Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Technical Discussion (http://www.chiefdelphi.com/forums/forumdisplay.php?f=22)
-   -   Moment of inertia and turning performance (http://www.chiefdelphi.com/forums/showthread.php?t=105309)

Al Skierkiewicz 05-04-2012 11:54

Re: Moment of inertia and turning performance
 
Quote:

Originally Posted by Alan Anderson (Post 1152966)
Have you ever tried to turn an active Segway by manhandling it instead of using the steering control? It has plenty of resistance. :D Granted, it's mostly due to software controlling the wheels and not to strictly mechanical tendencies.

I think it would be better to say that a Segway has very little wheel scrub to worry about.

Alan,
Even though the Segway is disabled there is something electrically that is working on the motors, akin to 'brake' mode. When you turn it off completely, it acts as you would expect.

icyplanetnhc 05-04-2012 19:35

Re: Moment of inertia and turning performance
 
Quote:

Originally Posted by Al Skierkiewicz (Post 1154524)
Haochuan,
The "jumpiness" that teams describe is primarily the torque of your drive train overcoming friction with the floor. As a wheel breaks loose and the RPM jumps up, the result is a mini-wheelie when the tires again grab the carpet. In most cases with high friction tires (four or more wheels) this is the case when turning.

We hypothesized that the jumpiness of our turning is caused by the inherent rocking of the standard West Coast 6WD drivetrains. This rocking causes inconsistent resisting torque to the turning. Each time the robot rocks during a turn, the front or rear wheels would grab the carpet, creating an abrupt but sharp resisting torque to our turning before the robot begins rocking the other way. Each rock seems to amplify the jumpiness and bounce.

However, we noted that some teams are able to create very smooth 6WD drivetrains that are similar in configuration to ours (254's Slipstream, for example). We will analyze some high-speed footage to get a better understanding of why we are suffering from this problem.

Al Skierkiewicz 05-04-2012 19:47

Re: Moment of inertia and turning performance
 
Robots have been doing this for more than the 17 years I have been here and drop wheel drive trains have not been around that long. When you have a lot of torque and friction something has to give somewhere.

JesseK 06-04-2012 09:47

Re: Moment of inertia and turning performance
 
Do a CD search for "Trapezoidal Speed Profile". Also do a CD search for 254's 2011 code, as that was posted as well. They've refined their code to optimize a couple of things. There may also be a difference in turning response with brake versus coast mode on the Jags/Vics.

MrForbes 06-04-2012 09:54

Re: Moment of inertia and turning performance
 
Quote:

Originally Posted by IKE (Post 1154095)
Kit bot=Noodle frame. It does have a fair amount of rock, so it works out well.
1114's Kitbot on Steroids improves this using the robust base-plate.

Super structure design can make a frame stiff.
Welding can make a frame stiffer.
Gussets can make a frame stiffer.
Additional rivets at attachment locations.
Gluing.
In general, Triangulation beats Boxification (I made the last one up).

Using tubes instead of channels for the fame members is really the final solution to the noodle frame problem.

mehrkam 06-04-2012 22:29

Re: Moment of inertia and turning performance
 
I have not read through all the threads, but many 6 WD drives drop the center wheels to rock back and forth. This is typically in the robot 101 build stuff.

Instead we raise the front wheels and shift the CG towards the back. This allows the robot to turn on a smaller profile. There is no rocking motion except for fast stops. The height that you raise the front wheels will depend on the wheels that you use.

Good luck for those continuing to St. Louis.

Chinmay 07-05-2012 15:54

Re: Moment of inertia and turning performance
 
Here is an update from the team regarding the "rocking while turning" problem:

"In our effort to determine why our robot 'dances' when it turns quickly, our first test was to find the horizontal center of mass/gravity (c.g.)

To our surprise, we found the c.g. to be very close to the center of the middle wheel. It is about 1" forward of the center wheel, as determined by balancing the robot on blocks.



Instead, we believe the problem is in the distribution of the mass.

The mass has large densities like below:


(S)---
|
|
|
|
|
----(B)

where (S) is the shooter, and (B) is the battery.

While the mass distribution is more complicated, the point is that as we rotate about the center vertical axis, the shooter wants to tilt the robot forward, and the battery wants to pull the back outward.

