Here is a little video showing off the Johnny 5’s dexterity. More to come… 8)
I was wondering when you were going to use utube. This does not help my long day at work. Gona hafta wait till I get home. 
Youtube’s a nice utility to use when we don’t want to have to mess with video types, codecs, encoding methods… Makes it easier for everyone to be able to watch the videos without having to download 'em. 
Spreading the robot video love, one youtube upload at a time… 
It’s nice to be able to embed the videos, too. I’m working on a page for lynxmotion.com that’ll have all the videos with our 'bots that I’ve been able to find.
Sure, and why not let someone else pay for the bandwidth…
Can J5 bend to the floor to pick stuff up?
i Was just wondering that same thing
This rendition of Johnny 5 is not able to bend low enough to reach the ground. I’m sure it could be modified to do so, but I wasn’t interested in doing that. I wanted the two waist servos to be for showing expression.
I can’t wait to see this when I get home. 8)
It looks like he wobbles back and forth a bit. What is causing the wobble?
Inertia…
Looking at the video again, I’d have to say “physics” is the cause of the wobble. Although there was a bit of jitter at the end of large moves with sudden stops, I really didn’t see any more “wobble” than I’d expect to see in a large system of physically interconnected parts, acting as levers on each other.
You might be able to eliminate some of it with velocity ramping routines, which would probably be fairly tricky to do accurately in a hobby-servo-based system without direct position and velocity feedback from each individual joint. A more reliable way would probably be to subject each joint to some sort of pre-loading, as a form of anti-backlash mechanism. That way, you would always have at least some load on each joint, and though it would be less prone to oscillation at the end of each move, each and every pre-loaded axis would always be under some sort of force from a known direction, and consuming current accordingly.
Unless you’re willing to go with something that’s massively overbuilt, using harmonic drives, full bearing support, and/or mechanical braking on each axis, you’re probably going to have at least some jitter, wobble, and slop, just because of basic physics.
Yes the injection molded base could be made a bit less prone to these extreme forces by adding an aluminum disk to the top rotation part. The base is very stable for arms, but this is extreme. Note we are just pounding some cool moves out, We aren’t really trying to polish it too much.
Physics, really? I wasn’t much interested in smart ass answers to my legit question.
It sounds like it’s flex in the top of the rotating base. Got it, thanks.
Hey we’re just goofing off a bit. I certainly didn’t mean to offend, and I’m sure Seamus didn’t mean to either. 
I’m real sorry…
No problem. My question was to you, and you obviously got what I was asking because you answered it. 
Sorry I didn’t explicitly say “which specific part or parts are flexing to create the wobble effect” … 
I wasn’t trying to be a smart aleck about it either. I was just trying to convey that it didn’t appear to me as though there was anything more than the usual mechanical play involved.
In the interest of reducing the amount of play and slop involved in the turntable mount: i have found that, both on the original L5 arm, the Lynx-6, and probably now with the molded turntable assembly, that it’s more stable if the standoffs/rollers/bearing surfaces are under a small amount of compression. I achieved this on the L5 by shimming the servo downwards, and on the L6 by shimming the standoffs up slightly with thin washers. The new molded plastic base looks as though it would shim the servo downwards, but since I haven’t had any hands-on time with it, for all I know, it may not need shimming at all, in order to compress the rollers and ensure positive contact.
Even with all of this talk of “slop” and “play” in the various axes, this is not to imply that there is anything inherently loose or inaccurate about the Lynxmotion assemblies. They’re as accurate and tight - if not moreso - as I’d ever expect from anything built around hobby servos. Even the best “dual ball bearing” servos on the market place those dual bearings so close together on the output shaft as to almost negate any advantage over a single bearing. On the other hand, when I recently got my first set of metal-geared servos, with digital circuitry no less, I was blown away by the smoothness of their operation. I’d always used nothing but servos with mylon gear trains and analog drive circuitry, and I’m still impressed every time I turn the things on.
I didn’t see any smart answer. A question of flexing was asked and an answer was given.
As for the video, it was awesome! I did not see any wobble that would not be considered normal. The moves were swift and changing from one position to another which I would expect some flex & wobble as a result. Even if an aluminum plate was used, it’s still going to flex from the waist and torso mass.
Well done Jim, I give it a 9 
Brrrr.
The track base movement is good. Is that the lowest speed it will move at?
BTW Nice Job…