Home Creators Posts Import Register Favorites Logout
haven't archived this post yet. have a subscription? use the importer!

Content

Here's my progress on the new openDog foot, which has both an analog pressure sensor and a switch so i know it's definitely on the ground.

The 'pressure sensor' is in fact a load cell which has two strain gauges attached and an small amplifier board which gives me some data. It's one of these (along with the code example):   https://learn.sparkfun.com/tutorials/load-cell-amplifier-hx711-breakout-hookup-guide/all 

It's not really supposed to be mechanically mounted the way I have it, but it seems to work ok to detect small amounts of flex in the aluminium extrusion which of course bear all the mass of the robot. 

My only setback so far is that the board only outputs data 11 times a second, which is no where near quick enough, so I've ordered some AD7705 boards which are a lot quicker.

This is the next video due for Patrons, although it may be later next week due to some other commitments as we as awaiting the new amp boards (and making them work)

Files

Previews only

Comments

Adam Purdie

I've found FSR's to be a faster alternative to strain gauges, they're as responsive as your reader can read. In the teensy 3.6 I was up around 4000 reads per second :)

XRobots

I've tried them in the past, they aren't very linear and in this case it's a challenge where to stick it ;-)

Adam Purdie

You're right, I had to come up with some pretty interesting spring layouts for my spider-feet to use the FSR and they're non-linear. At least in my case, i treat them like a basic reading of [ not touching, light, normal, heavy, problem!] with the goal of comparative measurement rather than absolute. This is a bit of a side effect of the need for the robot foot to touch-down up to 30 degrees in any direction wrt it's leg so the precision is immediately n*sin(0..30) anyways.

XRobots

Yes openDog's leg will also touch the ground at various angles. A more 'upright' post will cause less of a reading on my load cell, but then the leg would be less springy that way anyway because the knee would be straighter, so it might all work out.

alicetaur

I'm not sure if im not good on magnet theory but what if tou put two apposing magnets so that float apart when the magnets are forced closer they warp the fleid to trigger a hall effect sensor

XRobots

I think I'd have to try it and see what happens, it might cancel out the field?