I was designing an arm around the HSR5995tg servo Now I know better. The thing is I got the sprockets and .25 chain, even found some extra rigid super lightweight extruded plastic for the structural elements. So now the idea, for the elbow and shoulder I’d like to do something similar to these planetary motor servos lynxmotion is preparing. I was thinking if I used planetary motors with a very high gear ratio and a high resolution potentiometer to control the motion at the joint might I not be able to pull it off with a standard H Bidge? I mean it wouldn’t move fast, I don’t need or want fast, I want powerful. What I’m wondering is if using a standard motor controller and microcontroller will be able to control the feedback fast enough to keep the arm from trembling or bouncing to compensate and control position. I’m pretty sure I could get it to work under no load movement thanks to the high gear ratios. So any opinions? Anybody try anything like this? I might just try it anyhow, I don’t know how long it’ll take for lynxmotion to get their new motor servos working and on the market, keeping busy keep me from getting impatient.
Jim
I’ve hinted at some possible ways to control larger motors using servos in the long gear head motor thread. In particular I was looking at an easy way to control a larger arm operated by liner actuators (like little hydraulic cylinders). There might be some creative things to do depending on the type of arm you are making and its size.
One quick correction. I think you mean high resolution A to D not pot. The idea is sound, but there is one gotcha. When looking at the specs for any gear head motor you will find there is a point where additional gear reduction in the gear head does not add torque, and only slows the motor down. Fo example with the 22mm motor we are using for the smaller servo the allowed torque didn’t increase from 231:1 ratio to the 310:1 ratio. So the allowed torque is dependant on the strength of the final gear, and adding additional reduction will not increase the available torque. Now if you were to take the 22mm motor and add more gear reduction to the output via some really strong gears you could add strength while sacraficing speed. But then you have to work out the mechanics involved. There is no free lunch. 
yep , theres the loss from friction and gear strenth , so
to double the strenth you probly need too triple the motor size and
double the gear strenth,
i got a car window motor and i,m waiting for a couple more to try , they are super strong but they have probly 10degrees of play in them (might be because there old but would still work for an arm or hex leg just got to figure out how to attach them … 
Yeah, I was on to doing the reduction by installing the motors on the arm and rotating the joint (with bearings an .25" shaft) using sprockets and chain. I’m pretty certain torque will not be so much the problem as reading the the potentiometers on the joint through A to D on the microcontroller card and having that control the motor controller fast enough to be able to hold the arm still when it comes under load. I imagine a 100 to1 gear ratio would slip slow enough (even under moderate load) to make it capable of compensating position without to much shake. I was just wondering if anybody else was clever (desperate, nuts, crazy, stupid) enough to try this. Those 100 to 1 planetarys are noisy as your grampas black and decker drill but they sure got some torque to 'em.
I almost forgot, anybody see those new firgelli linear actuators
I have no idea how to design the geometry for an arm based on these babies but I’m anxious to try. Soon as I got some extra cash I’m gonna get me some of these to play with. Maybe I should see how backhoes are set up, done lots with servos but I got no idea how to rig these…
Here is my progress on our PGHM based arm.
should put a boxing glove on it.
Fit that on a tracked base (Johnny style) and boy could you cause some mischief.
I bought 4 of them for a prototype device I was working on for my job. In summation - you will likely be disappointed with the PQ12 series either the high speed or high force. They are on the weak side.
It looks… Powerful
Jim,
I know, I’m a persistent pain in the butt, but… When are you planning to make PGHM available to the public? I mean, love that prototype Jim, but, the more I see of it, the more I realize that it’s just what I need.
cwkoehler,
Thanks for the heads up, I think I’ll just wait for the L series to try the new firgelli linear actuators. According to specs’ it’s got 50N at 2mm/sec travel. I’ve already scratched out a few hexapod leg designs. Been looking at heavy equipment hydraulics to get my geometry inspiration for the arm. I think linear actuators will probrably be more efficient and cost effective, for certain types of articulation, in the long run.
Jim, you get to have all the fun!
fantastic!
Another great example of Lynxmotion innovations .
I made a decision a long time ago to show a lot of the prototyping process in the forum. For many it seams as if we are taking way to long to make things ready for sale, but in reality the process for us is really fast. The difference between Lynxmotion and other companies is we show all of the process, and others do not. The smaller motor is on the website. The larger one is not because I am waiting to receive some more samples with a higher gear reduction. I promise we are not dragging it out. We are going as fast as we can. 
All the better for us!