OCCRA
Go to Post So remember, we may be envious of teams who can do water jet cutting and has every possible monetary and engineering advantage, but there are many more people who are envious of every team who has a chance to compete. - Rombus [more]
Home
Go Back   Chief Delphi > CD-Media > Photos
CD-Media   CD-Spy  
portal register members calendar search Today's Posts Mark Forums Read FAQ rules

photos

papers

everything



swerve in tube

Justin 5553

By: Justin 5553
New: 06-27-2018 09:44 AM
Updated: 06-27-2018 09:44 AM
Views: 1304 times


swerve in tube

Hi everyone,
I've been working on this for a while and finally finished it.

This design has been heavily constrained by the limited capacities of my team in regular machining. Except for some tapped shafts, this can all be done with laser cutting, to which we have great access.

Driving gear ratio: 18:60
Module spinning ratio : 70:1
Wheel diameter : 2.5 in
Max "real life" speed : 14.1 ft/s

I use the Cimcoder (am-3314a) for the drive and the VP integrated encoder (217-5046) for the module control.

I would love some feedback !

Recent Viewers

Discussion

view entire thread

Reply

06-27-2018 11:48 AM

ClayTownR


Unread Re: pic: swerve in tube

I have relatively little experience with laser cutting, and no experience with cutting aluminum, so I thought Id ask: When you laser cut the tube, would you cut one face, then flip to the opposite side to cut the other one, or can the laser do both faces at once?

As a slightly more related question, what are your plans for mounting this on a frame?



06-27-2018 11:53 AM

Skyehawk


Unread Re: pic: swerve in tube

Huh, interesting. I'm impressed. Are the swerve forks just a piece of C-channel?



06-27-2018 11:53 AM

Bryce2471


Unread Re: pic: swerve in tube

The design looks great. I like the packaging a lot.

Some questions and suggestions:
I would recommend using a bevel gear with more than 1:1 gearing in the wheel module.
It's hard to tell from the picture how you are supporting the thrust and moment loads from the wheel, so I can't comment on that.
If you are planning to use an integrated versaplanetary encoder for the steering, I think you'll want your belt ratio after the gearbox to be 1:1.
How close is the bag motor belt to the floor? I would put a 3d printed cover over it, so it doesn't catch carpet debris.



06-27-2018 01:13 PM

GeeTwo


Unread Re: pic: swerve in tube

Quote:
Originally Posted by ClayTownR View Post
As a slightly more related question, what are your plans for mounting this on a frame?
Usually "drive in tube" means that the drive is mounted in the tube which IS the frame. That is, this is the last foot or so of a frame structural member.

Quote:
Originally Posted by Skyehawk View Post
Are the swerve forks just a piece of C-channel?
They sure do look like a piece of the tube, cut open on one face.

Quote:
Originally Posted by Bryce2471 View Post
How close is the bag motor belt to the floor? I would put a 3d printed cover over it, so it doesn't catch carpet debris.
+1 on this. What is the ground clearance, and what is the ground clearance angle (angle from the contact patch of the wheel to an obstruction, measured from the floor)? These are critical if there are any bumps on the floor, or are likely to be small items (e.g. game pieces) which can get under there and leave the wheels spinning in air.

What are the tube dimensions? Given the width of the tube compared to the CIM it looks like 4"x2", with about an 0.2" wall. Heavy.



06-27-2018 02:10 PM

Jon Stratis


Unread Re: pic: swerve in tube

With a little modification, I think you could overcome the ground clearance issue. If you updated the lightening triangles on the side, you could probably move the large pulley inside the tube, against the top. That would let you flip the CIM motor over, mounting it to the top of the tube sticking up. Similarly, you could flip the bag motor assembly over and mount it to the top of the tube, running the shaft through the middle of the belting for the CIM motor and down to a matching bearing in the bottom of the tube for a little additional support, leaving the pulley against the bottom of the tubing as it is now. That would effectively put all of the motors and gearboxes above the tube, with the only thing sticking down being the wheel assembly.

