Chief Delphi - How fast was the fastest mecaum?

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)

- Technical Discussion (http://www.chiefdelphi.com/forums/forumdisplay.php?f=22)

- - How fast was the fastest mecaum? (http://www.chiefdelphi.com/forums/showthread.php?t=132732)

asid61

12-01-2015 22:35

How fast was the fastest mecaum?

Out of curiosity, has anybody actually geared their mecanum drives for anything above 15-16fps? Generally I see speeds of 10-12fps with mecanums.
I'm asking because I always see the claim that mecanums can't go fast, but I've actually never heard of anybody who has tried. There's nothing that limits it in theory, so I wanted to know the practical limit.

Jay O'Donnell

12-01-2015 22:38

Re: How fast was the fastest mecaum?

2052 has done a pretty good job of a fast mecanum bot: http://youtu.be/Ogy65hEPIXk

asid61

12-01-2015 22:48

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by Jay O'Donnell (Post 1427122)

2052 has done a pretty good job of a fast mecanum bot: http://youtu.be/Ogy65hEPIXk

That's a really well done mecanum. I can't find the actual speed listed anywhere though, although it definitely looks pretty fast.

mschwab013

12-01-2015 22:54

Re: How fast was the fastest mecaum?

At 291 we usually have pretty fast mecanum. We build custom gearboxs and see about 13-14 fps

SpeedFreed

12-01-2015 23:20

Re: How fast was the fastest mecaum?

For a summer project, we built an octanum drivetrain powered off of 4 CIMs and 4 Mini-CIMs, and geared the mecanum wheels to 20.5 FPS free-speed. We saw speeds of about 17 FPS, on the field, and 19 FPS in our school's hallways.

In addition, we used the OLD Vex 6" mecanum wheels (the ones that break), and still had them last well enough for whole off-season competition. We have since switched to the updated versions, and they look like they should be fine (haven't tested through a competition though).

asid61

13-01-2015 00:17

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by SpeedFreed (Post 1427143)

WOW see now that's what I'm looking for! That's actually a really fast speed for any drive, especially mecanum.
How did strafing go? Even if it was slower than straight, how fast was it? Did you run a suspension?

Ether

13-01-2015 01:37

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by SpeedFreed (Post 1427143)

geared the mecanum wheels to 20.5 FPS free-speed. We saw speeds of about 17 FPS, on the field, and 19 FPS in our school's hallways

That's 93% "speed loss constant" in the hallway and 83% on the field. Interesting.

How did you measure the actual speed? e.g. video, stopwatch, etc?

Would you be willing to share the raw data?

Jay O'Donnell

13-01-2015 06:23

Re: How fast was the fastest mecaum?

Here's another one, keep in mind they aren't on carpet. http://youtu.be/O_vvVU4OBT4

Jay O'Donnell

13-01-2015 06:26

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by SpeedFreed (Post 1427143)

Was this due to never having enough space/time on the field to accelerate to your full speed? That would be a problem this year since the field will get cramped and there won't be room to accelerate much.

Ether

13-01-2015 08:51

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by Jay O'Donnell (Post 1427213)

Was this due to never having enough space/time on the field to accelerate to your full speed?

He said 20.5 fps free speed, which presumably meant "vehicle speed at motor spec free speed". The reduced actual top speed presumably due to JVN's "speed loss constant".

Chexposito

13-01-2015 09:15

Re: How fast was the fastest mecaum?

Team Titanium's 2008 robot comes to mind.

Jay O'Donnell

13-01-2015 09:18

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by Ether (Post 1427249)

He said 20.5 fps free speed, which presumably meant "vehicle speed at motor spec free speed". The reduced actual top speed presumably due to JVN's "speed loss constant".

I was asking about the difference between the school tested speed and field speed. I'm assuming with a bigger runway one would have more time to accelerate than on a real field that has obstacles.

Ether

13-01-2015 09:26

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by Jay O'Donnell (Post 1427259)

I was asking about the difference between the school tested speed and field speed.

Pronouns are the cause of many misunderstandings.

The difference in hall vs field speed could be due to the difference in compliance of the flooring (carpeted field vs tiled hallway).

FrankJ

13-01-2015 10:19

Re: How fast was the fastest mecaum?

In a frictionless physics lab I would expect the free speed between the field (carpet?) & hall wall (tiles of other hard surface?) to be the same. In the world the wheels may not be perfectly aligned. The macanum rollers turns more freely on a hard surface rather than carpet. All this create side forces. The side forces net out to zero but consume energy thus a lower speed.

silverD

13-01-2015 10:37

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by asid61 (Post 1427129)

That's a really well done mecanum. I can't find the actual speed listed anywhere though, although it definitely looks pretty fast.

Thanks :)
Our 2013 robot, Silverfish, used 8" mecanum with a 12.75:1 reduction for a free-speed of 14.53 ft/s (11.77 ft/s adjusted).
Our 2014 robot, Dragonfly, again used 8" mecanum, but with a 10.71:1 reduction for a free-speed of 17.3 ft/s (14 ft/s adjusted).

Ether

13-01-2015 10:38

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by FrankJ (Post 1427281)

In a frictionless physics lab I would expect the free speed between the field (carpet?) & hall wall (tiles of other hard surface?) to be the same.

I wouldn't.

The compliant flooring (carpet) moves slightly as the mec rolls forward over it, due to the tranverse force component exerted by the roller. This motion is accompanied by roller rotation, which subtracts from forward motion somewhat.

That's the theory anyway. It would be enlightening to see a controlled test to investigate the magnitude of this effect quantitatively.

