View Full Version : pic: Team XBot Mecanum Practice Chassis
Fred Sayre
15-11-2006, 16:02
[cdm-description=photo]25886[/cdm-description]
Tom Bottiglieri
15-11-2006, 16:05
Looks awesome! The modules look very robust and pretty. Some plating on that thing would make it look like a beast.
I have some questions though. Did you notice any flex in the frame when driving? It looks to me like it is all bolted together. I know people have had problems with frame flex in mecanum drive trains. One of the wheels would lift off the ground, messing up the resultant total force vector, sending the robot in another unforeseen direction.
Lil' Lavery
15-11-2006, 16:18
I have some questions though. Did you notice any flex in the frame when driving? It looks to me like it is all bolted together. I know people have had problems with frame flex in mecanum drive trains. One of the wheels would lift off the ground, messing up the resultant total force vector, sending the robot in another unforeseen direction.
I beleive that they have a suspension (similar to 40's) on each wheel module to attempt and accomodate for any bending, imprecision, or minor terrain features.
Certainly a great looking protoype, how much does it weigh?
Fred Sayre
15-11-2006, 16:22
Looks awesome! The modules look very robust and pretty. Some plating on that thing would make it look like a beast.
I have some questions though. Did you notice any flex in the frame when driving? It looks to me like it is all bolted together. I know people have had problems with frame flex in mecanum drive trains. One of the wheels would lift off the ground, messing up the resultant total force vector, sending the robot in another unforeseen direction.
Check out http://www.chiefdelphi.com/forums/showthread.php?t=49213 for a better description.
Each wheel/gearbox pod pivots. The tennis and racket balls are what we are testing out for suspension springs. Right now, we are having more of an issue making the friction in the drive pods equal (we had one really beat up CIM in this chassis). If one wheel takes a second longer to start spinning then the same issue occurs.
Overall though our drivers were starting to be able to compensate for the quirks in the movement, so when we get it nice and tuned it should only get better :)
Looks very nice.
How well does it handle stairs? :)
-dave
Looks very nice.
How well does it handle stairs? :)
-dave
It's too much fun to drive and a lot of fun to watch. If it needs to climb stairs, I'll make it climb stairs.
deshirider430
15-11-2006, 16:48
Lavery as in this years game is going to have stairs!? Just playing, thats a really nice frame and set up.
Ben Piecuch
15-11-2006, 16:48
Beautiful moves! I can't wait to see what the drivers can do with some more practice. One question, any idea why it seems to be so noisy (in the video?) With your custom gearboxes I figured they would be nice and quiet. Or, are the AM Mecanums just that loud? (Sorry for the insinuation Andy/Mark...)
<offtopic>
Dave, it's about that time of the year when the real hints start coming our way. Is this your first "plant" here at the forums for us to deal with?
</offtopic>
Please keep us update with both your mechanical tuning and software tweaking (and methodology, if you will...) Thanks!
BEN
Beautiful moves! I can't wait to see what the drivers can do with some more practice. One question, any idea why it seems to be so noisy (in the video?) With your custom gearboxes I figured they would be nice and quiet. Or, are the AM Mecanums just that loud? (Sorry for the insinuation Andy/Mark...)
<offtopic>
Dave, it's about that time of the year when the real hints start coming our way. Is this your first "plant" here at the forums for us to deal with?
</offtopic>
Please keep us update with both your mechanical tuning and software tweaking (and methodology, if you will...) Thanks!
BEN
The first gearbox we assembled is markedly louder than the remaining three and it's also the one that's slowest to start, so we're going to tear it down and rebuild it. We think the first handful of shafts we made are inaccurate, so the gearbox isn't square and that's contributing to some binding. The chain is also a bit loose and I screwed up the tensioning mechanism some, so that'll be fixed in the next iteration -- whenever there's time for that.
The wheels, even on carpet, are a bit bumpy because the roller profile only approximates an elliptical shape. More weight may sink the chassis deeper into the carpet and alleviate some of the uneven ride; but that's also contributing to some of the noise you hear. Right now, sans battery, I'd say it weighs about 50 pounds. We'll get an accurate measurement of that tomorrow, we hope.
Edit: There's also no grease in the gears. We need to fix that, too. :)
Looks very nice.
How well does it handle stairs? :)
-dave
HINT HINT OMG HINT!!!!!!!!!
lol something that could climb stairs hmmm
*tears apart an iBot* MINE CAN NOW
Rich Kressly
15-11-2006, 20:12
I'll never get tired of watching one of these guys move. Excellent job getting to this point for sure. As for Dave and the stairs, just ignore him like the rest of the FRC GDC does... ;)
Billfred
15-11-2006, 20:24
I was looking at the full-size version of the picture, and it looks like the actual frame is just four lengths of box tubing and two C-channels bolted together. I like the layout--is there more to it, or am I just not seeing it?
Rob2713g
15-11-2006, 20:43
Each wheel/gearbox pod pivots. The tennis and racket balls are what we are testing out for suspension springs.
Could you please provide an up close picture of your suspension or explain it in a little more detail (I'm new to suspensions). Thanks a lot!
Fred Sayre
15-11-2006, 21:13
I was looking at the full-size version of the picture, and it looks like the actual frame is just four lengths of box tubing and two C-channels bolted together. I like the layout--is there more to it, or am I just not seeing it?
Nope, thats it! The underside of the C channel is milled out to accommodate the suspension, holes were drilled for where the drive pods rotate, and 8 holes were drilled to bolt the tube and C channel together.
