Air Cannon

[RyanN]

We're using a freon tank on our robot now. We used a PVC tank for over 3 years and never had trouble with it, but the team, including myself, were waiting on it to blow like a bomb anytime. Since this robot's main purpose is to demonstrate around people (most of the time little children) we changed the tank to metal with a blowout hole in case something goes catastrophically wrong. We are still using PVC for the shooter, and regular flexible hose for delivering the air, but all the pressure is behind the valve and really never on the cannon. We've also limited the amount of time that the valve can remain on for. I think its like 1/10 of a second or something so it uses less air, and also reduces the chances for something to go wrong. I think we will limit the pressure to 100PSI even though the tank can handle 300PSI. I'm not sure how much pressure our sprinkler valve can handle, and I don't want to find out either . With 60PSI I can shoot a t-shirt over 300 feet over a tree with with a 1/10 of a second shoot. Here are some pictures of our setup:

[FourPenguins]

I've decried the use of PVC tanks before, and I stand by that. I would never use PVC for storage of any more than atmospheric pressure. I've seen what it can do, and i know someone with the scar to prove it.
However, the odds of damage to the barrel of the cannon are fairly low as the air is bound to follow the path of least resistance. In a sealed tank, this would be any crack or weakness in the tank. In an open barrel or the pipe to an open barrel, this would be the opening in the barrel, with only the resistance of a light shirt.
Is there some risk of a leak, a crack, or a blowout? Yes. Would that risk exist with any material? Also yes, although perhaps it would be a smaller risk with heavier/stronger materials. Like everything, it's a case of acceptable risk.

[Rob2713g]

The air cannon we built is pretty simple. We already had a sprinkler valve (which limits flow to 110PSI) and used PVC rated at 280PSI. I'll bring up changing to ABS. I uploaded a picture here. Thanks for the advice!

[Otaku]

Definately change the air tank to ABS. What you (or whoever) was effectively holding was a bomb.

[Pat Arnold]

Originally Posted by artdutra04

It's actually against OSHA guidelines to use PVC for any kind of (non-buried) compressed gas use.

And easier way to think of the PVC vs. non PVC debate for air cannons is to compare acrylic to Lexan. We all know that acrylic may work for many uses on our robots, but we one needs to baby it in order to prevent it from cracking and shattering. If they used Lexan on the other hand, they don't have to worry as much about catastrophic shattering.

The comparison given is very helpful; I was unfamilar with OSHA's PVC use restrictions. ABS is clearly the better choice.

Since MARS can not meet the school's team liability insurance requirement, there are no current plans to remake or publicly use our launcher.

One additional safety consideration we had with our design was preventing someone being shot pointblank or directly, so we mounted it at an approx. 45 degree angle on our 2005 bot and discussed use of a remote enabler that an adult would control.

The launcher was a fun summer building project; liability issues, probably more than the build, have opened eyes about all the stumbling blocks encountered when trying something new.

[Jim E]

Liability Insurance.
Thats a biggie. Is it worth $1500/yr or more for $1M policy to shoot a T-shirt into the stands? I think not!

I've seen industrial accidents happen with stored pressure. They ARE NOT PRETTY. Fortunately, nobody was seriously injured, but the insurance is there for a reason.

For tank safety, there is an aluminum-welded tank on ebay for about $16.00. This should solve that issue. Contact me if interested,for the link.

Of Course the kids-will-be-kids factor is always in play. At my high school, they fired an air cannon after touchdowns. It was located on the flat roof of the announcer's booth just to limit who had access.

[Mike Nawrot]

When it comes to air cannons, there's one general rule of thumb I keep in mind: PVC+Pressure= Explosion and blood stains on the metal shop floor. Just don't do it. If you want a pressure tank without spending too much money, sears offers a $20, 5 gallon craftsman tank, rated for 135 psi. That's the equivalent of 13.6 feet of 3" PVC. That's a lot of air. Our cannon, which is based on this tank, shoots t shirts a good 200 feet, and we have a relatively bad barrel to tank ratio (there's a lot of wasted air, but having a massive barrel is more trouble than it's worth. We're working on it). The operating pressure is limited to 100 psi by the kit's auto-shut-off switch, and just in case, the relief valve is at 120 psi. The only PVC component that holds any pressure is the sprinkler valve, which is rated for 150 psi, so there's a decent safety factor. It's all mounted on a robot, so it's remotely operated. I'd consider it safe, but there's no such thing as too safe when it comes to pressurized air. So general rules:

1. A safety factor of 1 generally isn't good enough. It's like having exactly 0" clearance between moving parts. It works on paper, but that's it.

