$2.64? For his sanity?! 
Thatās an insult. At least give him a fiver! 
Better and cheaper? Iād say the wait will be worth it. 8)
Sighā¦ I knew I was leaving myself wide open on that one. 
Well ok, but I will have to have a little more for shipping. Looking for a small box to put my sanity inā¦
just looking for contact or force applied? the legs are aluminum so you could possibly use a self biased hall-effect sensor and a floating or hinged steel pad of some sort that closed distance as the foot is placed. doesnāt work well for much more than contact sensing though as linear output hall effect sensors are a bit pricey.
whatās the smallest diameter FSR weāve seen? you could put a pin in the leg end, then allow it to compress an FSR mounted perpendicular further up the leg as pressure is applied. you could put a slot in the leg to allow for the FSR mounting.
| |
|-----]| <- FSR in a slot at right angle to pin going out bottom of foot.
| o |
\ | /
\ | /
\+/
'Actually, I DO have a foot switch in mind.
It would provide X and Y analog āsenseā, and a contact closure for Z. I still need to determine the downward force a leg exerts, and if it is sufficient to actuate the switch (Operational Force 800ĆĀ±200 gf).
It would entail modifying the aluminum leg below the servo (radically). No drawings (or actuator parts ordered) yet.
Several advantages to this sensor design if the footās force is sufficient.
A simple push-rod to a Micro Switch could also suffice. Switch could conveniently bolt to the side of the leg. A sliding foot cap would actuate the switch.
Email me if youāre interested.
Alan KM6VV
I want to do a ground contact switch. I think it would be easier to implement than pressure sensors and a lot can be done with simple switches.
The robot weighs 1.8kg so worst case (1.8kg/6) each foot needs to be able to switch āonā when as little as 300g (10.56 oz.) is applied. Adding a 30% fudge factor means the switch needs to throw with about 200g (7.00 oz.) of pressure. The foot also needs to be somewhat rugged as it needs to support or withstand the 11 oz. along with the added stride pressure.
Waterproofing would also be a nice addition.
Yeah, me eitherā¦
The big problem with this method is the switch would have a very limited angle that it could switch with. Iām thinking more along the line of a round hole at the end of the tibia that has an insert with rubber insulator, that self centers a center rod that holds the actual foot. The leg and center rod make up the two contacts of the switch. If this sounds vague, that because it vague in my head at this point. lol But if I can work out the details it should āswitchā with little force from any direction.
I have an envelope for yaā¦
8-Dale
Doh!!! 
What about a prox sensor? No need for mechanics then.
Edit >>
Bad idea, that would not work.
My āPetā idea would have most of the Tibia below the servo deflect upon contacting an obstacle. Touching something in X or Y would result in the lower part of the Tibia would result in an analog output (200 ĆĀ± 80 gf.cm at 10ĆĀ° deflection). It would probably deflect with a Z force as well. Iām anxious to see it in action.
The Z switch, as I feared, would need more like 600G to actuate the switch, unless I can alter it. Or leave it out, and put a separate switch lower on the Tibia just for Z.
I have been considering having a thin brass tube slide up and down over the lower 1/2" or so of the Tibia. This travel could actuate a MicroSwitch (single Z direction, 'tho).
To expand on your idea, if the lower end of the leg were extended with a tube, as I think youāre suggesting, it could be divided into 4 or perhaps more sections along itās length; and then when the floating center contact was deflected, it would contact one or more of the āquadrant contactsā, and thus give a directional indication of the force exerted on it.
Alan KM6VV
Is the leg one piece or double stacked? If the leg is made of two thiner pieces of aluminum the middle gap can be utilized for a sensor configuration. Hmmmm, Iām thinkingā¦
Single piece of 1/8" aluminum.
Alan KM6VV
Hi,
Iāve also been thinking of making a sensor on the tibia. Since Iām using plastic parts for the tibia I have been playing with a thought of using FSR inside or as a part of the plastic part (tibia) simply by cutting a long thin slot across the upper part of the tibia (under the servo). About 2/3 of its width has to be cut, then Iām thinking of putting the FSR sensor (and some glue 
)inside this slot. Since the plastic material has som flexibilty I hope the FSR would give the tars some āfeelingsā 
But I understand that you want to use a mechanical switch. What about looking at nature? Making a tars part at the end of the tibia as a hinge. And let the hinge/tars activate a microswitch?
Just some thoughtsā¦
Hi KĆ„re!
I think the FSR has to be compressed between two pieces of flat stock? Could make a 1cm square tars out of aluminum, with a little hinge to the tibia then a sandwich of FSR, aluminum, and some sheet rubber for traction. Kinda big.
Just that FSR needs a fairly big surface area to work. Are you suggesting that the little microswitch be HINGED on one of the two little holes, and the lever (with some rubber pad) be used as the tars? That might work!
Smallest microswitches Iāve seen are about 12 mm x 6 mm, (Panasonic AH1 microswitches 1/2" x 1/4"). These little switches actually bolt up quite nice to a wide (1/2") lower Tibia. When I made my Tibia, I was thinking of a āsliding tubeā over the end of the Tibia, hence the lower end of my Tibia is quite thin 3/8" (9,5 mm). The tube would push up against the lever of the switch. A spring inside the lower end of the tube would be compressed to add/adjust the resistance, if desired.
Iām still hoping my āpetā idea will give me some REAL force recognition in X Y! And if I can add analog force sense in Zā¦
Alan KM6VV
Perhaps a new thread specific for the development of the contact sensor is needed. A contest with a prize for those that actually make a working demo model might add some inspiration (but could get into āidea stealingā issues) and make sure cost are reasonable. Also a basic design specification would be needed to make sure the gizmos would perform the desired function.
Ok, this is just one design that might not be cost effective, but I did this to perhaps spark new ideas.
I though about hinging the tip of the leg but it just did not seem right. I also thought of a design that would require a rounded rubber boot similar to a design Jim made awhile back but the problem is the stifness of the rubber as well as the plunger not staying put in the rubber boot might be a problem. Then I thought of a push button and came up with this idea below. A very light weight spring would need to go around the plunger shaft to keep pressure off the sensor.
http://img158.imageshack.us/img158/2753/fsrswitchuf1.png
http://img206.imageshack.us/img206/8812/fsrswitchdwgwe1.png
Here is another design but it does not have a spring. The plunger is āfree floatingā
Maybe this thread could be used: lynxmotion.net/viewtopic.php?t=319&postdays=0&postorder=asc&start=0
What ever happened to the good olā tape a force sensor to the bottom of the foot? Or use an IR sensor to detect distance from surfaceā¦
KISS - No need for fancy red and blue parts =P
How about some links to working examples.
Why tape it? just use Elmerās glue and yarn to attach it. 