DIY 2-stage light gas gun
Dave
make a 2-stage light gas gun when we visited NextFab. Apparently sling
shots and BB guns (along with anything else that's fun ;) are illegal
in Philadelphia and most surrounding areas. So we'll scratch that. I
plan to continue with the work but it looks like it'll have to be on
my private property in safe enclosed spaces. Laws are funny. I can
legally have 50 pounds of black powder at my house and build rockets
that could literally launch me over a building, but sling shots are
illegal.
For anyone interested, I've got a good start on a prototype built
already. Picture a big syringe -- load a BB into it and smack the
plunger with your hand, and if you hit it hard enough, the BB shoots
out with enough speed to become embedded almost flush into a block of
wood. Kind of amusing for something so crude. Once it has the rest of
the important components (light gas, burst disk, energy storage, and a
driving mass), it should be pretty rad. I guess I will consider
breaking Mach 1 some measure of success, but ideally we should be
talking about a couple of times that without too much trouble. Mach 30
is possible but this project isn't my life's work. ;)
I'm getting close to the point where I want to measure the muzzle
velocities so I can quantify the improvements as I make them. I don't
suppose anybody on the list would have access to a real chronograph,
would they? We could build one but it's annoying getting 2 sensors to
trip when a BB (0.177", or 4.5 mm diameter) flies by without risking
the sensors getting hit by the projectile (been there!). I've used
trip wires before with fairly poor success, though I'm sure I could do
better. I am kind of tempted to just keep it old school and make a
ballistic pendulum. Basically you weigh a block of wood and you weigh
the projectile. Suspend the block of wood on string to make a pendulum
(maybe 4 strings of equal length so it stays parallel). Fire the
projectile into the block of wood and measure how high the block of
wood and newly-embedded projectile swing, and you can calculate the
velocity. Nowhere near as accurate as a true speed measurement but
it's almost silly how easy it is to set up. Anybody got any other
thoughts?
Maybe something like an array of tiny photovoltaic cells or
photodiodes that are in a series configuration (photoresistors don't
respond fast enough). Shine a bright light at all of them and the
signal going through them should be uninterrupted, but break the beam
to any one of them and the output will shut off. You'd probably need
to use point light sources to ensure that the BB cast a good shadow.
Maybe if the point source was relatively close to the line of fire,
and the sensor array was further away, the BB would cast a large
enough shadow to guarantee that it occluded at least one sensing
element. Maybe I could just use LEDs as the sensing elements, since
they're so readily available.
Any thoughts? I kind of want to try the ballistic pendulum just
because it's so dead-easy, but I know it won't be that accurate.
-Dave
pezman
pulse.
I do like the photo-sensor idea -- easy enough to project a line onto
a row of photocells. if you have an even number of the,, then any
shadow connected to the middle "tap" would cause a voltage blip up or
down.
A single-sensor stereo-photograph (using mirrors) with a fixed shutter
speed (say 1/1000 second ) could be used to simultaneously compute the
distance from the camera and the speed of the object (based on blur
length). If the shot were taken immediately upon exit from the gun,
then the distance would be fixed, so no stereo image needed. Mach 1
is almost exactly 1 foot per ms, so the blur length in feet is the
approximate Mach number. Alternatively, could use a close view and
faster shutter speed -- bbs are pretty small.
I have been thinking about a "virtual" burst disc -- say a super-
magnet covering a port in a steel disc. When the pressure gets to a
certain point, the magnet gets blown free (and maybe retained in its
free position by another steel disc or a companion magnet or
whatever. In design, I imagine a right-angle port. The magnet holds
a small plastic rod in a cylinder of steel with a relatively small
bore in its center. When the pressure gets high enough, it pops the
magnet from the face of the rod -- far enough to expose a port at
angles to the rod. This port communicates with the bb barrel and lets
the pressurized, light gas do its thing.
Matt Chernak
--
To post to this group, send email to hive76-d...@googlegroups.com
To unsubscribe send email to hive76-discuss...@googlegroups.com
For more awesome goto http://groups.google.com/group/hive76-discussion?hl=en
Dave
numbers don't like turning corners though, it makes a huge difference
in gas guns. Right now I'm using a magnet to hold the BB in the barrel
and I was figuring I could just use a super strong magnet and the
projectile could basically BE the burst disk. My current barrel was
just some tubing I had laying around tho, so it's not a very close fit
and will let lots of gas escape. Maybe still worth a try tho. It won't
be anything like a stainless steel burst disk, but it might still be
cool.
