|
|
|
 Sloppy thinking leads to sloppy designs...
- Joe Johnson [more] |
 |
|
|||||||
|
||||||||
This is a offseason project I worked on much earlier this year, just as a personal CAD project. I wanted to wait until I had 1 real module to post, but Kickoff is creeping up, so I'll post that when I have one.
While I liked the very compact and Lightweight setup of the CIM in wheel, I didn't want to deal with the wires ever again, and would rather avoid mercury on the robot (sliprings). This lead to the fun mental exercise of "if I were to make another swerve, what would it be?"
This module is the answer.
Showed this to Kevin on 2451 and he took off running and now has a fancy 2 speed version.
This was purely for fun, not a VEXpro item, sorry ;-)
Will type up specs sometime and post here.
Discussion
view entire thread
22-12-2014 13:30
Re: pic: Swerve again, for fun
So when my poem said a new swerve from Aren Hill, it wasn't kidding!
Looks great! What is it geared for? I can't tell from the section view, but it looks like it's cutting close on the clearance for the wheel. Do you know off hand the amount of clearance it is?
22-12-2014 13:47
Ether
Re: pic: Swerve again, for fun
|
Originally Posted by Aren_Hill
...would rather avoid mercury on the robot (sliprings).
|
22-12-2014 14:08
Bryce2471
Re: pic: Swerve again, for fun
A few questions immediately pop up for me:
How is the CIM shaft so short? Isn't it against last year's rules to modify motors?
Where would you get the bevel gears, and how much would they cost?
What's the weight?
Other than that, it looks very good.
22-12-2014 14:13
JesseK
Re: pic: Swerve again, for fun
From the cut-out in the other picture, I'm having a hard time figuring out how the large turning gear is mated to the grey-ish ring that acts as the weight-bearing thrust bearing for for the entire robot. It seems like all of the robot's weight is hanging from this ring which is hanging from the gear?
22-12-2014 14:14
Jared Russell
Re: pic: Swerve again, for fun
|
Originally Posted by Ether
Are there not suitable sliprings for this application which contain no mercury?
|
22-12-2014 14:40
Andrew Schreiber
Re: pic: Swerve again, for fun
|
Originally Posted by Jared Russell
A few years back I found it very difficult/impossible to find a non-Mercotac COTS slip ring that was simultaneously (a) rated for FRC current levels and (b) within single-components COTS cost limits. It is possible that you could make your own spring-loaded slip ring, but YMMV with inspectors/rules in a given season.
|
The Mercotac ones are far lighter though.
22-12-2014 14:43
orangemoore
Re: pic: Swerve again, for fun
|
Originally Posted by Aren_Hill
This was purely for fun, not a VEXpro item, sorry ;-)
|
22-12-2014 14:54
tim-tim
Re: pic: Swerve again, for fun
|
Originally Posted by Bryce2471
How is the CIM shaft so short? Isn't it against last year's rules to modify motors?
|
22-12-2014 14:56
Thad House
Re: pic: Swerve again, for fun
|
Originally Posted by tim-tim
By that definition, wouldn't it be illegal to cut/strip CIM motor wires to add connectors of your choice on the ends?
|
22-12-2014 15:04
Kevin Ainsworth
Re: pic: Swerve again, for fun
It has to be the most compact and lightest swerve to date!
Professional grade designing, I looked through the entire design earlier this year and thought it was brilliant. We tried to design something totally unique but nothing we came up with was as elegant. Thanks for the inspiration!
22-12-2014 15:14
Bryce2471
Re: pic: Swerve again, for fun
|
Originally Posted by Kevin Ainsworth
It has to be the most compact and lightest swerve to date!
|
22-12-2014 15:19
Bennett548
Re: pic: Swerve again, for fun
Is the weight of the robot sitting on the cim pinion? I can't figure out the wheel pivot bushing/bearing for this design.
Edit: I see the button heads now. Is that a silverthin bearing?
22-12-2014 15:35
nuclearnerd
Re: pic: Swerve again, for fun
It's compact for sure, but I can't figure out the bearing design.
As far as I can tell, all of the vertical loads are taken by the CIM shaft (!), and I can't really tell what takes the horizontal loads. There's an upper steering bearing on the CIM shaft, but the only lower bearing I can see is the white ring below the steering gear and that doesn't have much radial contact. What am I missing?
22-12-2014 16:40
asid61
Re: pic: Swerve again, for fun
|
Originally Posted by Bryce2471
What makes you say that? Do you know the weight?
|
Plus, making it into a shifting version requires a lot of weight unless you don't use a COTS shifter shaft.
EDIT: Of course, if anybody could make the design lighter it would be Aren. So it could be lighter actually.
Very nice swerve Aren. It's cool seeing your drives. Are you using a dead axle? Is it possible to flip the cim?
