Stingray Robot Kit Changes (#28980)

Comments

• erco Posts: 20,250

  2011-10-05 16:49 edited 2011-10-05 16:49

  How about lower-geared motors, at least 100:1 or lower?
• william chan Posts: 1,326

  2011-10-06 01:51 edited 2011-10-06 01:51

  Can EDDIE board be used on this new Stingray kit?
• Mike G Posts: 2,702

  2011-10-06 07:04 edited 2011-10-06 07:04

  I was kina thinking on the same lines... if you made the chassis more like EDDIE's manufacturing cost would be less. Plus there could be a stackable platform as apposed to a box.
• Chris Savage Parallax Engineering Posts: 14,406

  2011-10-07 07:40 edited 2011-10-07 07:40

  erco wrote: »

  How about lower-geared motors, at least 100:1 or lower?

  
  erco, the motors used on the Stingray are compatible with many similar motors sold by various vendors. Trust me when I say, 100:1 would be way too slow...probably about BOE-Bot speed. 50:1 is what the original spec was and these motors had good torque and speed. The current motors are 30:1.
• Chris Savage Parallax Engineering Posts: 14,406

  2011-10-07 07:43 edited 2011-10-07 07:43

  william chan wrote: »

  Can EDDIE board be used on this new Stingray kit?

  
  I suppose the EDDIE board could be used, but the whole point of opening up the Chassis Kit was to give customers more choices. THere were many who use non-Parallax boards who wanted access and not be forced to buy our board. This gives them that opportunity. So many customers that provided feedback on the MSR1 had various ways of configuring the boards. All these ideas were good, but that many variations was just not possible. Now the customer can configure things exactly as they want.
• Chris Savage Parallax Engineering Posts: 14,406

  2011-10-07 07:45 edited 2011-10-07 07:45

  Mike G wrote: »

  I was kina thinking on the same lines... if you made the chassis more like EDDIE's manufacturing cost would be less. Plus there could be a stackable platform as apposed to a box.

  
  Let's remember, the Stingray platform is smaller and yet faster. Compared to EDDIE there really isn't any comparison. The Stingray platform has its place as the BOE-Bot and EDDIE do. We'll see where customers take it in the near future and re-evaluate things down the road as always, based on customer feedback.
• erco Posts: 20,250

  2011-10-07 08:43 edited 2011-10-07 08:43

  Chris Savage wrote: »

  erco, the motors used on the Stingray are compatible with many similar motors sold by various vendors. Trust me when I say, 100:1 would be way too slow...probably about BOE-Bot speed. 50:1 is what the original spec was and these motors had good torque and speed. The current motors are 30:10.

  
  Maximum speed is usually neither good nor useful in robotics; control is the goal. The current motors are definitely undergeared and uncontrollably fast. Search forum posts for "baseboard damage" or "jumped off workbench". I stand by my assertion that 100:1 is an appropriate ratio for useful robotic applications. Check out the Youtube videos on Stingray. There isn't any precision maneuvering to be seen (roaming with PINGs isn't precision, it's random obstacle avoidance, and BoeBot does it equally well), mainly just people driving it around like a radio controlled toy. Speed is OK for that limited use only, but wouldn't Parallax like to see something more sophisticated? As long as you're open-sourcing the Stingray design and letting people buy what they want, why not offer 100:1 motors and give them the choice?
  
  I tried unsuccessfully to obtain some from the manufacturer. If you have some 100:1 samples, send me some, and I'll put some encoders on a Stingray and show you some maneuvers.
• Duane Degn Posts: 10,588

  2011-10-07 14:04 edited 2011-10-07 14:04

  @Chris,
  
  This is great news. I think I have most of what I need to complete a Stingray Chassis Kit. For $149 dollar, I'm likely to buy one. I do wonder about encoders? I remember Kye suggested some that work great with the Stingray motors.
  
  @erco, can't you just slow the speed of the motors with PWM? Do you plan to carry a heavy load or travel up steep inclines with the extra torque? I hope you don't just need slowed down motors so you can use them with your relays. Don't get me wrong, I like relays (and wood) but relays aren't very good when it comes to precision control using PWM (although I do this to control my lab ovens).
  
  Duane
  
  Edit: I added a link to Kye's suggestion.
• erco Posts: 20,250

  2011-10-08 09:43 edited 2011-10-08 09:43

  No intended relay use, the motors are definitely undergeared and PWM is a poor bandaid. Fine, slow motor control from a stopped position is nearly impossible. All the electronic trickery in the world can't make up for low torque output from undergeared motors.
  
  Try 'em and get back to me with your thoughts. I've seen lots of posts about adding encoders to Stingray, yet not a single example of fine motor control exists. No posts, no videos. Scribbler 2 is 1000X better at motion control. Can a Stingray move accurately enough to write a single legible letter? With all that sophistication, shouldn't it be able to? And no, letters I and O don't count.
• Duane Degn Posts: 10,588

  2011-10-08 10:27 edited 2011-10-08 10:27

  erco,
  
  What about these motors?
  