AdamHeard
15-11-2006, 21:25
Nope, thats it! The underside of the C channel is milled out to accommodate the suspension, holes were drilled for where the drive pods rotate, and 8 holes were drilled to bolt the tube and C channel together.
How durable is that setup?
Cody Carey
15-11-2006, 23:03
That appears to be the max size of a drive base according to last years rules, right? For a competition bot, would you scale that design down somewhat in order to add a protective structure around the outside? or would you just rely on the fabric bumpers? It would seem to me that the channel used on the outside would bend/dent easily, and no amount of suspension will keep a Mecanum drive operating correctly if one or two of the wheels are suddenly bent at some odd-angle. The video looks very nice indeed, and I can't wait to see what the drivers can do with a bit of practice :)
Fred Sayre
15-11-2006, 23:09
How durable is that setup?
It hasn't broken yet ;)
The frame seems plenty rigid and sturdy. We have not really ran into anything yet, but with bumpers and such it should be fairly protected. How it deals with the strains of the rest of the weight of the robot we still have to evaluate.
That appears to be the max size of a drive base according to last years rules, right? For a competition bot, would you scale that design down somewhat in order to add a protective structure around the outside? or would you just rely on the fabric bumpers? It would seem to me that the channel used on the outside would bend/dent easily, and no amount of suspension will keep a Mecanum drive operating correctly if one or two of the wheels are suddenly bent at some odd-angle. The video looks very nice indeed, and I can't wait to see what the drivers can do with a bit of practice :)
The chassis is about an inch short of last year's maximums in all directions. If you look at some of our past work, you'll see that I don't typically add structure to things for protection, but since these wheels are quite expensive, this chassis was kept simple to facilitate accepting bumpers. Also consider that it's for practice and won't be seeing competition, so there was no use in wasting time and resources on making it stronger than it needs to be for learning, practice and demonstrations.
I wanna see a Mecanum drive at Portland this year! I could spend hours looking over this robot in the pits.
GMKlenklen
16-11-2006, 14:37
alright, some things to try:
See what effect tightnening the rollers has compared to haveing them loose.
Drive up some ramps.
Drive up some stairs!
Oh, and did you incorperate the suspention system? I can't tell from the picture.
okok, now you gotta try driving it up some stairs like the airtrax did!!! Please? oh, and ramps too...
Btw, I can't quite tell, do you have suspension on it? or did that get left out?
If we have a chance, we'll try running it across all sorts of weird surfaces and objects to see how it behaves. We have our ramp from last season around somewhere and I'm already considering mechanisms that'll allow it to climb curbs and stairs and the like.
Worse comes to worst, we have a reliable design for a simple 6WD drive train waiting in the wings. I can post CAD models of that as well, if folks are interested.
The suspension is implemented, but not quite as we'd originally intended. Because we wanted to get this thing running, we improvised a bit and used both tennis and racquet balls as suspension springs. You can see the bright green and dark blue balls in the video and photos. The tennis balls give a better ride height, but don't compress as easily as the racquet balls, so the jury's still out on which we'd use for competition. More testing is required.
GMKlenklen
16-11-2006, 14:48
oh, that's cool!
Crazy Ivan
16-11-2006, 15:20
Looks awesome! That was cool idea using the racket/tennis balls for the suspension.
Kingofl337
21-11-2006, 18:52
If you want to be able to rotate the base has to be square. Your current base will not rotate well when you load another 100lb on it.
If you want to be able to rotate the base has to be square. Your current base will not rotate well when you load another 100lb on it.
When I'm back after the holiday, we'll add weight and do some more playing around with things to see how it goes. Bringing the wheels closer together so the footprint is square is an easy change, if it's needed, but it's by no means a given method of achieving that discussed result.
Alan Anderson
22-11-2006, 00:25
If you want to be able to rotate the base has to be square. Your current base will not rotate well when you load another 100lb on it.
The required torque to get a Mecanum wheel to move sideways certainly increases with additional weight. I think that can be addressed somewhat by making the rollers spin a little freer.
Why do you think squareness is an issue? The center of mass should probably be as close as practical to the center of the 'bot in order to keep weight equalized on all the wheels, but I think a long or wide rectangle ought to work fine.
viking1902
22-11-2006, 01:26
Why do you think squareness is an issue? The center of mass should probably be as close as practical to the center of the 'bot in order to keep weight equalized on all the wheels, but I think a long or wide rectangle ought to work fine.
I can verify that squareness is not an issue. I have built two holonomic drives on team 1083 in 2004 (http://www.chiefdelphi.com/media/photos/17064) and 2005 (couldn't find a pic). Both drivetrains were rectangular and worked just fine.
Lil' Lavery
22-11-2006, 08:27
I can verify that squareness is not an issue. I have built two holonomic drives on team 1083 in 2004 (http://www.chiefdelphi.com/media/photos/17064) and 2005 (couldn't find a pic). Both drivetrains were rectangular and worked just fine.
116 built a rectangular holonomic drive in 2005 as well, rotation was definately NOT a problem.
Kingofl337
22-11-2006, 09:21
Let me update my post. It doesn't matter if the robot is square but the wheels should be equal distances from each other.
Our test wheelbase was inadvertently square and drove fine. Our final bot was rectangular and had an extremal difficult time rotating. The weight load was similar to our test platform, also WPI had a square robot 2005, and a few other teams with holonomic drives had square robots. None had no problem rotating, leading us to conclude the squareness of the robot effects the rotational ability.
I would recommend making sure you load the system with at least 140lbs throughly testing it before you commit. Rotation is a big part of this drive system.
vBulletin® v3.6.4, Copyright ©2000-2017, Jelsoft Enterprises Ltd.