I actually did some stereo high-speed BB photography years ago, and
it's hard to capture. BBs are tiny and really reflective, so you just
see a tiny dot. For my best shots I painted the BBs with bright orange
spray paint, otherwise they were impossible to see. It made a cool
effect though because on impact the paint would blow off in what
almost resembled sparks or a small explosion, and then I'd find the BB
with no paint left on it. But shutter speeds like 1ms are way too slow
to catch anything good unfortunately. I had to dial my BB gun down
almost all the way to get very much at all. It's a harder problem than
it sounds. I eventually cheated for my shots of unlit light bulbs,
running a very small current through the filament and triggering the
camera flash when the circuit was broken. The BB itself didn't have to
even hit the filament because the glass debris would take it out every
time. Fun stuff.
I'm leaning towards using a string of LEDs as photodiodes and then a
point or line light source (laser pointer through a glass rod, or
whatever). It's gotta be sensitive and it's gotta be seriously fast
tho. Some photodiodes may not even be fast enough. Will have to
test...
One good idea I saw online is to just pipe the sensor output to an
audio input and record. 44kHz is the low end of standard these days
and that's pretty fast and low effort. Not as fast as a
microcontroller but seeing the whole history of samples could be
helpful in debugging.
-Dave
Dave
terminal velocity of a BB is something like 60 mph or so, very low.
When projectiles that are as draggy and light as BBs 'splash' into the
atmosphere they slow down really fast. I did some back of the envelope
rough calcs and it was looking like if you shot a certain BB-like
projectile at, say Mach 10, it might only get like a hundred yards
before it was only going Mach 1, and in another hundred yards it would
be down at its terminal velocity. Kind of interesting.
I'm not sure how the mic would work. Seems possible to get fooled if
there were high-speed gases in front of the projectile, or if it was
subsonic (the 'bang' would be faster than the projectile).
Certainly it would be possible to put a mic or piezo element or
something on a plate that you fire directly at... that tells you when
to stop your clock. For steel BBs, you could use some sort of Hall
effect or other magnetic disturbance sensor to detect when the BB was
at a point inside the barrel to start the clock. Forget that, maybe
just put a photo interruptor right on the end of the barrel itself,
where you know for certain the BB will pass? Seems reasonable
enough....
-Dave
On Dec 12, 9:59 pm, Matt Chernak <matt.cher...@gmail.com> wrote:
> What sort of distance will there be between the gun and target? Could audio
> work instead of light? Maybe set two separate mics, one at the muzzle, the
> other at the target. If you record them onto two separate
> tracks simultaneously, you can zoom in *way *close (48 or even 96 kHz) and
Matt Chernak
Jack Zylkin
usbtypewriter.com
Philadelphia, PA
Dave
once and couldn't get anything usable out of it (very small voltages).
I also kind of lean towards the ideas that don't lock you to a certain
type of projectile (ferrous, metallic, shiny, etc).
-Dave
On Dec 13, 12:21 am, Jack Zylkin <jzyl...@gmail.com> wrote:
> The only problem with pez's induction loop idea is that the eddy
> current will slow down the BB.
>
> On Tue, Dec 13, 2011 at 12:06 AM, Matt Chernak <matt.cher...@gmail.com>wrote:
>
>
>
>
>
> > If the distance (one foot) is that low, then yeah, mics would probably not
> > work. I was imagining this gun firing through a back yard or a field.
>
Dave
as you envision it and I can try a subset of it if I have the
materials on hand and a means to test it. Maybe I'm picturing it
differently. Even if it drops the speed by a few feet per second,
that's no big deal to me, it's probably still a lot more accurate than
a ballistic pendulum.
-Dave
pezman
the Mach number.
> ...
>
> read more »
pezman
> ...
>
> read more »
Dave
picture taken at the right time will be tricky on its own. But that,
or even just an old-school shadowgraph, would be some pretty cool
documentation for the project.
Apparently that 'trillion fps camera' is more like a stroboscope so it
doesn't help if you are trying to capture a single non-repeating
event.
-Dave
On Dec 13, 10:03 am, pezman <mikehoga...@gmail.com> wrote:
> ...
>
> read more »
pezman
Dave
to synthetic) schlieren imaging is crazy, like capturing the heat
rising off a person's hand. I can't afford the mirror you'd need to do
something like that though. Provided I was in the trans-sonic region
or higher, synthetic schlieren should work (like in that rifle image).
Documentation-wise that would be incredile though. Probably not a
super precise way to measure speed on its own (I'm guessing on par
with a ballistic pendulum, it'll get you in the ballpark), but
definitely cool as hell.
If you want to do an experiment let me know.
-Dave
On Dec 13, 2:45 pm, pezman <mikehoga...@gmail.com> wrote:
> http://www.sciencephoto.com/image/89118/large/C0024857-Rifle_shot,_sc...
> ...
>
> read more »
pezman
make).
> ...
>
> read more »
Dave
-Dave
> ...
>
> read more »