CHEAP air cylinder positioning with 43oh and flow meter

[swampdonkeykami 13]

Hi Guys! I've been toying with an idea, that I  didn't think would work, and set up a test jig.

Using an air compressor, a solenoid valve, a flow meter and a 43oh I found that we can get a repeatable cylinder position, like +-2.5 mm! 

The flow meter is measuring the volume of air that we've fed to the cylinder and the 43oh opens or closes a solenoid valve depending on the cylinder position.http://arduinoforgoodnotevil.blogspot.ca/2013/10/accurate-ish-pneumatic-cylinder.html

[bluehash 1,578]

Pretty cool!

 

Attaching video:

http://youtu.be/Gmt7iVCwDKY

[swampdonkeykami 13]

Now that we know it works, I'm bustin' out more cylinders and joystick controls! :thumbup:

[chibiace 46]

looks dangerous, fantastic!

[t0mpr1c3 91]

I like it! Have you tried it under varying loads?

[Rickta59 589]

So is going to be part of some gold mining machine?

[swampdonkeykami 13]

Glad you like! More fun in the works!

 

No, this came out of a gold mining machine. I'm re-purposing for something completely useless. ;-)

 

As for varying loads, yes and no. 

If you are talking lifts various weight of items, shouldn't make a difference because the cylinder volume is constant. You put in 3 litres of compressed air @ 30psi, the cylinder moves 500mm. You put in 3 litres of compressed air @ 200psi, the cylinder moves 500mm. The trick is, the 43oh needs to return to a known position (full retract, full extend, or limit switch) prior to changing the load.

 

But if the load varies with stroke, like a pivoting boom, then there needs to be some fancy maths to figure it out. Because air is compressible, the 3 litres you put in @ 30psi gets compressed as pressure required to support the load increases.

 

And, it'll be sloppy because air acts like a big spring under accelerating and decelerating loads.

But I've got a few tricks up my sleeve to deal with that...

[swampdonkeykami 13]

*thinking* or we could install a pressure transducer and have the 43oh calculate the volume of air @ X pressure and base all movements on that baseline pressure, then don't need any fancy geometry algorithms for axies!

 

THANKS FOR THE INSPIRATION GUYS!!!!!