2. Nobody stands in front of the barrel. Everybody should be behind the cannon, except for those in the crowd, but by then, air resistance should have slowed the projectile enough to prevent it from hurting anyone.

3. Test everything countless times before using the cannon in any crowded places.

4. Having a safety relief valve is a good idea

5. PVC DOES NOT SAFELY HOLD PRESSURE. Even if it says it's rated for pressure, that's at standard operating conditions. PVC will shatter easier in the cold, and even if it's at the rated pressure, any sort of unusual stress or strain will cause it to explode, leaving nasty cuts on anyone in the blast radius. From what I hear, stitches on your face aren't fun.

Overall, accidents can be prevented with the use of common sense, and by reading the labels. Have fun, be safe, avoid hospital trips, medical bills, and law suits.

[Chris Fultz]

think about these things -

what happens when you launch at the wrong angle and shoot someone (maybe a small child) with the rocket er. I mean shirt, at close range?

Are you willing to stand in front of the launcher and take a direct hit to the stomach, leg or head? If not, why do you want to risk 'letting' some by-stander?

Most sporting events have returned to throwing shirts into the crowd - there is a good reason.

There is a hole in a ceiling tile at the top of a high school gym (40+ feet up) where a shirt hit. Suppose that had been someones face?

[Schnabel]

Originally Posted by Chris Fultz

There is a hole in a ceiling tile at the top of a high school gym (40+ feet up) where a shirt hit. Suppose that had been someones face?

Actually this adds to that point. The sad thing was that the ceiling tile that broke from our cannon was just air. The shock wave from it did that at only 60psi. I know when we did it at IRI, we would make sure everyone in front of it almost to the second level was sitting down plus we would block the crowd off both ways. I still felt uneasy shooting where we did.

[Rob2713g]

WOW! I'm glad I asked this question. I definitely didn't know parts of this were so dangerous. We are changing to ABS or a non-PVC tank and will do extensive safety tests.

Ours doesn't launch with lots of power. At a 45 degree angle and 110PSI, a shirt goes approximately the height of a 2 story house. Last Thursday, we tested it by first hitting a mostly empty trashcan, which it did not knock over (but did rock). One of the parents then volunteered to get hit at close range (10ft) and was fine. Then some kids got hit at catching distance and were fine. The lack of extreme power in our launcher may be due to the barrel being so much larger than the tank and the shirt not being too tight in the barrel (allowing air around it). We'll try to be careful and scrap the plans if it looks dangerous.

For those that do launch at football games, pep rallies, etc. what other safety precautions have you taken?

[Otaku]

Originally Posted by Rob2713g

Ours doesn't launch with lots of power. At a 45 degree angle and 110PSI, a shirt goes approximately the height of a 2 story house. Last Thursday, we tested it by first hitting a mostly empty trashcan, which it did not knock over (but did rock). One of the parents then volunteered to get hit at close range (10ft) and was fine. Then some kids got hit at catching distance and were fine. The lack of extreme power in our launcher may be due to the barrel being so much larger than the tank and the shirt not being too tight in the barrel (allowing air around it). We'll try to be careful and scrap the plans if it looks dangerous.

110psi = Lots of pressure. I wouldn't go above 60, or hell, anything above 1 bar with PVC (1 bar = average atmospheric pressure, 14.7psi).

Keep in mind, the short length of the barrel is why your launcher isn't "powerful". Firearms can acheive higher projectile speeds with longer barrels because of the physics involved, and the same thing is applied here. The longer barrel means that the projectile (in this case, a shirt) has pressure applied to it for a longer time. The short barrel means that the shirt gets a quick blast of pressure (which I'm sure some of it is lost because I'm assuming you don't put the shirts in a break-away casing that would create a better seal) before the air can escape the barrel and dissipate, no longer imparting it's energy upon the shirt.

Now, for a much safer system...