_-\
(S) \
\
\ _(B)
\-

In short, we need to pay more attention to the mass distribution for dynamic balance. While we are well balanced statically, our 2012 mass distribution results in poor dynamic balance."

Hopefully this can provide some new insight into this issue. There are so many complications that make creating a stable and smooth drivetrain harder than it seems :)

IKE 08-05-2012 07:15

Re: Moment of inertia and turning performance
 
Try adding about 10 lbs to the front, center, then rear. High and low. Take notes about how thw cg movement effects driving performance. These 6 trials will tell you a lot and will make a good experiment.

MichaelBick 08-05-2012 09:05

Re: Moment of inertia and turning performance
 
Another problem, that I noticed this year, was the field. I think many fields were not completely level due to maybe a budget decrease. I know for a fact we were "bouncing", with an 1/8 in drop, and very stiff chassis(we overbuilt the drivetrain, using 3 x 1 tubing instead of 2 x 1). Furthermore, I even saw teams that are known for their drive bouncing(ie. 1538 Holy Cows).

jwfoss 08-05-2012 09:34

Re: Moment of inertia and turning performance
 
Quote:

Originally Posted by MICHAELABICK (Post 1167961)
Another problem, that I noticed this year, was the field. I think many fields were not completely level due to maybe a budget decrease. I know for a fact we were "bouncing", with an 1/8 in drop, and very stiff chassis(we overbuilt the drivetrain, using 3 x 1 tubing instead of 2 x 1). Furthermore, I even saw teams that are known for their drive bouncing(ie. 1538 Holy Cows).

I believe that the bouncing you saw in 1538s drive was caused by their use of pneumatic wheels.
I noticed many teams this year with the same condition.

Chinmay 08-05-2012 10:25

Re: Moment of inertia and turning performance
 
Quote:

Originally Posted by IKE (Post 1167935)
Try adding about 10 lbs to the front, center, then rear. High and low. Take notes about how thw cg movement effects driving performance. These 6 trials will tell you a lot and will make a good experiment.

;) great idea. This will help us see if our suspicions were correct. I will post the results here. It's really a pretty complex problem with a ton of variables, but I like your idea to isolate a few situations and compare them

roystur44 08-05-2012 14:00

Re: Moment of inertia and turning performance
 
Chinmay,

Try some smaller wheels and drop the bumper as low as you can get it. Geared correctly with the smaller wheels you will find the robot accelerates quickly. We design in the placement of the wheels close to the perimeter of the frame. The frame needs to be very stiff if you want the robot to run fast and turn gracefully. Try to keep the weight of the robot as low as possible and between the front and rear axles.

Here are some pictures of our 6 wheel sheetmetal chassis for 2012:

https://picasaweb.google.com/1177698...12BuildSeason#

One of 971 summer projects is to invite local teams to a question and answer to our design process, materials, fabrication methods. Shoot us a email if you are interested.


Roy

Chinmay 08-05-2012 15:12

Re: Moment of inertia and turning performance
 
Quote:

Originally Posted by roystur44 (Post 1168038)
One of 971 summer projects is to invite local teams to a question and answer to our design process, materials, fabrication methods. Shoot us a email if you are interested.

Thanks for the offer :) I'm in CT right now and will be here for the next 10 weeks still in school doing research, but I'll have some of the kids on the team shoot you guys an email and take a closer look at your drive train. I'd love to see the facility and the robot in person near the end of summer when i get back home :) I'll be in touch...

PAR_WIG1350 10-05-2012 23:16

Re: Moment of inertia and turning performance
 
Quote:

Originally Posted by jwfoss (Post 1167968)
I believe that the bouncing you saw in 1538s drive was caused by their use of pneumatic wheels.
I noticed many teams this year with the same condition.

We had it, it was reduced when we switched from a gamepad to a pair of joysticks.
Our wheel base was probably only about 14" due to our use of 12 inch wheels on a wide robot, but we were not in danger of tipping over. In one match we were up at nearly a 45 degree angle on a tilted bridge and we still recovered and balanced the bridge. The only time we tipped was when we drove off the side of a bridge in Boston.


All times are GMT -5. The time now is 16:35.

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