22-12-2014 17:05
JesseK
Re: pic: Swerve again, for fun
|
Originally Posted by Bryce2471
What makes you say that? Do you know the weight?
|
We can presume that Kevin from 2451 has a pretty good guess for how much it weighs when he says
| It has to be the most compact and lightest swerve to date! |
|
Originally Posted by Aren_Hill
Showed this to Kevin on 2451 and he took off running and now has a fancy 2 speed version.
|
22-12-2014 17:43
Cash4587
Re: pic: Swerve again, for fun
Just curious, is there any way we can get our hands on the CAD files to take a closer look at the module? I know this may be a stretch but worth a shot too: If a team wanted to build this, would the spec sheet include all the parts that come together to make this?
22-12-2014 17:57
AdamHeard
Re: pic: Swerve again, for fun
|
Originally Posted by asid61
+1. I have a swerve design exactly like this on my computer, except with a 2" colson instead of a 3.25" versawheel, ad it weighs a minimum of 5.3lbs. Still heavier than Bryce's swerve, albeit only very slightly.
Plus, making it into a shifting version requires a lot of weight unless you don't use a COTS shifter shaft. EDIT: Of course, if anybody could make the design lighter it would be Aren. So it could be lighter actually. Very nice swerve Aren. It's cool seeing your drives. Are you using a dead axle? Is it possible to flip the cim? |
I could go design a 4 lb swerve now, no guarantee it will work or hold up. Many of the recent designs posted aren't robust enough to survive a season (I think Aren's here would though, as he's fielded 5+ swerves at this point I think).
22-12-2014 18:08
Kevin Ainsworth
Re: pic: Swerve again, for fun
|
Originally Posted by AdamHeard
I don't think theoretical weights are fair to compare. If the system isn't made and working, it doesn't matter how light it is.
I could go design a 4 lb swerve now, no guarantee it will work or hold up. Many of the recent designs posted aren't robust enough to survive a season (I think Aren's here would though, as he's fielded 5+ swerves at this point I think). |
22-12-2014 19:51
Tyler2517
Re: pic: Swerve again, for fun
I have to ask were some of the concepts for this taken from my teams designs i posted over the summer? Looks really clean and well thought out.
http://imgur.com/a/H6Qcm
22-12-2014 22:50
asid61
Re: pic: Swerve again, for fun
|
Originally Posted by AdamHeard
I don't think theoretical weights are fair to compare. If the system isn't made and working, it doesn't matter how light it is.
I could go design a 4 lb swerve now, no guarantee it will work or hold up. Many of the recent designs posted aren't robust enough to survive a season (I think Aren's here would though, as he's fielded 5+ swerves at this point I think). |
Given that, I think it's possible that this weighs more than many designs posted previously. However, it is much more compact, and for many situations that can be more valued than the weight.
If you believed that none of the recent designs can survive the season, then you really should have said so ealier, when those designs were posted.
Also, I found this yesterday:
http://www.chiefdelphi.com/media/photos/34178
So Aren beat himself some years ago.
EDIT: How is the bevel gear constrained axially on the shaft? Does it depend on the wheel to position itself properly?
23-12-2014 08:43
Gdeaver
Re: pic: Swerve again, for fun
This years game was hard on our swerve modules but they are still going strong. I would pick a robust module over a cute light weight module any day. So I say to those who have caded all those modules in the off season, Make them and beat them. Do they perform and survive? Then it is a good design. Until then it's just a concept. We made a CVT module this off season. Do we do it this year? I don't know. I'm leery of adding complexity to an already complex system. I guess it depends on the game. But, we made it and tested it. Made revisions and beat it. First has become brutal. Will 2015 continue this trend?
23-12-2014 11:08
Kevin Ainsworth
Re: pic: Swerve again, for fun
|
Originally Posted by asid61
We're looking at a theoretical model right now, so it makes sense to compare theoretical weights. Kevin's theoretical weight for his shifting design posted a few months ago was ~7lbs IIRC, whereas Bryce's models for shifting designs weighed around 6.3lbs. Even if it wouldn't survive a season as in the CAD, there's at least 0.5lbs to play with to buff it up. And Kevin's design used only small steel miter gears, whereas this design uses much larger steel bevel gears.
Given that, I think it's possible that this weighs more than many designs posted previously. However, it is much more compact, and for many situations that can be more valued than the weight. If you believed that none of the recent designs can survive the season, then you really should have said so ealier, when those designs were posted. Also, I found this yesterday: http://www.chiefdelphi.com/media/photos/34178 So Aren beat himself some years ago. EDIT: How is the bevel gear constrained axially on the shaft? Does it depend on the wheel to position itself properly? |
The larger gear in Aren's design might be larger but the smaller gear is much smaller. I would go off tooth size and not OD of the largest gear.
Our actual module weighs in at 7.5 lbs with no additional lightening of the gears, etc. This is too be expected since all the steel screws weren't added to the CAD model. We'll post a video of it driving around just as soon as it is, should have it powered up this afternoon.
23-12-2014 14:39
asid61
Re: pic: Swerve again, for fun
|
Originally Posted by Kevin Ainsworth
The larger gear in Aren's design might be larger but the smaller gear is much smaller. I would go off tooth size and not OD of the largest gear.
