I'm making a Hydraulic/Pneumatic thing, need ur ideas pls

[roid]

Anyone know about hydraulics here? Actually that doesn't matter i'm still going to ask.

I'm curious if there is a hydraulic joystick that instead of operating valves - it instead pushes directly on bladders/bellows.

What do u think?

i did find a pic that seems to be illustrating the idea http://patentimages.storage.googleapis. ... gf0002.png Note the joystick on the lower right and the balloons below it.
But i don't know of anything that actually uses this type of mechanism IRL.

There's various proof of concept engineering toys that use syringes, but syringes with their sealed sliding surfaces are dumb they have way too much friction. (EXAMPLES: https://www.google.com.au/search?q=hydr ... m&tbm=isch )
Balloons/bags/bellows are better.

Tips? Ideas? Known examples? Questions? I'm all ears
This project is in it's very early research stages.
What is this concept even called? "Direct Hydraulic Control" or something? i have no idea, could use some help with that too.

[flip]

Anything mechanical is going to be inferior to optics.

[Isaac]

Anything mechanical is going to be inferior to optics.

Let's duel: You use the laser gun and I'll use any bolt-action rifle.

[snoopy]

I think flip is right.

You will have to worry about your bladder wearing out, plus you're transferring the energy elsewhere where you still have to measure it somehow - why add the extra step of energy transfer when you can read position more directly? I guess maybe you'd want it for the "look and feel" - It might be a decent concept for force feedback (though then I'd worry about latency).

So.... why do it this way instead of how it's currently being done?

[flip]

Anything mechanical is going to be inferior to optics.

Let's duel: You use the laser gun and I'll use any bolt-action rifle.

You lose , I win on line-of-sight alone.



@Snoopy. Yeah I thought so too at first, until I read the post on .com. Seems Roid wonders what it's like to play with a joystick you can actually grab . Either that or he is reaaaaaaaaal lazy.

EDIT: After reading Roids post on .com and Snoopys reply here, I got tears in my eyes! LOL!

[flip]

http://www.youtube.com/watch?v=A818bCjNetw

Hehe.

http://www.youtube.com/watch?v=h69TrQTseHA

http://www.youtube.com/watch?v=kEuB6h1LjPw

[roid]

It's a low power system, the only power input will be the human operator. I'm merely using the hydraulics as a fluidic lever, it doesn't need amplification which is why i can do away with valves. I don't want to rip anyone apart, this is a toy.
Optics would require electricity yes? I'm avoiding that kindof complexity. If this thing can be entirely pneumatic or hydraulic then it will require no batteries and no circuitry, only tubes. Like the internet - just a series of tubes .

Oops, I should have mentioned this is not a joystick for controlling a computer game! It's for controlling a mechanical device, like those small desktop syringe-actuated robot arms i linked to in OP. The idea is that the entire device (including it's linked controller) requires no batteries, and it's cheap.
My application for it has the interesting property that it'll be made almost completely outof balloons and tubes, silicone, latex. It will likely end up quite soft and compliant. The joystick itself could even end up as a wibbly wobbly tentacle - where instead of the bellows being below the stick they are actually INSIDE the bendy joystick - they're actuated as you bend the stick. Doesn't have to be this way, but it's one possibility...

It's just that the whole notion of a hydraulic joystick that directly actuates bellows (instead of valves), is an idea i've had for a while but have never actually seen done before. I wonder if anyone else has come across anything like it. Maybe i've never seen it because it's a terrible idea and doesn't work too well? It is a mystery.

...gotta buy some of those twisty modelling balloons they use for making balloon animals... AFK for a sec.
(edit: ok got em)

[Isaac]

Anything mechanical is going to be inferior to optics.

Let's duel: You use the laser gun and I'll use any bolt-action rifle.

You lose , I win on line-of-sight alone.

Yeah, I'd be on foot and I'd be using the terrain. You might hit me for a second, you'd mostly be hitting rock, bushes, and or trees. All it takes is one of my rounds to render your gun useless, which is an easy fixed target, by the way. That's IF you could find me. Also I'm fast on foot. ..

That's it! DBB paint ball match! who's in?!?

[Duper]

Our engineers use pneumatic actuators all the time here for our test equipment. It's quick and cheap. And with thousands of cycles (if not in the 10's of K's) per year, maintenance is low if not non-existent with those parts.

Sounds like a neat idea roid.

There's always the steampunk value too. ;D

[flip]

That sounds good in theory Isaac!

[roid]

Our engineers use pneumatic actuators all the time here for our test equipment. It's quick and cheap. And with thousands of cycles (if not in the 10's of K's) per year, maintenance is low if not non-existent with those parts.

