Hi Jim,
I saw your test video on youtube. And like that it can detect ground from many angles (45 +/- deg and more). But could it be possible to mount the pressure sensor on the tibia too? So that you only need to tie up the wire and not the tube.
I think you can improve the performance by hard piping the sensors to steel capillary tubes. Basically, you use very narrow steel tubes everywhere except at the joints, this way; you only need a short length of PVC tube to allow for movement at the joints. This is the same concept used for automotive hydraulic brakes where the lines are all steel except a short length of rubber hose used to allow the suspension to move without breaking any lines. Small Parts has this type of thing and they also have manifolds that could be used.
I know this is prolly more complex than what you are looking for, however its one idea I thought I would throw on the table to spark new ideas.
First off, I added two more images to my earlier post.
lynxmotion.net/viewtopic.php?p=44743#44743
Yes Mike, I have experimented with some aluminum tubing as you have mentioned. I like the idea, but I also want to keep it as easy and inexpensive as possible.
Yes Zenta, I like the concept of putting the Freescale sensor on the leg somewhere. I’m at least trying to do it that way. Just getting started on that front.
Very solid looking! I like it!
I think that placing the sensors on the legs is a good idea. Let’s hope it fits 
Xan
I was thinking the same thing that it looks very solid and powerful. Jim there are several designs of the leg, which one did you wind up using?
Hi Mike,
If you’re referring to the CAD printout it’s actually 4 parts that make up the leg.
I have tried the aquarium valves, picked up two different kinds. They appear to be of such low quality they leak. 
I started looking through smallparts website but couldn’t find what I was looking for. It looks like they primarily use the Luer fittings which appear to be larger than the normal 1/8" barbed fittings.
I’ve successfully mounted everything needed on the leg, but I’m still needing a good, cheap, small, valve. 
So close, yet so far…
If the valve stem will screw out of the valve, you might remove the stem and put a little plumber’s grease or vasoline on the valve needle and/or valve seat and see if that will help the valve seal. Also, make sure the pressure direction on the valve is on the seat side of the valve.
I will try this if I get time. The valve is really symmetrical as in there is no apparent input or output.
I’m going to order some of these valves with the Luer fittings. If you can’t beat um, you may as well join um. At least I will know if it will work, then I can deal with getting everything to fit.
Just a quick test for the cheap aquarium valves: attach a piece of tubing to the valve, close the valve, and put the valve in a glass of water. Blow on the tube and see where the valve leaks. If the valve bubbles out the valve stem, then put the tubing on the other side inlet and try again. If this fixes the leak problem, then that is the side of the valve that goes to the sensor. If the bubbles come out the other side of the valve, then the valve has a seat leak and probably won’t hold pressure without possibly lubricating the seat. The aquarium valves are probably needle type valves and will probably only hold pressure on the seat side as opposed to the stem side (which would require a stem packing).
edit: The below shows the directional idea of these types of valves.
Hey Zoomy, The PDF is interesting. It will help me explain what the aquarium valves are. Ok, the drawing on the left. The opening on the right side is reduced as in the drawing, but it then carries straight through to the left side.
In the left drawing, the pressure is the arrow, with the flow going in, angling down, then straight up thru a round hole. The bullet shaped thing at the top of the hole is pushed down on to the hole stopping flow. when the valve is opened, the bullet thing is lifted off of the hole, and flow goes up around it and out the opening on the right side of the valve body. On the outlet side of the valve pressure is also applied to the valve stem area, where it will leak without having a substantial packing on the stem. The glass of water should be a quick test.
Eureka! It works…
I tried a different valve and it’s working better. When you open and close the valve it reads 43-45 from the A to D. We did a run through on the pressure test. Weird thing is after doing the test the zero pressure reading dropped to 20 and has been holding for 20 minutes. I’m not sure if this is a problem or just an artifact of using the setup as it is. I also noticed that the sensor readings will fluctuate a little when pressure is applied quickly, as in overshoot and bounce back. Again not sure if its a problem or just something to be aware of. Edit: After a few hours the sensor is now reading around 60 with no pressure on the pod. Opening and closing the valve brings the readings back to around 43. Still don’t know if the readings can be improved, or if it’s working the best that it can…
[code]pressure A to D
on scale reading
0.00 lbs 45
0.10 lbs 50
0.25 lbs 63
0.50 lbs 97
0.75 lbs 133
1.00 lbs 195
1.25 lbs 265
1.50 lbs 343
1.75 lbs 425
2.00 lbs 509
2.25 lbs 589
2.50 lbs 653
2.75 lbs 832
3.00 lbs 980[/code]
Hi! At university we bought these sensors to use them for the legs of an hexapod. I’ve seen the picture on the accessories, like valves, tubes etc. Where can we buy them? Can you suggest to me any good online seller? Or is it better looking for them in a normal shop that sells this kind of stuff?
Thanks in advance
Paolo
The post dates back over 5 years, and RobotShop has since purchased Lynxmotion.
We currently offer two foot sensors:
lynxmotion.com/p-626-hexapod … -pair.aspx
lynxmotion.com/p-728-tubing- … -pair.aspx
Thanks for the quick reply. We already bought sensors, we just need tubing etc.
So I’m giving a look at RobotShop website. Thanks