As a possible benefit to this change, you could potentially get away with smaller tubing. You would have the space to add in a second reduction on the drive motor and could switch to smaller pulleys. And since you won't be constrained by the diameter of the CIM motor, you could maybe get it down as far as 1"x2", although that might be pushing it. 1.5"x2" is probably more realistic.

Another thing to keep in mind when looking at this... bumpers. As it stands now, I believe the wheel assembly would hit any bumpers mounted to the side of the tube, unless the bumper height was sufficient - in recent years it hasn't been.



06-27-2018 02:43 PM

Chris is me


Unread Re: pic: swerve in tube

Does your sponsor specifically have access to a tube laser? It is a specialty piece of equipment that can cut through and rotate tubes to cut on all faces. If they do, that's fantastic, tube lasers are the coolest thing and you should consider yourself lucky. If they don't, then they're not going to be able to cut any of the tube or channel parts you are using.



06-28-2018 04:29 AM

Justin 5553


Unread Re: pic: swerve in tube

Thank you all for the feedback !

Quote:
Originally posted by ClayTownR
When you laser cut the tube, would you cut one face, then flip to the opposite side to cut the other one, or can the laser do both faces at once?
It cuts one face at a time, then flip to the other, and this allows a lot more freedom in design.

Quote:
Originally posted by ClayTownR
As a slightly more related question, what are your plans for mounting this on a frame?
Well, I could use that big aluminum extrusion as the frame, as suggested by GeeTwo, but it would be quite heavy, so maybe I will add some holes on this extrusion so as to mount it on a more classical frame (2x2x0.1in or 50x25x2mm)

Quote:
Originally posted by Skyehawk
Are the swerve forks just a piece of C-channel?
Yes !

Quote:
Originally posted by Bryce2471
I would recommend using a bevel gear with more than 1:1 gearing in the wheel module.
That could indeed reduce the width of the wheel module but I can't find a set of gears other than the one VEXPro sells that can be mounted on 1/2in hex shafts. So it would require a lathe or mill to make a more fancy shaft, which I don't have. That's the main reason why I have kept the gears sold by VEX.

But it would shorten the width of the wheel module and allow for a smaller reduction previous to the module (which is not really a needed improvement given the small diameter of the wheels).

Quote:
Originally posted by Bryce2471
It's hard to tell from the picture how you are supporting the thrust and moment loads from the wheel, so I can't comment on that.
I use a bearing flushed in the extrusion, this bearing is held by screws and nuts. These screws and nuts are inserted in holes around the bearing hole and the washers and screwheads get over the external ring of the bearing. (is that clear )

Quote:
Originally posted by Bryce2471
If you are planning to use an integrated versaplanetary encoder for the steering, I think you'll want your belt ratio after the gearbox to be 1:1.
I've read this multiple times but never understood why, what's the reason? (I have very little knowledge about encoders and code).

Quote:
Originally posted by Bryce2471
How close is the bag motor belt to the floor? I would put a 3d printed cover over it, so it doesn't catch carpet debris.
8.2mm, I will add a cover !

Quote:
Originally posted by GeeTwo
what is the ground clearance angle?
4.8 That's small, so this is definitely not good for games like PowerUp or StrongHold but it would be okay for Steamworks or AerialAssist I think.

Quote:
Originally posted by GeeTwo
What are the tube dimensions? Given the width of the tube compared to the CIM it looks like 4"x2", with about an 0.2" wall. Heavy.
100x60x4mm so appoximately 3.9x2.5x0.16in (I'm French so metric system reigns )
It weighs 412g (0.9lbs) according to Inventor.