  I would want to be able to add encoders to whatever motors I used. And I do care about low speed control. It would be nice if there were a gear/motor combination that allowed fine low speed control and also allowed for high-speed remote-controlled hot-roding.
  
  I assume Chris' "30:10" ratio was a typo and should be "30:1"?
  
  Duane
• erco Posts: 20,250

  2011-10-08 14:46 edited 2011-10-08 14:46

  Those look similar, they may or may not bolt up. If you're gonna pay $40+ to try other motors, don't waste the money on 50:1 units. Go for at least 80:1 to make it worth your while, and I still say 100:1 is preferable. I contacted the mfr of the Stringray motors long ago (Chien Shing, or something close) and they definitely have a variety of motors, including over 100:1. But they would not supply me with any samples. Pity, I think Stingray could be a useful platform with different motors.
• Roy Eltham Posts: 2,996

  2011-10-08 18:41 edited 2011-10-08 18:41

  I think with 50:1 motors it might be usable. I've ordered a couple of those Lynxmotion 50:1 motors with encoders. I'll put them on my Stingray and try it out.
  The 30:1 ones are just a little on the weak side at low speed which causes it to sort of stall randomly when going slow, and at high speeds it's harder to control (overshooting a lot), which makes it unable to do good dead reckoning.
  With 50:1 I think I could make it work well enough. Not as good as with 100:1, but still usable.
• erco Posts: 20,250

  2011-10-08 19:32 edited 2011-10-08 19:32

  Roy, you're an excellent peacemaker & diplomat. Thanks for hearing, I'll stop my diatribe now.
• Roy Eltham Posts: 2,996

  2011-10-08 20:11 edited 2011-10-08 20:11

  erco, I just enjoy trying to do things that are hard. I don't always succeed, but I don't mind that.
  I found that pololu also has some potentially usable motors with encoders that come in lots of ratios (including 67:1, 100:1, & 131:1). They have 6mm shafts so they can be used with the wheels that come with the Stingray, but I might have to drill some extra holes in my Stingray's motor mount plates to get them working. I'll try those later if I can't make due with the 50:1 ones from Lynxmotion.
  
  I will try to make it work and post results for you. Maybe we can convince Parallax/Ken to change some thing in the Stingray in the long run. I like a lot of things about the Stingray platform, the size, the looks, the ease of mounting things to it. The only real problem I have is the 30:1 motors. So I want to help make it better...
  
  It would be cool to pair it with that new Eddie controller board too.
• erco Posts: 20,250

  2011-10-09 09:47 edited 2011-10-09 09:47

  Roy Eltham wrote: »

  It would be cool to pair it with that new Eddie controller board too.

  
  @Roy: You were an early adopter of Ken's original Plywood bot, and did some excellent work on it. Since then, that bot has morphed into MadeUSA and Eddie, but you must always remember your robot's proud cellulose origins. Don't be seduced by the Matt side!
  
  I wish the whole Eddie effort well, and will stir up interest in robotics. But I'm banking that the next wave of successful consumer robots will be continue to be very simple; no MSRS, no PC.
• Roy Eltham Posts: 2,996

  2011-10-09 13:01 edited 2011-10-09 13:01

  I wasn't planning to use MSRS or a PC on the bot (I'll use the PC to download code to the bot as I do now). It's just that the Eddie controller board looks like a really nice Prop based robot controller, much better than the MSR1.
  My understanding is that you can use Eddie just fine without the PC laptop onboard or MSRS, it just becomes difficult to interface with the Kinect, but I don't mind that.
  
  I think you and I are more on the same page than not, erco.
• erco Posts: 20,250

  2011-10-09 14:01 edited 2011-10-09 14:01

  Roy Eltham wrote: »

  I think you and I are more on the same page than not, erco.

  
  Amen, Brother! BTW, I keep meaning to make a pilgrimmage down to Oceanside to meet Gordon McComb & Tom Atwood (ROBOT mag editor). I bet you would drive up for a long lunch meeting with those gents, huh? Might be fun to brainstorm some projects together.
• Roy Eltham Posts: 2,996

  2011-10-09 23:33 edited 2011-10-09 23:33

  Yeah, Oceanside is not far from me... Maybe 30 minutes drive depending on traffic of course...
• Whit Posts: 4,191

  2011-10-10 18:58 edited 2011-10-10 18:58

  The drive would be a bit further from Louisiana...
  
  That doesn't mean I wouldn't like to be there!
• Ron Czapala Posts: 2,418

  2011-10-10 20:56 edited 2011-10-10 20:56

  erco wrote: »

  Those look similar, they may or may not bolt up. If you're gonna pay $40+ to try other motors, don't waste the money on 50:1 units. Go for at least 80:1 to make it worth your while, and I still say 100:1 is preferable. I contacted the mfr of the Stringray motors long ago (Chien Shing, or something close) and they definitely have a variety of motors, including over 100:1. But they would not supply me with any samples. Pity, I think Stingray could be a useful platform with different motors.

