Basil, Level of experience? Rocket Science? Heck, if I can do this anyone can. A few years ago I didn't even know the difference between a DC motor and a servo, let alone how to control them. No collage degree hanging on my wall either. LOL
Dave Schulpius
On Sat, Jan 9, 2016 at 4:01 PM, 'Jim Easley' jimeasley.tech@gmail.com [B9Builders] wrote:
Not to fear Basil. It is a learning experience for all and you don t need a degree around here. From: B9Builders@yahoogroups.com Sent: Saturday, January 09, 2016 1:47 PM To: B9Builders@yahoogroups.com Subject: Re: [B9Builders] B9 Arm centering issue demo
In regard to skill levels within this group, this level of complexity is scaring the daylights out of me and I'm thinking I should go out and get a Robotics Degree or Rocket Science Degree or similar if intend to build "the big guy"... you guys are amazing Basil (Down Under) -----Original Message-----
From: 'Jim Easley' jimeasley.tech@gmail.com [B9Builders]
To: B9Builders
Sent: Sun, Jan 10, 2016 8:29 am
Subject: Re: [B9Builders] B9 Arm centering issue demo
Dave, I was not actually suggesting adding any switches to your design. In essence much like your existing optical interrupter only with a separate emitter and detector with one located at the slide platform and the other near the wrist. Having a direct line of sight within the rubber arm could be handled with a small diameter tube with an aperture at the receiving end. I hear you on your space requirements as well as your need for natural motion. Just fitting the components in the space you have available must have been difficult. The only way I know to achieve near absolute positioning is either a fixed stop (perhaps some sort of solenoid retractable pin) or a closed loop system with enough resolution to achieve the accuracy needed. As you have already experienced, gravity and hysteresis is not easy to eliminate in the real world. You may have already researched this but, take a look at multi-axis CNC control. I know it might seem like overkill for a hobby project however, this might be right up your alley. If your project ever becomes a vendor item, numerical control could open up additional possibilities along with pre-programmed motion routines. Please forgive me for rambling on. I think we all want to help and what you are trying to achieve is amazing and certainly promotes a higher level of skill within the B9 group. From: B9Builders@yahoogroups.com Sent: Friday, January 08, 2016 9:01 PM To: B9Builders@yahoogroups.com Subject: Re: [B9Builders] B9 Arm centering issue demo Jim, Your probably right. There is a bit of play in the worm gear. However I thought gravity and the weight of the arm would keep any slop out of the stopping points. Also your comment about the extra switches is spot on. However I'm not sure how to script for this. I want the arm to look somewhat real as it comes t rest. I really dont want it to be moving around a lot looking for a couple switches before it comes to rest. One of my biggest roadblocks is space. I just don't have a lot of room to fit sensors and feedback devices on this thing. I'm leaning towards trying to find an encoder that ill fit into my setup. It needs to be almost flat to fit my needs. One big drawback with useing an encoder in my robot arm is that when the robot is first started up the encoder must be homed before it can be used. Of course that could be done in a startup routine. Thanks for the help. Dave Schulpius On Fri, Jan 8, 2016 at 1:53 PM, 'Jim Easley' jimeasley.tech@gmail.com [B9Builders] wrote:
Dave, I watched your video and it was quite impressive. You have a lot going on in your control setup to be sure. The mechanical assembly looks really good. It has been my experience the mesh between gears can introduce some play in an assembly. I believe what you are trying to accomplish is much like a CNC operation. That being said, you should consider an encoder based loop rather than one based on voltage for absolute positioning. As you depicted in the video, given the arc of travel in the joint a little goes a long way and while gear motors provide lots of torque they do introduce positioning problems. I do think using the optical sensor was a good idea however that type of setup requires a separate trip point for each direction. As a suggestion, it would be simpler and more accurate if you used a reflective or two piece sensor at the joint reading the position of the wrist rather than a flag interrupter at the joint. This could be used as your home position for that axis. Just my 2 cents. From: B9Builders@yahoogroups.com Sent: Thursday, January 07, 2016 10:29 AM To: B9Builders@yahoogroups.com Subject: Re: [B9Builders] B9 Arm centering issue demo Hi David R. Nice suggestion about the stops. I'll have to think about it and see if that can be engendered. It may be difficult because the arm needs to swing back and forth past the center point. Thanks, Dave Schulpius On Jan 6, 2016 2:22 PM, "David Reabe dar19904@yahoo.com [B9Builders]" wrote:
Dave,can you put something like a bracket,or something that restricts it from moving beyond your stop points?
On Wednesday, January 6, 2016 12:57 PM, "David Schulpius dschulpius@wi.rr.com [B9Builders]" wrote:
Hi gang, I'm getting very close to finishing up my B9 Arm. Actually it's mostly done unless I come across a component needing a redesign. I'm in the coding and control phase right now. I thought I'd share this video I took last night. It's actually meant to send to Demention Engineering (Kangaroo X2) and Sptra Symble (Softpot) for troubleshooting help. It's long at 1/2 hour and kinda technical but you may enjoy it. I'm having a small problem with my arm not stopping at the same point depending on what direction it's heading. I send the same stop point from each side and it reports back a different value and over shoots. Anyway, feel free to comment and have fun. Dave Schulpius :
http://youtu.be/1HcrOEeRpw4
Obviously the Kangaroo is not able to auto tune such a heavy asymmetrical load as I have. So I took off the front claw section (about 4 pounds) and attempted another Autotune. This time I had a quick and successful tune. When I tested the tune I had nice smooth action with nice ramping effect. However when I reattached the 4 pound claw section the movement sucked.
Thank for nothing DE and SS! 
Dave
Obviously the Kangaroo is not able to auto tune such a heavy asymmetrical load as I have. So I took off the front claw section (about 4 pounds) and attempted another Autotune. This time I had a quick and successful tune. When I tested the tune I had nice smooth action with nice ramping effect. However when I reattached the 4 pound claw section the movement sucked.
Thank for nothing DE and SS! 
Obviously the Kangaroo is not able to auto tune such a heavy asymmetrical load as I have. So I took off the front claw section (about 4 pounds) and attempted another Autotune. This time I had a quick and successful tune. When I tested the tune I had nice smooth action with nice ramping effect. However when I reattached the 4 pound claw section the movement sucked.
Thank for nothing DE and SS! 
Obviously the Kangaroo is not able to auto tune such a heavy asymmetrical load as I have. So I took off the front claw section (about 4 pounds) and attempted another Autotune. This time I had a quick and successful tune. When I tested the tune I had nice smooth action with nice ramping effect. However when I reattached the 4 pound claw section the movement sucked.
Thank for nothing DE and SS! 
I attached the ezb last night and battle tested stopping at my set point from each direction. I'm going to call this issue RESOLVED! It's stopping either dead on or within 1mv of the set point. 
I couldn't be more happy, thankful and excited. Here's a youtube video I took last night of the results of the final testing. Enjoy :