xan, do u integrate it in your code?
for each leg u controll the i/o and when he stand up the he press down until it give contact and the body amount is reached.
u integrate it to all legs?
then u must say it that he goes to terrain the he must his feet higher
raise to go over refusal.
its possible to intagrate this to your ps2 code…
or what u want to do? a new code? how to controll?
Hey Xan, tossing something out here…
This is a bit out of the box thinking, why not use a stable PWM output signal then use the force sensing resistors in place of that 4.7K resistor on the output of the PWM signal, with 8 capacitors and have the BB2 read the analog voltage ? I noticed there was an RCTIME function in the Atom code could this help? Or a way of reading an analog voltage? Or possibly convert the resistance into a sine wave output? and read the frequency.
http://www.ontrak.net/Pwm1.gif You would have to change the value of the cap to be within your spec.
Not sure if this helps, but currently im trying to find a more efficient way to read pwm data with the bb2 without using assembly or getting overly complicated.
*edit, scratch that, what about a circuit that does voltage --> frequency then have the bb2 read a frequency input?
–Aaron
Hi t4ndu,
I’m planning to integrate the terrain adaption into my code. The idea is to first release a faster version of the code (2.0). The speed is needed for the terrain adaption. I’ll include the terrain in 2.0 after it is released. Properly as 2.1 using conditional compiling. But you have to be patient 
Hi LikWidChz,
Why should I use this strange method to read the Inputs when I can do the same with just 2 resistors. 
Which are probably much more accurate anyway. And the Schmitt trigger converts it to digital signals. I’ve got it already up and running. Works perfect!
I’ll post some pictures soon.
Xan
Lack of knowledge on which electronic devices to use, although that Schmidt trigger is pretty slick!
Does that just output a change state at a certain level of pressure on the foot?
–Aaron
Hi All,
I had those pictures lying around for a couple of weeks but didn’t had time to post them. 
I did some testing with the setup on a breadboard before I found the right resistor value. The value needs to be very precise to let it work. With other values the rubber caps give enough force to prevent the Schmitt trigger from switching. I came out with some 6k8 resistors and they work pretty good.
I’ve made a home brew PCB with the needed components. I miss the time that I went to school and simply could walk to the lab and made my own “professionalâ€
A lot of interesting experimentation, but looks like in the end you are going back to using digital input for surface contact determination. If this is the case, looks like the analog part of the setup could be skipped saving time, complexity, and $$$. If the final position output is to be a 1 or 0, then a simpler contact switch approach may work just as well. I’d save the force sensors for applications where digital input won’t get the job done.
Hi Xan,
Great to see some pictures! Very cool picture of your Phoenix 8)
I liked how you covered the 9v battery, very neat.
I used the same schmitt trigger too, but I never got so far as making a PCB board.
As Zoomcat mention the FSR could be saved and a microswitch solution could be used. But I believe its easier to integrate a FSR onto the tiba then using a mechanical switch. Especially if you want it to look neat and tidy.
Another soulution would be a total redesign of the tibia like the one Jim posted or a solution like Matt Denton used on his BF hexapod… 
Hi Zoomkat,
I understand your doubts about adding analog sensors, making a PCB and converting them to digital signals when you can easy do it with a switch. But on this point I don’t have the need for analog signals. I only want to know if the leg touches the ground. So having digital signals running in to the BAP is easier and it saves processor time. Zenta already mentioned another important reason. I didn’t find another small and good looking contact switch that I could easy mount on the leg without redesigning the leg.
About the $$$ I think the whole set (6 sensors, Schmitt trigger and some resistors) cost about $40-$50. That isn’t to expensive I think…
But don’t get me wrong, I totally understand your point. I’m just explaining the choices
Thanks!
Hi Zenta,
Thanks! A friend of mine got a real good camera and Phoenix had a whole photo shoot last Sunday. I think you know why And this is one of them. The pictures came out very cool!
I used the same material on the back side of the SSC to protect my Lipo from the sharp soldering points. It’s nothing more then the back side of a black plastic cover that was lying around. LOL
http://www.a1-office.nl/catalog/images/products/512363.jpg
The guys from trossen robotics give me some good links for good quality PCB’s I’m planning to check them out and see what the PCB will cost me. Maybe it’s cheaper to order a bunch of them. If you’re interested off course
Both Jim’s and Matt’s solutions are very neat! What I like on Matt’s design is that you can pick up one leg by hand and the sensor feels the hand. This makes it possible to let the other legs return in to place. Very cool!
Thanks for the comments guys!
Xan
Hi,
It was time to resume this part yesterday. I had to put it on hold because I heated the heat shrink so much that it melted the two wires from the FSR to each other. LOL The defect FSR is replaced with a new one.
I’ve added the first part of the terrain adaption to the code. This already gives it more life! When the phoenix is lifted from the table all legs will stretch out. Once back on the table phoenix will get back in his “normalâ€
While you are thinking about what type of switch to use, you may want to check the below post. A simple (and inexpensive) setup simulating foot switches for detecting when a a foot has made contact with a surface. Possibly a way to continue software development while researching what to use for the foot switch hardware.