Use the FIRST air tanks and mount them to a dedicated launcher-bot (simple frame, tank drive, 2/4 air tanks, etc) that has a pneumatic system for an ARM instead of a cannon. Do the math and figure out the general trajectory of the different pressures (one bar at a time, ie 15psi, 30psi, 45, 60, 75, 90, so forth) as well as the boom ratios (1:2 ratio? 1:3? What can your pneumatics handle? How well does it work?) and such.

This way you know the trajectories it launches at, about how far and such, and it's as controlled as it can be every time.

-Otaku

[artdutra04]

As for the air cannon safety issue, the answer is pretty simple. Just don't wrap elastic bands around the tee shirts. If you just roll them up, if you shoot them at a high angle* shot (like a mortar), they will just open up in the air and flap and drift downward at a very safe speed. You'll only get about half the range of what you normally would, but it'll be a lot safer.

* Another safety benefit of shooting tee shirts this way is that with very high angle shots, it becomes almost impossible to hit anyone at point blank range. A cannon with a three foot barrel, held four feet above the ground, at 60 degrees above the horizontal, means the end of the barrel is six and a half feet above the ground. Unless you're Yao Ming, is a fairly safe launch position.

Originally Posted by Otaku

Now, for a much safer system...

Use the FIRST air tanks and mount them to a dedicated launcher-bot (simple frame, tank drive, 2/4 air tanks, etc) that has a pneumatic system for an ARM instead of a cannon. Do the math and figure out the general trajectory of the different pressures (one bar at a time, ie 15psi, 30psi, 45, 60, 75, 90, so forth) as well as the boom ratios (1:2 ratio? 1:3? What can your pneumatics handle? How well does it work?) and such.

The FRC-legal pneumatics are too puny for air cannons. You need at least 3/4" NPT fittings to get a flow rate good enough to shoot a tee-shirt a considerable distance; the biggest fittings in the KoP are only 1/4" NPT. (From being bored at a meeting experience, no matter how many tanks and hoses in parallel you add, if you limit yourself to only KoP components, don't expect to shoot much more than a ping pong ball.)

[Qbranch]

Originally Posted by artdutra04

The FRC-legal pneumatics are too puny for air cannons. You need at least 3/4" NPT fittings to get a flow rate good enough to shoot a tee-shirt a considerable distance; the biggest fittings in the KoP are only 1/4" NPT.

Uh oh... i was trying to design one of these air cannons on a dime using a 3" internal diameter barrel 12" long (abs plastic, .25" wall thickness) and was going to just use four of the 16cu in clippard air cylinders as air storage.... I was not however using standard IFI air solenoids, i got my hands on some used (free) Mac Valves with a Cv constant of 30, 1/4" npt.

With the hose included, if me and my Machinery's Handbook calculations were right, I should get an output rate from each clippard/valve pair of 38CFM, or 152CFM for all four pairs. If i switch that over from CFM to cubic inches/second it is 4377.6 cuin/sec. The volume of my barrel is 84.823cu in, so that should fill the barrel (and eject the tee shirt) in 0.0193 seconds... which is 35mph.

Was this velocity not high enough? It stands to reason if i double the cylinder/valve pairs i should be able to get this up to 70mph if it isnt. Or, is there another cheap alrternative (remember clippard cylinders are cheap and the valves i'm using are free (have 10))? What's the cheapest way to switch that much air on with one valve? Whats this I hear about irrigation valves?

Sorry, i'm new to the T-Shirt cannon (high flow pneumatics) business... Any help would be much appreciated.

-q

[Otaku]

I didn't mean for you to make an Air cannon with FRC parts. I meant to make a pneumatic catapult. Since it doesn't have to be FRC-legal, you could bump it up to 1/2" ID tubing if you need to (But I'm sure everybody has tons of the 1/4 stuff)

[Qbranch]

Originally Posted by Otaku

I didn't mean for you to make an Air cannon with FRC parts. I meant to make a pneumatic catapult. Since it doesn't have to be FRC-legal, you could bump it up to 1/2" ID tubing if you need to (But I'm sure everybody has tons of the 1/4 stuff)

Well, yeah thats my problem. I have a lot of 1/4" fittings and such... however, I'm thinking I'm going to switch to a Craftsman 5gallon pressure tank instead of the clippard cylinders... so i'll probably be moving up to 1/2"

So.... does any body know where i can get a high flow solenoid for 1/2" for cheap?

thanks,

-q