For those that are interested in tinkering with foot contact programming and use, there may be a simple setup for proof of concept testing. I made the below simple rig to test getting servo position when foot contact (or any other similar event) is made. It works fairly well and points out some issues to be addressed. A similar simple setup could be made by those with hexapods that want to do some testing. In the below setup, the piece of aluminum foil that simulates the floor is tied to ground on my ssc-32. The paperclip “foot” is connected to one of the analog/digital pins on the ssc-32. A +5v pin on the board is also connected to the input pin thru a 22k resistor to keep the input pin high and not floating when foot contact is not made. When the paperclip touches the aluminum foil, connection to ground is made and the analog input pin goes from +5v to 0v. My test program stops the servo and gets the last position sent to the servo during a timed move. One should be able to easily adapt this setup to a hexapod for testing. First thing to do is to attach/glue sheets of aluminum foil to a flat surface like a 22"x28" or larger piece of poster paper or cardboard to make a conductive walking surface. This aluminum surface is connected to the board ground. For the foot contacts, a 4"x8" piece of aluminum foil is folded into a 4"x4" piece, then the hexapod foot centered in the piece and the piece folded up around the foot and secured with rubberbands. A jumper lead is used to connect the aluminum foil foot cover to the input pin setup for each foot. To test walking on uneven surfaces, something like a thin book (or other objects) can be wraped in aluminum foil and placed on the conductive walking surface. This is just an idea for doing proof of concept foot contact testing for little time and $$$ investment.
Nice setup. Does make testing easier. Think I’ll use a similar setup to do force feedback testing with. I want to be able to have each leg to keep the same force applied to the object its in contact with, as all the legs with the surface, as it walks/climbs over something. In other words, if the bot is sitting on a table top, powered on in the “1500” position, and each leg has say 50 oz/in of force applied to the table top to hold it up in the “1500” position, then, when it moves, I want it to apply that same force to the object each legs encounters as it is moving. Does that make any sense?
There are several conditions that need to be explored. This requires more thought than I’m ready to put in to it. Hopefully, I’ll have the time in a couple of weeks. I’ll start a new thread addressing those issues then.
Thanks for posting this idea as a starting point.
One first approach to take could be observing the pickup point of a contact on a foot, then see how much more added servo movent is required to generate the desired force to bring the bot to the desired level position. As an example, contact is made at position 1500, but an added 30 increments are required to account for leg bending and such inorder to achieve a desired level position. Then for general walking one could add 30 extea servo increments to the reported servo position to send the servo to its desired position. Idealy when a contact is picked up, the foot is pretty much firmly placed and only leg flex has to be accounted for.
Hmm. . . That is definitely something to consider. I can’t do anything until I get 6 more HS-645mg’s. I used 6 of the 645 from my AH3 to build my Phoenix with. Replacements are due in sometime this week. Now I wish they’d hurry up and get here.
Just made a quick video (below) of the simple contact sensor setup, and bottom is the output of the program controlling the servo via the ssc-32. The servo starts at #0P2200 and then is given the command #0P700T6000 to start moving. The VC command is sent to the ssc-32 at 20ms intervals, and when a reply of 228 or less is received, the STOP 0 command is sent to the ssc-32. Then the QP0 command is sent to obtain the last position command sent to the servo from the ssc-32. The returned value is is the multiplied by 10 to give the approximate current position of the servo. Then the test cycle is repeated.
youtube.com/watch?v=3zQalB0bfNE
step 1
starting 2 second start delay
voltage byte is 0
position byte is Ã…Â
servo position is 1380
step 2
starting 2 second start delay
voltage byte is 0
position byte is ‡
servo position is 1350
step 3
starting 2 second start delay
voltage byte is 3
position byte is Ã…Â
servo position is 1380
step 4
starting 2 second start delay
voltage byte is 19
position byte is ‹
servo position is 1390
step 5
starting 2 second start delay
voltage byte is 6
position byte is ‹
servo position is 1390
step 6
starting 2 second start delay
voltage byte is 2
position byte is ‹
servo position is 1390
step 7
starting 2 second start delay
voltage byte is 0
position byte is Ã…Â
servo position is 1380
step 8
starting 2 second start delay
voltage byte is 1
position byte is Ã…Â
servo position is 1380
step 9
starting 2 second start delay
voltage byte is 0
position byte is Ã…Â
servo position is 1380
step 10
starting 2 second start delay
voltage byte is 6
position byte is ‹
servo position is 1390
done
It works great! All you need now is a popsicle apparatus using these components to make a contact sensor for a leg.
Probably easy to do. You know the old saying “those that can, do, those that can’t…, populate the peanut gallery”. 
Voila!
Eye dropper rubber end cap, dowel leg, two copper carpet tacks, two pieces of wire, and a rubberband. Video isn’t so good as every thing (cam, ssc-32, servo, USB serial adapter, etc.) is on a single unpowered USB 1 laptop hub making things a bit jerky.
While at Radio Shack, I took another look at the below 5.0mm tactile switch. I think it might work well attached to the bottom of the leg under the rubber end cap.
has anyone tried using quantum tunnling pills ?
as foot sensors i am thinking about giving it a go just thought i would see
if anyone else had tried it first
I can’t find any post-production sensor for sale. You have any links?
Seems like it’s still a developing technology. Probably should wait until the product mature and offer in more common and small package.
hi they sell them at are local maplin electronics aka (radio shack)
look here maplin.co.uk/module.aspx?moduleno=44202
also have a look here peratech.co.uk/
and the are cheap at £1.49 for 4 i think they could be good
but i am no good at programming so i dont think i will get to far with doing the project , but i am going there today for other parts i will get some just to testand see if would be a worth while project
Wow, the size is very impressive. Actually, my idea of small package is something like an IC, with pinout. With pre-define input/output specfication, sensitivity etc. And most importantly, they have done large amount of sample testing.
End consumer just need to know what application (i.e. max force) and there is a component which fit the purpose! Sure a bit lazy to do modding myself, 
especially with this kind of stuff. My patience probably won’t last long. 
I’m not familiar with quantum tunnelling pills. I’m in the process of adapting a recipe for fabric force sensors (maplin.co.uk/module.aspx?moduleno=44202), sounds like a very similar material. I plan to post details on my project in the near future.