Our actual module weighs in at 7.5 lbs with no additional lightening of the gears, etc. This is too be expected since all the steel screws weren't added to the CAD model. We'll post a video of it driving around just as soon as it is, should have it powered up this afternoon. |
The weight comes not only from the teeth but the face of the gear I think, as well as the hub. As the diameter of the gear increases the area/ volume of the gear increases considerably.
You actually made your module? Super cool! I'll be looking forward to seeing it driving.
23-12-2014 15:23
Kyler Hagler
Re: pic: Swerve again, for fun
Great looking swerve Aren!
When you get it working Kevin, it would be awesome to see a video of it working!
23-12-2014 17:48
Dillon Carey
Re: pic: Swerve again, for fun
|
Originally Posted by asid61
Also, I found this yesterday: http://www.chiefdelphi.com/media/photos/34178 So Aren beat himself some years ago. |
Example: I led the design on this module as well as the 2010 6 wheel swerve.
23-12-2014 19:51
Aren_Hill
Re: pic: Swerve again, for fun
|
Originally Posted by Dillon Carey
Not everything that 1625 has ever done was designed by Aren, many things were, but his senior year on our team was 2008. He helped out up till about 2011, he moved on to help 3928 and 148 after. He did a lot for our team, but their were (and are) a lot of talented people who deserve credit for 1625s success.
Example: I led the design on this module as well as the 2010 6 wheel swerve. |
(And if I made that tiny module, the lightening pattern would've been way cooler looking
)
24-12-2014 03:16
asid61
Re: pic: Swerve again, for fun
|
Originally Posted by Dillon Carey
Not everything that 1625 has ever done was designed by Aren, many things were, but his senior year on our team was 2008. He helped out up till about 2011, he moved on to help 3928 and 148 after. He did a lot for our team, but their were (and are) a lot of talented people who deserve credit for 1625s success.
Example: I led the design on this module as well as the 2010 6 wheel swerve. |
How did 6 wheel swerve work out?
24-12-2014 03:22
pwnageNick
Re: pic: Swerve again, for fun
|
Originally Posted by asid61
OMG I'm so sorry. I didn't double check the author.
How did 6 wheel swerve work out? |
also:
Finalists at GKC along with Engineering Excellence,
Semi-Finalists at Midwest along with Regional Chairman's
Galileo Division Champions
Never ranked lower than 3rd after quals at any regional or champs, overall season record of 36-13-1
sooooooo...
*Fun fact, only 6-Wheel Swerve to ever be on Einstein
24-12-2014 14:24
Chris is me
Re: pic: Swerve again, for fun
Aren never ceases to break my head with his designs. The non-coaxial power transfer is so cool. I like how you use the 16-style bearing thing to avoid using any shafting at all holding the module up. I like how your crazy ideas make complete sense in terms of the problems you're trying to solve.
It doesn't look like you have any sensor feedback on the wheel (velocity, you have a thing for angle i think). Do you have something planned for that? This seems like a cool application for that on-CIM encoder that's been going around CD these days.
| If you believed that none of the recent designs can survive the season, then you really should have said so ealier, when those designs were posted. |
24-12-2014 14:46
AdamHeard
Re: pic: Swerve again, for fun
|
Originally Posted by Chris is me
He's not objecting to the designs, he's objecting to comparing the weight of untested CAD models. He had no reason to make this point until this happened, so I don't think it's a fair criticism.
|
Good design is focusing on eliminating items before pocketing them. Design for manufacture is key.
I didn't comment as I've learned to avoid (or try to...) providing criticism to people in that stage, as they generally don't listen to it anyway.
28-12-2014 17:24
Aren_Hill
Re: pic: Swerve again, for fun
|
Originally Posted by Chris is me
Aren never ceases to break my head with his designs. The non-coaxial power transfer is so cool. I like how you use the 16-style bearing thing to avoid using any shafting at all holding the module up. I like how your crazy ideas make complete sense in terms of the problems you're trying to solve.
It doesn't look like you have any sensor feedback on the wheel (velocity, you have a thing for angle i think). Do you have something planned for that? This seems like a cool application for that on-CIM encoder that's been going around CD these days. |
The module support is a 1/4" section silverthin bearing, the forks that hold the main wheel dead axle overlap the inner race of the bearing and clamp it. This is what handles the weight of the robot. No more custom bearing 16 style stuff, just a big COTS X-contact awesome bearing.
And that was my first thought after seeing that nifty CIM-encoder setup Chris, it'd work quite well as I currently have to cut the CIM shaft anyway.
This is by far the "most legitimate" swerve I have designed, and no pocketing etc has been done, I am confident it would last through a season and then some, while still being easy to work on.
Fun Fact, it only has 1 nut on it, and I could remove that one as well if I really wanted to. (Wheel dead axle)
-Aren
view entire thread
Tags
loading ...