First time building a flipper robot

[Danny B]

After attending last year's Doncaster event, and being hugely impressed by Eruption, I've decided to build a flipper robot, but I've never worked with pneumatics before.
I need a basic checklist of the parts I'll need, and what size and rating would be ideal for a featherweight. I do know that a 12oz paintball tank is ideal for storing the CO2, but that's about all I know.

[maddox10]

I hope you realise that a machine like Eruption uses a ram that is total custom build, and that the force of simular machines varies between 400 and 2000 kg on the ram.

[Max]

I would read through this thread:
http://www.fightingrobots.co.uk/thre...ght=Pneumatics
It has tonnes of information and questions already asked and answered about pneumatics. Also search for similar threads and read them.
You should especially read through this website:
http://www.teamdavinci.com/understanding_pneumatics.htm
It is absolutely brilliant and explains everything you need to know.

Once you've done some research then we can answer more specific questions.

[Danny B]

Thanks for the links, they've kind of helped steer me in the right direction. I did still have a couple of questions I was a little stuck with.
- I plan on running my system at 10bar. Does that count as LP or FP?
- One thing I'm worried about is the system freezing. Is there anything in particular that I need to do to avoid this?
- I want to work out how much force the ram will generate, but I'm really struggling to get my head around the mathematics. Can someone simplify things a little?

[typhoon_driver]

pressure = force / area

Take the area of the ram and multiply by pressure to get the force.

What's with the creepy pony? You aren't one of those oddball bronys are you?

[maddox10]

10 bar is Low pressure.
Most commercial pneumatics are most happy @6 to 8 bar.

Some brands have a line of higher pressure equipment, and those are set at 16 bar max.

Anything over 16 bar is seen as high pressure.
Full pressure means unregulated CO2, or 1000psi compressed air. (don't know a machine that is using that atm)

Freezing is the result of using CO2. It's 3/4 liquid in a full tank, and it needs heat to evaporate. This cools everything down, lowering the pressure CO2 needs to be liquid (55bar @20°C, dropping to 0 bar @-79°C).
This fact, combined with the inability of commercial pneumatics to cope with liquid or cryogenic temperatures makes a system "freeze".
Several ways are used to combat this. Main thing is to avoid liquid CO2 entering the setup.
Syphon tubes, orientation of the bottle, buffertanks, heatsinks and so on.
Some machines cope fine, others struggle.

Force of a ram.

Most rams are round (better for pressure reasons and easier to make). Like Gary writes, surface*pressure = force.
Radius*radius*Pi is the surface of the piston. Bar = 1 kg per cm²

Example. 50mm bore ram. 2.5*2.5*pi*10=193 kg of force.

[Danny B]

How much force at the end of the arm do you think would be enough to be competitive amongst other feathers? I'm not trying to set records, I'd just like enough power to turn someone over.

[Danny B]

So presumably a ram with a longer stroke would need a larger buffer tank? And should I be worried about longer ones bending?

[typhoon_driver]

Ok i'm out. Have fun.

[Anttazz]

Not always, its really dependant on volume rather than stroke length. If they're designed to handle the pressure you run them at you shouldn't worry too much about it bending. It may happen, but so might anything else when you're fighting other bots. Not sure if you've seen our flipper on the Facebook page, but it runs a tiny ram on the lowest pressure possible on a Trev regulator and it still manages to turn robots over, maybe not every single time but because it uses so little gas you get a fair few cracks at getting them over in our demo environment.

[maddox10]

Buffertanks.
Industrial guideline is that a buffertank has at least 5 times the volume of the used volume per action. But that would make it very large for our purposes, so I give myself the guideline of 3 times used volume.

In a low pressure setup, the buffertank doesn't do a lot to combat freezing as it has the be in between the regulator and valve.

Ram stroke and bore.
A long narrow ram or a short fat one deliver the same amount of power, if the volume is the same.
Problem is, commercial long narrow rams have smaller inletports, and those restrict flow.
As you remark, long narrow rams are more vurnable to bending forces.
I myself try to keep stroke to a minimum, with stroke up to twice the bore (a 50mm ram will have 100mm stroke or less.)
On the other hand, that's not commercial stuff.

Amount of force on the tip of the arm.
Hannibalito 3 does about 200 kg. Hannibalito 4 goes over 400 kg. But those are Full Pressure setups.
I have seen robots with linaks (electric "rams") pushing 25 kg on the tip and win fights. Not fast, nor powerfull, but very controlable and easier than pneumatics.

[Danny B]

Ok, I think I have a basic design figured.
It's inspired by American middleweight T-minus, and I think it should provide a relatively compact shape capable of both running inverted and self-righting.
The whole machine should be about 35-40cm long, with a 50mm bore, 100mm stroke ram powering the weapon. According to my calculations, a 10bar system should deliver about 60kg of force at the tip of the arm.
Robot outline.png

[maddox10]

I think, if I follow the drawing, that it will be more force on the tip.

How high goes the tip according your plan?

[Danny B]

Approximately 30cm.

[maddox10]

Sounds plausible.