Do they solely use the linear actuator types (pistons), or do they also use the flexible bellows type?
The bellows types i've seen all seem to be very heavy duty, their design is closer to what i'm talking about simply because they are bellows, but i don't need anything that heavy duty. I guess i'm looking for the lightest duty bellows actuators around. I'll probably be making my own entirely custom, but i'm looking for inspiration so i don't just repeat others' mistakes, and so i have an inkling of what kindof cheap-ass materials i can get away with using (and which ones i should avoid).

This is gonna be some super light duty pneumatics, even the pressure comming outof a tiny one hand balloon pump would be in the high end of the acceptable pressure range. Coz it's looking like balloons are going to be both the actuators and the control inputs, probably side-by-side 3 at a time in a sleeve.

it'll likely end up rather similar to this

[roid]

http://www.emeraldinsight.com/content_i ... 206010.png
this image seems to illustrate the original JOYSTICK idea quite well. Imagine that the tube outputs of those 4 bellows were connected to 4 other bellows and a 2nd joystick, so as you move the first joystick around, the the 2nd one mimics all it's movements. No power needed.
Now replace that 2nd joystick with whatever, a robot arm, etc.

The 2 balloons problem is really messing with my idea though, i want these things to inflate/deflate 1:1 in a predictable, haltable, LINEAR fashion dammit >:(
Maybe the elastic covering which will inevitably surround the whole structure will provide enough back-pressure to keep these connected balloons behaving themselves.
if i have to replace my balloons with inflatable silicone tubes it will likely require more pressure (and thus human strength) to inflate noooooooooOOOOdflkdjfasdf.

my main planning difficulties are in the actuator portion (ie: in the robot being controlled), not in the controlling joystick portion.
The easiest option for those actuators still seems to remain to use the balloons to operate a PULL mechanism rather than push. They can pull on a string (attached to the arm i'm actuating) when they DEflate. Bleh, i had my heart set on push, so it would operate similar to Festo's handling assistant But maybe ideas will change with tomorrows coffee...
whoops who left this mental broadcast running.


update: Experiments with threading a drinking straw across the middle of a long twisty balloon, inflating one end, and then squeezing that end to see what happens... have been going better than expected . This mechanism is proving to be quite promising.

[Isaac]

I knew it was only a matter of time before Roid would start showing-off his balls on the internet.

[Duper]

They use piston.

If you're using soft material from switch to device, be sure to account for loss of pressure between both points. i.e., don't set your bladder pressure @ what is needed. Tubing will fatigue over time and stretch a bit possibly causing a problem with the switch design.

[roid]

i finally found the joystick idea in use! (albeit it uses pistons, not bellows)
Behold: the De Fonbrune Pneumatic Micromanipulator
https://www.google.com.au/search?tbm=is ... anipulator
It has some pretty impressive capabilities, like single micrometer movement precision. They use it for manipulation of things under a microscope, like injecting stuff into cells.

[Isaac]

I don't mean to one-up you or anything, but http://imgur.com/NIHpSD4.png

[roid]

yeah push-pull wires, heavy, stiff. Similar to the tendon&bone mechanical model, typically used for control of small simple manned aircraft. AFAIK it has more friction than pneumatics, and IMHO it's unnecessarily heavy and (relatively) expensive. I'm not sure it's necessary for it to be so heavy duty for my purposes. Specifically: Introducing rotating parts and PULLEYS into my design may be troublesome.
Obviously the De Fonbrune Pneumatic Micromanipulator is also heavy and expensive, but it's design specifically called for heavy duty & high precision. My design doesn't.
(apparently the De Fonbrune design used pneumatics because it was high precision, perhaps it eliminated slack introduced by wire pulleys?)

Tendons are indeed an option, perhaps pulling against inflatable bones (physical compliance, weight, and cost advantages), but i'm trying to avoid it if i can. Balloons in their deflated state do act somewhat as tendons, i'm still seeing what advantages i can milk from these feature-combos before i decide to add extra parts.

There's many options.

[Krom]

You have to consider compression losses in the pneumatics, which could be upwards of equal or even greater than the friction losses from cables. Air can be compressed to take up less space, which means that you can't pass even a moderate amount of force through to the other end without compressing a large volume of air down significantly. The air will just compress in the lines if you are attempting to push something too heavy on the other end, which is the reason industrial pneumatics use an air compressor turned up very high to operate.

If you want to really pass more than a moderate amount of force down the lines you need hydraulics, because fluid cannot be compressed. IIRC hydraulics are actually the most efficient method of force transfer known, they have a lower loss than gears, wires, chains, belts, pneumatics or pretty much anything else. Perfectly sealed hydraulics also have some other benefits, since the volume of a fluid cannot be changed, as long as there are no gasses in the lines a hydraulic can be used to push and to pull with just one cylinder. Of course the disadvantages are weight, and higher material costs because otherwise the tubes/pipes themselves can change volume under stress which can reduce some of the benefits.

