Lynxmotion Tech Support

Yes we will add them with the pressure sensors and stuff to make a pair of foot sensors. This has been a long time comming.

Wow the first third party support for the Servo Erector Set.

stelk: I've been giving some thought to you approach and have a few thoughts to share, but all should be taken with a big heap of salt because I don't have and idea what I'm talking about

Magnetic susceptibility of the spring will cause drift over time, though it should have a peak.

The air gap isn't that big of a deal because if the measurements are accurate, then the size of the gap is known. The location of the tip of the foot can be known with some certainty.

The more the air gap is decreased, the more the small changes have to be amplified. Considering environmental factors (walking over a screw in the floor) a larger air gap will yield a better signal to noise ratio.

What about interaction amongst the sensors when the legs are nearby each other? Will the apparent force increase greatly when another leg's magnet is in range of the HES, or is the range small enough that this is a non-issue?

This may all sound like I'm not a fan of the idea, I really am. I've just fought and failed against this one for long enough to be suspicious of anything that looks like it will work This solves the main issue I was having by adding a spring. I couldn't use one with FSR and had all kinds of problems getting them to reliably reset to zero.

Andy,

Those are some excellent points. Here's my initial response, I now have some thinking to do on my commute.

1: Magnetization of the spring. Although not shown in the pictures the yellow "cap" is threaded and can be removed to replace the spring. My thinking was that over time the spring may loose tension and the user may wish to replace it. The springs are rather cheap and you may wish to have a variety of them on hand to compensate for different loads. Also if the spring becomes magnetized over time it may be replaced with a material not susceptible to such. I am currently using piano wire.

2 & 3: Air Gap: The larger the air gap the stronger the required magnet. The stronger the magnet the more inference from other legs is possible. This is a very good point on your part, right now I have been doing the majority of my testing on a thrust cell. I'll put a layer of ferrous material on top and let you know what I observe. One idea is to place a ferrous "cage" around all but the top of the magnet. I'll have to read up abit more on this as I think this would also concentrate the flux. I'd like to keep it simple if possible.

Thanks!
Mark K

hey robot dude we talked about this on a another thread but Why didn't you tell me about the foot sensors that you put in the catalogue I don't really have the tools neccesary to make my own sensors. how ever I have thought about it and if you had a tube slide in and out of the larger tube for the SES then put a small light sensors inside the tubes witha couple of LED's mounted on the outside of the leg that were on. then drill a hole where the tubes would be when the leg was on the ground. then when the leg was on the ground, the light would shine through the hole and the light sensor would get a good reading and tel the Atom that the foot was where it should be. this would only give you a fairly primitive yes/no feed back but for my purposes (and hopefully somebody elses) it would do just fine and dandy.

Oh yeah forgot to say something. the inner tube would be sprung out to it's full length at all times except when ground contact was made.

Unless Jim's using an expensive pressure sensor (which I doubt), his way is much cheaper.

You can still use them as simple on/off switches.

The neat thing, though, is that you can tell the exact amount of pressure that the foot is undergoing.

So, on an uneven surface, you can make the hexapod balance by having it try to attain equal pressures on all of the sensors.

That way, your hex could (theoretically, at least) crawl right over a rock without problems.

maybe I should look again but it looked like he was using custom machined brackets and stuff. I Have NO MACHINE SHOP. I can not do 3-D rendering and rapid prototyping and stuff like that. I am 14!!! I sthere anything I could do with Bolt on Parts?

Never mind wrong person. I really like Robot Dudes (Jim??) approach. can I get a list of parts needed to build one of these sensors? and also, where did you get those rubber caps, not the lynxmotion feet, but the one that is inverted inside the foot pressing the sensor? Thank you! Now I can actually start ordering parts since I know everything I wil need.

I have not lookd but if Lynxmotion does not have the round rubber feet I asume your are talking about, Digi-Key sells them.

www.digi-key.com

Jim's working on the complete package (whilst working on 20 other projects at the same time, no doubt).
If you wait a bit, you won't have to wonder if part X from one site fits part Y from another site.

I have had my share of "part x not fitting part y" it would be a good idea to wait but then if you can't , I can relate to that as well.

Right now I'm doing the Biped Scout. Not working on anything else till it's done, but it's close anyway. The tubing foot pressure sensor kit will not take long, but I have to finish this thing first. Just need a little more time.

I can wait a few months. need moolah But stil I want to get an order form typed up so the moment I make my last payday for this thing I can jump on this computer and order the parts I so can't wait!!

Oh, don't get me wrong, I wasn't saying that you couldn't do it with simple on/off switches.
I was simply saying that knowing the pressure is potentially more useful than just knowing whether the foot is down or not.

The only problem with pressure sensors is that you'd need 6 A/D converters.