Xan,
I can’t wait to see videos of this thing of yours walking over uneven terrain… It’s creepy the way you describe how you can pick it up and its legs fall… I think I had seen a video of some other phoenix like robot on trosson robotics site…Did you ever post the code on how you did that?
I have an idea on how you might eliminate the sticking of your feet to give erroneous results in your setup, I’ll draw something up and lemme know what you think about it, I’ll post it here.
–Aaron
I was in Radio Shack yesterday and looked at the the pushbutton switches they have available. The below switches appear to be robust enough to be used on the foot of a hexapod. The mini one might fit inside the rubber foot boot, and the others arrached to the foot using a modified rubber foot boot. The bottom tactile switch is very small and might make a good gripper switch.
radioshack.com/product/index … Id=2062539
radioshack.com/product/index … Id=2062543
radioshack.com/product/index … Id=2062496
Hi Guys,
Thanks for the support!
The only video from a hex doing that was the one from Matt. He’s got a video from his V4 doing this.
All ideas are welcome!
I think the spring load in the switches should be enough to make sure the rubber cap is pushed back in to place. I hope they are not to strong because with the weight spread over 6 legs the spring load could be to hard…
Did you also think about a way to mount them? I want to find a good balance between decent reliable sensors and keeping the phoenix sexy. Who would want a lady with huge feet? 
I’m also thinking of making some kind of lever to move the sensor a bit up and out of the rubber cap. More info when I’ve got some drawings ready.
Xan
i think i will use this one or others like this
distrelec.com/ishopWebFront … 65599.html
small print switch its 6mm x 6mm
you can put it under the End Caps.
its have a small spring. its only closed when phoenix stand on the ground.
for mount it with glue…
T4Ndu –
I think those switches are quite tiny enough, but I just don’t think there is another solution to an alternative to the force sensing resistors.
My theory is to move the FSR from the less obvious place of under a foot and enable all possibility of it sticking by putting it between 2 hinged pieces of aluminum or on the top of a metal tube where its not close to the ground but simply getting smashed by the phoenix’s own weight…
I’ll draw up a photo of my idea and who ever can run with it or shoot it down, either way might as well… I was going to draw it up in adobe Illustrator or something…
Btw those SPDT switches go bad after a while, I have a notebook that uses them in the left and right mouse buttons and one of them has stopped clicking, the switch physically still works but the actual click has stopped, might be a bad thing if you glue it to your legs of your robot :\ I think the FSR is an excellent item as there is no moving parts, the moving parts you have to add to it to allow it to work, which could be as simple as a rubber foot between it and metal, seems pretty bulletproof.
–Aaron
Never say never I think Matt Denton has proven you wrong already since he used microswitches as feet sensor on his B.F hexapod.
That sounds like a really good idea! By placing the FSR on another place than under the foot can be a good idea. I’ve experimented with similar thoughts too. Also similar to what I did when placing the sensor for A-pods mandible.
Looking forward to see a drawing!
Drawing coming sometime, been distracted lately, although
I think this form factor is quite a bit more forgiving then a pad and looks like it will last a whole lot longer too!
So that’s the image of the product, here is a link to the digikey part number along with the data sheet.
search.digikey.com/scripts/DkSea … 02-1211-ND
Funny, this company CUI is actually about 20 minutes away from my house… and at 5.18 cent’s each… They are cheaper then the others
This handles up to 3.30 lbs or 1.5 Kg… so I’m not sure if that would work for this application although it might work good if you desired a higher sensitivity level ???
here’s another by the same company
search.digikey.com/scripts/DkSea … 02-1227-ND
http://img26.imageshack.us/img26/990/16059987.jpg
So first time using illustrator, anyway that’s kinda the general idea not exact, I really dislike the idea of using rubber bands or springs on a robot… anyway… I was thinking it would be interesting if you could make a new femur or something that was 2 parts…?
http://img26.imageshack.us/img26/2070/femur.jpg
I think you could make something somehow that wouldn’t require springs or rubber bands or any hokey crap like that… I hope this sparks some “Different” Ideas perhaps?
Illustrator is pretty neat.
–Aaron
Good idea, but how stable is it still the whole?
When one cuts through the legs and fixed with a rubber…
well yeah … That’s just the idea here I think that the femur or tibia would have to be 3 thicknesses of metal/plastic to really accomplish this like some sorta slots like a futon frame where it can only wiggle back and forth a little bit…
The clearances would have to be very very close so that you couldn’t visually see the flex but if you felt it you could figure out that they are loose, you wouldn’t want so much wobble that when your robot picked its leg up you could see it droop down. it would be just enough flex to make your FSR resistor read nothing…
That’s the basic idea though I’m sure this will spark other manufacturing ideas although its very rough as you can tell… the basic idea is to get the FSR off the ground and eliminate any possibility of the FSR getting inaccurate results
