GGDT/HGDT Request Thread

[High-PSI]

One last thing;

The muzzle velocity goes up with a smaller diameter barrel that is longer (but the same volume). The velocity goes WAY up. Is tha because the air has a straighter shot into the valve?

Matt

[Mr.Tallahassee]

It's because the longer barrel gives the projectile more time to accelerate. This only works to a point.

[High-PSI]

Oh, geeze, I am sorry, I meant longer chamber (tank).

I understand a longer barrel increases FPS, but I was unaware than a longer and narrower tank would do the same thing.

Matt

[MrCrowley]

It might have something to do with the flow into the valve. If you have a chamber diameter much larger than your valve diameter, it would probably create turbulence and may disrupt flow whereas a chamber that is the same diameter as the valve can easily flow through it with minimal turbulence.

Just a hypothesis off the top of my head, I know little about flow dynamics.

[Mr.Tallahassee]

Actually, that does make sense. It doesn't have to squeeze in several directions as much so all the air is going in the same direction at the same time if the chamber is the same diameter as the barrel.

[High-PSI]

That is what I was thinking. Kind of like cramming too many people through a narrow door at the same time. If they are single file, they go through faster.

Anyway, it is quite noticeable when I modelled the gun.

Oh, hey check this out, I ran the numbers on my big 2 inch cannon; Turns out, with a 4 inch tank (instead of my 6 inch tank) at 400 PSI, that big cannon will put out 2,200 ft pounds of energy!

Geeze, I think that would hurt my shoulder.........

Matt

[ramses]

Actually, that does make sense. It doesn't have to squeeze in several directions as much so all the air is going in the same direction at the same time if the chamber is the same diameter as the barrel.

I'm going to go out on a limb, and say that GGDT should not recognize that, since it is a 0D simulation (gas pressure in the chamber is assumed to be a constant.

I don't now know if his valve flow model would include that effect, but it would take a 2D model at least to recognize turbulence directly.

[Mr.Tallahassee]

Here is the website for GGDT. It's a 1D simulation. It doesn't take turbulence into account in it's programming but it does take choking flow into account which could also be the reason smaller diameter chamber works better. This is based on what I read in the FAQ section on the site.

[DYI]

While the website claims that GGDT is a 1D simulation, it is only in the most literal sense. In the terminology of internal ballistics models, it is, as Ramses stated, 0D. GGDT treats the gun as two bodies of gas, the chamber and the barrel. At a given time step, pressure throughout one of these gas bodies is taken to be constant. This implies an infinite speed of sound and a gas which has no mass. This is the main cause of GGDT's problems with trans-sonic flow - while it takes at least a 2D model to give the problem a proper treatment, this model does not represent anything approaching even an ideal gas, much less a real gas. GGDT is essentially an unsteady orifice flow calculator combined with some adiabatic expansion based ODEs to represent the chamber and barrel gas volumes. It is reasonably good as 0D models go (except for its apparent inability to adjust its time step appropriately to prevent this), but it is NOT a 1D model in the true sense of the phrase.

The reason for that phenomenon with high L : D aspect ratio chambers is, if I remember correctly, an issue with the flow choking. I don't think GGDT allows for flow choking by the chamber, and the effect you're seeing will not show up in reality. Quite the opposite, in many cases. This may, however, not be the exact cause. There are a lot of "fudge factors" in the program to attempt a better representation of realistic behaviour than the 0D structure naturally allows, and this may be caused by one of them. Mr. Hall is very tight-lipped on the matter, but there's a thread in the old SGTC archives that shows the "main loop" GGDT uses to solve the ODE system representing its model, if you're interested in taking a look.

TL;DR: the effect you're seeing in GGDT is not realistic, and neither is GGDT. It models low performance pneumatics well, but don't go trying to draw general conclusions about fluid dynamics from its outputs

[jackssmirkingrevenge]

The muzzle velocity goes up with a smaller diameter barrel that is longer (but the same volume). The velocity goes WAY up. Is tha because the air has a straighter shot into the valve?

Could you post screenshots, there might be a parameter you're/we're missing.

[Selador]

Wanting to make a compact full-sized pnuematic coaxial.

I don't really think that ALL of this data is necessary to answer my basic question, (What is the optimum barrel length for a 1/2" pvc barrel, in a ~18 inch length 2" pvc coaxial chamber.), but I'll do my best to fill in the data requested.

Reservoir:
Gas = Oxygen obviously.
Temperature = ~65 F
Pressure = ~130 lbs
Outer Diameter = 2" (ID of pipe.)
Inner Diameter = 3/4" (OD of barrel)
Length = ~18"
Volume = ??? Isn't that one of the things we are trying to determine here ?