Quote:
Originally posted by Jon Stratis
With a little modification, I think you could overcome the ground clearance issue. If you updated the lightening triangles on the side, you could probably move the large pulley inside the tube, against the top. That would let you flip the CIM motor over, mounting it to the top of the tube sticking up. Similarly, you could flip the bag motor assembly over and mount it to the top of the tube, running the shaft through the middle of the belting for the CIM motor and down to a matching bearing in the bottom of the tube for a little additional support, leaving the pulley against the bottom of the tubing as it is now. That would effectively put all of the motors and gearboxes above the tube, with the only thing sticking down being the wheel assembly.
I'm not sure to understand everything, but I'm scared that this would make the swerve a lot less compact, since a lot of space in the tube is lost, especially under the CIM motor. The ground clearance is an issue for some games but for lot of others, I think the current clearance is big enough (with a cover to protect from debris).
Yet, if this was to be used in PowerUp or even RecycleRush, you idea would be very interesting.
I think I will anyway make a V2 with a better ground clearance and a better use of space. I will aslo modify the wheel module spinning system, I'm not really satisfied with it.

Quote:
Originally posted by Jon Stratis
Another thing to keep in mind when looking at this... bumpers. As it stands now, I believe the wheel assembly would hit any bumpers mounted to the side of the tube, unless the bumper height was sufficient - in recent years it hasn't been
You're totally right on this, I hadn't the bumpers in mind. On the V2, I will choose a way to mount this on a frame in a way that bumpers can be mounted at the same height they were during the recent years.

Quote:
Originally posted by Chris is me
Does your sponsor specifically have access to a tube laser? It is a specialty piece of equipment that can cut through and rotate tubes to cut on all faces. If they do, that's fantastic, tube lasers are the coolest thing and you should consider yourself lucky. If they don't, then they're not going to be able to cut any of the tube or channel parts you are using
Yes, yes, yes and yes !
We've been working with them for now two years and this is a great opportunity for my team, they have great capacities in laser cutting on sheet and tubes and are very flexible.



06-28-2018 08:19 AM

AriMB


Unread Re: pic: swerve in tube

Quote:
Originally Posted by Justin 5553 View Post
I've read this multiple times but never understood why, what's the reason? (I have very little knowledge about encoders and code).
To put it simply, an absolute encoder tells you what angle its shaft is pointed at, from 0-360. If the encoder is geared 1:1 with the module, the angle from the sensor will correspond directly with the angle from the module. If it is geared up (2:1 for example), the sensor will read 0 both when the module is pointed at 0 and 180. Because of this, when the robot starts up, there is no way of knowing exactly which direction the module is pointed. You can get around this by adding in a separate zeroing sensor that only triggers once per revolution of the module and running a zeroing routine at the beginning of auton every match, but that adds (generally) unnecessary complexity.



06-28-2018 01:32 PM

Justin 5553


Unread Re: pic: swerve in tube

Quote:
Originally Posted by AriMB View Post
To put it simply, an absolute encoder tells you what angle its shaft is pointed at, from 0-360. If the encoder is geared 1:1 with the module, the angle from the sensor will correspond directly with the angle from the module. If it is geared up (2:1 for example), the sensor will read 0 both when the module is pointed at 0 and 180. Because of this, when the robot starts up, there is no way of knowing exactly which direction the module is pointed. You can get around this by adding in a separate zeroing sensor that only triggers once per revolution of the module and running a zeroing routine at the beginning of auton every match, but that adds (generally) unnecessary complexity.
Ok thank you !
all encoders are absolute encoders ? can't they do both? (giving movement and position)



06-28-2018 01:49 PM

GeeTwo


Unread Re: pic: swerve in tube

Quote:
Originally Posted by Justin 5553 View Post
Ok thank you !
all encoders are absolute encoders ? can't they do both? (giving movement and position)
No, not all encoders are absolute encoders, but these are preferred for systems (e.g. steering or arms) where the specific angle that the device is oriented is the desired quantity. If only relative angle (e.g. a drive wheel) is needed, a relative encoder (e.g. quadrature encoded) is usually used. Use of reduction after the encoder means that the encoder is effectively a relative encoder, even if it is an absolute encoder.



view entire thread

Reply
previous
next

Tags

loading ...



All times are GMT -5. The time now is 08:19 PM.

The Chief Delphi Forums are sponsored by Innovation First International, Inc.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi