, the frame is definitely part of it and maybe the Bees were onto something with the Omni wheels. Maybe the frame doesn't need to be as stiff and that weight can be used elsewhere with the omni directional drivetrains. Or you have to go in the other direction of stiffness. It's always a tradeoff.

Adam Freeman 01-08-2014 20:52

Re: paper: 2014 HOT Tech Notes
 
Quote:

Originally Posted by AdamHeard (Post 1395049)
This is all conjecture, but it looks like your chassis might have the ability to twist some vertically.

I've had a running theory for a while that a chassis such as that gets shoved into the carpet a bit odd under high defensive forces, contributing to the "locking up" affect.

Any thoughts?

Never thought about it that way. I would guess no, since the entire upper structure tubing was mounted through the chassis so I doubt there was much flex in the drivetrain. We had an issue during the build season where the claw was misaligned to the frame....it was almost impossible to flex the frame and upper structure to straighten it. We had to seperate components to get any movement.

But, anything is possible we've never looked at a corner to corner stiffness in the vertical direction.

My honest opinion is that we put ourselves in bad positions during the match which lead to a lot of easy t-bones for the defense.

magnets 01-08-2014 21:34

Re: paper: 2014 HOT Tech Notes
 
When some teams are in t-bones, their robot vibrates and shakes as its pushed sideways from the wheels getting and then losing traction with the ground, and it's almost impossible to get out of these pins. From what I've seen, it happens to robots that don't rock, like swerve (4 wheels in contact at all times) and 8WD (4 wheels in contact at all times).

I agree that avoiding defense is pretty important. We had issues with getting pushed in circles from the side and we had a traction limited 6 CIM drive with 2" wide McMaster blue nitrile tread on our wheels, like 254 does. 254 weighed well under 120 lbs, but didn't have too many defense issues because they were very smart drivers. 118 had a similar strategy. They were geared VERY high and had a single speed gearbox that would most likely trip breakers very quickly if they got into a pushing match, but they focused on avoiding defense, and it worked pretty well for them.

waialua359 01-08-2014 21:36

Re: paper: 2014 HOT Tech Notes
 
Quote:

Originally Posted by Adam Freeman (Post 1395071)
Never thought about it that way. I would guess no, since the entire upper structure tubing was mounted through the chassis so I doubt there was much flex in the drivetrain. We had an issue during the build season where the claw was misaligned to the frame....it was almost impossible to flex the frame and upper structure to straighten it. We had to seperate components to get any movement.

But, anything is possible we've never looked at a corner to corner stiffness in the vertical direction.

My honest opinion is that we put ourselves in bad positions during the match which lead to a lot of easy t-bones for the defense.

This past weekend, I got a chance to work with 16 as our partner at OZARK during eliminations. I took notice how easy it was for their swerve to just move away from a t-bone and drive away.
Even though we practiced a lot with 368, I never took notice because it wasnt a full field of robots and with defense from multiple robots.
Swerve is another direction we might move towards, as we tinker with drivetrain setups during the off-season. We already built a swerve back in 2009 but never implemented it into a real design for an FRC challenge.

Andrew Schreiber 02-08-2014 12:05

Re: paper: 2014 HOT Tech Notes
 
Quote:

Originally Posted by waialua359 (Post 1395079)
This past weekend, I got a chance to work with 16 as our partner at OZARK during eliminations. I took notice how easy it was for their swerve to just move away from a t-bone and drive away.
Even though we practiced a lot with 368, I never took notice because it wasnt a full field of robots and with defense from multiple robots.
Swerve is another direction we might move towards, as we tinker with drivetrain setups during the off-season. We already built a swerve back in 2009 but never implemented it into a real design for an FRC challenge.

As you may have noticed watching "6" a lot of that is drive practice too. JT knows that robot really well and controlling it is almost second nature.


All times are GMT -5. The time now is 22:54.

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