Valve:
Type = ??? What do you call a valve that seals against the end of the barrel in a coaxial ?
Flow Coefficient = ??? Beats the heck outta me ! (In other words, I am the manufacturer, and for the life of me, I can't get myself to write up some literature with that answer in it. My customer service is obviously not the best in the biz. I think I'm going to have to subtract a star or three when I go to rating my experience with me, as the manufacturer.)
Seat Diameter = Ok, I can kinda answer that one. It will be a one inch valve, riding inside a 1-1/4" pipe, sealing up against the end of a 1/2" barrel.
Open/Close Time = Purty durn quick. IOW: As fast as I can make it do so. LOL
Dwell Time = Forever. IOW: it opens and dumps the contents of the res. Then it stays open forever. Until, of course, I fill it again, which closes the valve again.
Dead Volume = I have no idea what that means. I don't think this thing will bring down a zombie, if that's what you mean.
Piston Diameter = Easy enough. 1-1/4"
Piston Mass = It ain't catholic, so I don't think it's going to be going to mass. As for the overal weight and volume... I have no idea. I haven't made it yet. I'm waiting for this request to be answered, before I even start.
Vent Diameter = Big. Going to be eixiting via a 1-1/4" pipe, and sprinkler valve. I'll make the volume behind the piston as small as possible. I don't want JSR to be stomping in my cheerios about wasted volume. LOL Bottom line, I'll be aiming for dumping the volume behind the piston almost instantly.
Pilot Volume = Like I said as small as possible.
Hamr Force = enough to give you a bruise if you are foolish put some part of your anatomy in front of it.
Hamr Dist = Actually, this isn't even a hammer valve, is it ? Ok, never mind these last three. LOL
Hamr Mass = See last answer.

Barrel:
Bore = 5/8" (Hey, I never knew 1/2" pvc pipe was actually 5/8" inside diameter !)
Length = ??? This is the question I want answered. What's the optimum length barrel, for a coaxial with an 18" x 2" chamber ???

Projectile:
Friction = Almost no pressure at all, to get it started moving. (Works REALLY well, just as a simple blowgun.)
Mass = Ok, ok, I'll consider trying to get my projectiles to convert to catholisism. I don't know why you guys are so hung up on going to church ! Meanwhile, they weigh maybe 1 to 3 ounces at the very most ??? A drywall screw and little plastic caplike thingy.
Diameter = Almost EXACTLY the same as the bore of the barrel. So, 5/8".
Init Position = snugged in against the piston surface once the piston is closed against the barrel.

Compute Performance
When you've got all your input data to your satisfaction, push this button and GGDT will do it's thing! Be patient though. GGDT takes a fair amount of CPU cycles. It may take a few seconds for it to return with it's answer.

If this takes more than a few seconds, I want my money back !




Thank you very much, in advance.

[DYI]

With a 1 oz. projectile, optimal barrel length is 570" returning a muzzle speed of 573 ft/s. With a 3 oz. projectile, optimal barrel length is 703", returning a muzzle speed of 379 ft/s.

Now, seeing as this information is entirely useless to you, perhaps we should address the more fundamental problem of why you're unable to run GGDT for yourself. Emulators are available in most cases of operating system conflict, or you could run on a computer which already has Windows installed. If you already have a relatively recent version of Windows, it may be that UAC is preventing GGDT from being run. This is usually solved by running the program as an administrator, although at one point I had to disable UAC entirely.

The reason these numbers are useless to you is that you're left with no feel for how performance varies as a function of barrel length. Quite frankly, unless you're trying to make one of those lovely hissing pneumatics the Mythbusters seem so fond of, a chamber as large as you're using is entirely excessive for a 0.625" bore. It would do you much more good to develop a solution allowing you to use GGDT for yourself.

As to the few questions you had: the valve model you're looking for is a barrel sealing piston valve, the dead volume is the valve volume downstream of the valve which is not yet in the barrel, the flow coefficient will be around 40% (I used 45%, but it doesn't make a great deal of difference with excessively long barrels), and the seat diameter is the outer diameter of the pipe the piston seals up against (in the case of a coaxial gun, typically the same as the barrel OD).

I'm also curious as to your propellant gas; pure oxygen is, as a rule, NEVER used in pneumatics, for safety and practicality reasons, and is not even included as a standard gas in GGDT. Using it in the model necessitates entering a bit of additional data - is that where you got hung up?

[Lockednloaded]

2"x36" barrel
3/4"x20" chamber
168g Clementine CD close to .6
And
1.68" x 36" barrel
46g golfball CD Of .3

Both at 500 psi and 300 psi

Valve:
valve type - generic
# of valves - 1
Flow Coef. - 5
Seat Diam - 0.75 in
Open Time - 20 ms
Dwell Time - 999 ms
Dead Vol - 1 in3

Thanks in advance for the help

[Gun Freak]

I put 5 and 40 for the flow coef becuase I don't think you really mean 5...

[Lockednloaded]

Thanks GunFreak, I think I just copied and pasted that info from somewhere else, but it was supposed to be a 3/4" QEV