  
  I used 50:1 Pololu motors with encoders on my homemade stingray and got pretty good control but I was doing 50% PWM. (see video)
  
  100:1 motors would probably be an improvement but I'm not going to spend another $80 to swap them out...
• Zoot Posts: 2,227

  2011-10-10 21:11 edited 2011-10-10 21:11

  I've seen lots of posts about adding encoders to Stingray, yet not a single example of fine motor control exists. No posts, no videos.

  
  These are *not* my projects, btw.
  

  
• erco Posts: 20,250

  2011-10-10 23:36 edited 2011-10-10 23:36

  @Zoot: Fun videos to watch, for sure. They fairly reinforce my point about lacking fine motor control for navigation. There is noticeable oscillation & overshoot in the soda can pickup videos, even doing its simple line following routine, which is not real navigation. The center vid is Ron's (2 posts back) homemade bot, using 50:1 gearboxes (stock stingray is currently 30:1). He does note that 100:1 would be an improvement.
  
  I'll pit my simple BS2/relay/properly-geared Retrobot against allcomers with 30:1 Propeller Stingrays in accuracy/repeatability contests: http://www.youtube.com/watch?v=PX0IhUqnwrk
  
  My whole point is to say that lower gearing (and more torque) makes any system more controllable. Higher max speed is not a virtue when it sacrifices torque and control.
• Chris Savage Parallax Engineering Posts: 14,406

  2011-10-12 11:24 edited 2011-10-12 11:24

  erco wrote: »

  I stand by my assertion that 100:1 is an appropriate ratio for useful robotic applications. As long as you're open-sourcing the Stingray design and letting people buy what they want, why not offer 100:1 motors and give them the choice?
  
  I tried unsuccessfully to obtain some from the manufacturer. If you have some 100:1 samples, send me some, and I'll put some encoders on a Stingray and show you some maneuvers.

  
  erco,
  
  What I failed to mention when I said the 100:1 motors would be too slow was that the original Stingray design I was mentioning also had smaller whells. The increased speed motors along with the increase in wheel size compounded the speed vs torque problems.
  
  50:1 motors with the original wheels would be perfect! I know some like the larger wheels because or carpet, so perhaps there is a compromise. I will look into it further.
• Chris Savage Parallax Engineering Posts: 14,406

  2011-10-12 11:26 edited 2011-10-12 11:26

  Duane Degn wrote: »

  I assume Chris' "30:10" ratio was a typo and should be "30:1"?
  Duane

  
  Typo corrected...yes, 30:1
• Mr. Lewis Posts: 24

  2011-10-14 08:46 edited 2011-10-14 08:46

  I'm a high school science teacher that has used BOE-Bots, Sumo-Bots, and some accessories in my class. I just purchased my second Stingray over the summer and can't tell you how excited my students are about trying to program them and compete with some mini-challenges.
  
  That said, for them (and myself) the predesigned Stingray was a great option, they can just plug in components and test them out without having to build a whole board and it allowed them some application from the propeller education kit. Is there any possibility you will either A) continue manufacturing/selling the propeller boards used for it, or consider developing an education robotics kit based off of this platform (either Stingray or propeller BOE)? I am teaching myself and my students how to use the propeller as we have time and I have found that this kit was great for an easily accessible advanced robot (plus it moves much faster which the students love).
  
  Thanks for putting together such great products and designing educational materials for them - I have about one of every educational kit and allow my advanced students to choose what/how they study based off of these.
• Roy Eltham Posts: 2,996

  2011-10-15 00:10 edited 2011-10-15 00:10

  I received my LynxMotion order today with the 50:1 motors and some encoders. I'll be trying those out with the stock wheels from the Stingray this weekend, but I'm also going to try and find some smaller wheels that'll work with the Stingray to try in the future.
• charleyshf Posts: 165

  2011-10-15 04:53 edited 2011-10-15 04:53

  @Roy, for the different size wheels, take a look at http://www.trossenrobotics.com/c/robot-wheel-hubs-castors.aspx , those are the same brand wheel that the Stingray uses and there are a few different sizes. Earlier this year I had gotten a used Stingray, and was constantly running into speed issues(had some wall smashing accidents), just like others had said, you can slow it down, but the motors are not geared high enough for low speed and would stall often. I had looked at trying some of the Polu motors, however there were quite a few people complaining about how much play is in the motor shaft and low quality the motors are. Before I stopped working on it I was thinking about a pair of large servos modified for the drive wheels, but never looked any further into it.
• Roy Eltham Posts: 2,996

  2011-10-15 13:57 edited 2011-10-15 13:57

  charleyshf, thanks for the tip, I've ordered some wheels from them to try out.
• erco Posts: 20,250

  2011-10-15 14:33 edited 2011-10-15 14:33

  charleyshf wrote: »

  ... Earlier this year I had gotten a used Stingray, and was constantly running into speed issues(had some wall smashing accidents), just like others had said, you can slow it down, but the motors are not geared high enough for low speed and would stall often.

  
  Amen, Brother charleyshf!
• charleyshf Posts: 165

  2011-10-16 02:02 edited 2011-10-16 02:02

  NP, hopefully you will have some good results with the different motor/wheel combo..