Ether

13-01-2015 10:44

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by silverD (Post 1427293)

(11.77 ft/s adjusted)...(14 ft/s adjusted).

did you do any actual speed measurements?

silverD

13-01-2015 10:49

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by Ether (Post 1427298)

did you do any actual speed measurements?

Ether,
Sorry to report we didn't. These numbers are based on the JVN calculator.

Ether

13-01-2015 11:01

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by silverD (Post 1427299)

Sorry to report we didn't. These numbers are based on the JVN calculator.

I suspected as much, but hoped otherwise.

We are trying to establish whether or not the 81% "speed loss constant" is a reasonable value to use for mec on carpet.

silverD

13-01-2015 11:45

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by Ether (Post 1427305)

I suspected as much, but hoped otherwise.

We are trying to establish whether or not the 81% "speed loss constant" is a reasonable value to use for mec on carpet.

I agree, this would be great to know. We hold driver practice on competition carpet. I suppose we could mark out the carpet, capture video, and try to get some numbers.

Bob Steele

13-01-2015 15:54

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by FrankJ (Post 1427281)

In a frictionless physics lab the mecanum drive would not move the robot.
velocity would be zero

Ether

13-01-2015 16:07

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by Bob Steele (Post 1427533)

In a frictionless physics lab the mecanum drive would not move the robot.
velocity would be zero

I assume this was intended to be humorous. Frank certainly did not intend to imply that the friction between the floor and the wheels is zero.

Bob Steele

13-01-2015 16:29

Re: How fast was the fastest mecaum?

You might take it to be humorous... but as a physics teacher I would like specificity.... I see many statements on CD that sound "right" but make little sense. (note... not from you Ether... you are quite good at getting to the specificity of the issue)

Do the side forces from the mecanums really "cancel each other out"
The net force accelerates the robot. side forces oppose each other but that energy goes somewhere. Has anyone ever felt mecanum wheels after prolonged driving? Conservation of energy is real.

Just a comment... not a condemnation....

Ether

13-01-2015 16:45

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by Bob Steele (Post 1427566)

Do the side forces from the mecanums really "cancel each other out"

Yes they do.

Quote:

side forces oppose each other but that energy goes somewhere.

Force is not energy. If there's no motion in the direction of the force, there's no energy involved.

Quote:

Has anyone ever felt mecanum wheels after prolonged driving? Conservation of energy is real.

That's due to friction in the rollers, roller axial free play, and floor compliance.

Ether

13-01-2015 16:50

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by Bob Steele (Post 1427566)

as a physics teacher I would like specificity.

I can relate to that. As a graduate student at a major engineering university back in the 70s, I taught physics to EE students.

FrankJ

13-01-2015 17:05

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by Bob Steele (Post 1427533)

In a frictionless physics lab the mecanum drive would not move the robot.
velocity would be zero

Ether is right. I use the term "frictionless physics lab" to mean I am ignoring trivial & sometime not so trivial factors to simplify the discussion.

In a truly frictionless lab, velocity for mecanums on a hard surface would be zero. On carpet it would be somewhat less than zero since the compliance in the carpet would give some forces not dependent on gravity.

The angled wheels gives you side force vectors. If the forces are not canceled out, then the robot accelerates in reaction to those forces. Especially in a frictionless physics lab.

My high school physics teacher thought I only took physics to play with the toys. He was largely right. :)

SpeedFreed

14-01-2015 01:05

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by asid61 (Post 1427172)

How did strafing go? Even if it was slower than straight, how fast was it? Did you run a suspension?

Quote:

Originally Posted by Ether (Post 1427191)

How did you measure the actual speed? e.g. video, stopwatch, etc?

Testing was, well, far from scientific, in that we ran it down the hall way a couple times and timed it with stopwatch on a phone. I'd give it a good ± 1 FPS or so. A similar approach was used on carpet.

Quote:

Originally Posted by Jay O'Donnell (Post 1427213)

Both carpet, and non-carpet tests were done over similar distances. I am pretty sure it was due to the extra loss in energy to the carpet (or perhaps to the gobs carpet fuzzies in the gearboxes) Regardless, I 100% do not recommend this set up for this year.

Chief Hedgehog

14-01-2015 01:25

Re: How fast was the fastest mecaum?

I would also second checking with 2052. They seem to have Mecanum down pat.

asid61

14-01-2015 02:51

Re: How fast was the fastest mecaum?

Quote:

Originally Posted by SpeedFreed (Post 1427967)

Strafing was pretty slow. Keep in mind this was with single speed gearboxes (yes single speed, geared to 20.5 FPS, and mecanum), so strafing was probably no more than 8-9 FPS (this number is entirely guestimate). This was, however, on the set of mecanum wheels with nearly 75% of the rollers broken. All this being said, we still ran it with FOD (nav6), and I (the driver) thought it was great.

Also, we didn't use suspension. I find that making the frame loose, adding suspension, etc, to be pretty unnecessary. The Vex mecanum wheels have that extra flex in them which I think is sufficient to replace the above (at least it was fine for us). I would still make sure it is as flat as you can make it when putting it together though.

Testing was, well, far from scientific, in that we ran it down the hall way a couple times and timed it with stopwatch on a phone. I'd give it a good ± 1 FPS or so. A similar approach was used on carpet.

Both carpet, and non-carpet tests were done over similar distances. I am pretty sure it was due to the extra loss in energy to the carpet (or perhaps to the gobs carpet fuzzies in the gearboxes) Regardless, I 100% do not recommend this set up for this year.

Those are heartening results. We'll have to order a few different gear and sprocket sets for ours then.

All times are GMT -5. The time now is 15:51.