[Duper]

EXACTLY what Krom said. That was what I was trying to get out, but failed.

Static pneumatics aren't the best solution. If you're using air, you should be plugged into an active line. Then you need to worry about oil and moisture build up which is fairly easy to deal with, just adds more to your gizmo. If you do use an active line, don't use statics on the switch, use a bypass valve system.

[roid]

I don't want to rip anyone apart, this is a toy.
I'm not expecting losses to be a problem since such low pressures* are being used. But if i do have problems with losses - I plan on just increasing the size of the controller and it's multiple "input/controller" bellows, and the human operator will have to apply a bit more force.
I doubt i'll be approaching the limits of the comfortable level of exertion one can reasonably expect from a human operator to control this thing, but that limit itself would be a fascinating discovery (it's tempting to see what's possible, maybe another time...). And anyway there's contingency plans if it comes to that - so it won't be a show stopper.

*(i expect all of these pressures to be only barely above 1 BAR, certainly well under 2 BAR, ordinary balloons seem to meet my current needs *knock on wood*.)

Static pneumatics

Oh is that what this concept is called? Cool. But it's hard to eliminate hits from pages regarding static electricity from search results, urgh static electricity is a common pneumatics issue.

edit1:
woa, sudden realisation: if needed, i could use a combination of pneumatics and hydraulics within the same single-channel line, separated by a fluid-tight membrane located somewhere along the line. It would be useful if i needed a tweakable combination of elasticity and non-elasticity (ie: if i wanted hydraulics, but with a bubble of air for slight shock absorption. Or comparatively if i wanted mostly air for weight savings, but with a bit of hydraulics to reduce the shock absorption slightly).
i doubt i'll need this idea but whatever THE INTERNET IS MADE FOR TYPIN'!

edit2:
hmm, maybe i could use hydraulics if i could find thin enough lines. My previous thinking was that i was using balloons, and if i went with hydraulics i'd also use balloons, but now that i re-think it that's not really true. In actuality hydraulics would enable me to use much smaller membrane-type actuators than balloons, but still be flexible. If i use a thin, reversible (can routinely turn inside out), non-elastic tubing i could use it as a long, very thin, linear expanding hydraulic bellows actuator. Although smaller than a balloon it could have the same (or greater) power density, and because of it's thinness the contained hydraulic fluid wouldn't be much and thus wouldn't add much weight.
It would look like a long completely uninflated (ie: flat) balloon, it can inflate modestly but a major difference is that the material would NOT stretch to inflate to a full balloon's size. It would inflate like a BAG, not like a BALLOON. I'd have to source the material :-/. Any ideas? It'd probably be a plastic film, perhaps MYLAR tubing (cheap packing material), Mylar doesn't stretch much right? Probably the diameter of a pencil... hmm.
Flow turbulence through thin tubes may be a problem though, it may slow down actuation and make the whole assembly less nimble than i wanted. Increasing the tube diameter would help, but it all could end up a tradeoff between weight and nimbleness. That's ok though, i imagine i'll have a lot of room to maneuver in that particular tradeoff, i wouldn't expect problems.
heh i just remembered some balloons are made from Mylar, so i could still end up using "balloons" afterall.

[roid]

Hey if anyone's got some unused heat-shrink tube around, can you please try something out for me, try to turn it inside out? Either way (outside itself, or inside itself). Lubricant optional - may as well.
Different diameters would likely have different results, i'd expect the bigger sleeves and the flat types to have more success. I'm just curious if it's mechanically flexible enough for the tube to turn at least half-way inside out (ie: halve it's length).
From pics of it, some types seem promisingly thin :]. Of particular interest is the 10mm flat types.

[Sergeant Thorne]

What you should be asking is...

[Duper]

no... but it will make a great lint collector.

I saw that exact image with something else splashed across the front the other day. lol. gotta watch those movies again soon.

[roid]

[roid]

Hey if anyone's got some unused heat-shrink tube around, can you please try something out for me, try to turn it inside out? Either way (outside itself, or inside itself). Lubricant optional - may as well.
Different diameters would likely have different results, i'd expect the bigger sleeves and the flat types to have more success. I'm just curious if it's mechanically flexible enough for the tube to turn at least half-way inside out (ie: halve it's length).
From pics of it, some types seem promisingly thin :]. Of particular interest is the 10mm flat types.

update: heat-shrink tube doesn't work very well at all. :T
Too hard & completely non-elastic. Needs to be somewhat elastic to go inside out.

[flip]

http://www.activeforever.com/latex-rubb ... 7AodNyUARw