Page 1 of 3

Regulator flow rates

Unread postPosted: Sun Dec 13, 2009 12:35 pm
Author: kjjohn
I am currently working on designing and building a very advanced rapid-fire hybrid cannon, and I need to know if the regulators will be able to fill my chamber quickly enough. I will be using a 20x mix of MAPP and pure oxy inside a 10 cubic inch chamber. I need my MAPP and oxy regulators to be able to fill this chamber with 36psi (30psi oxy, 6 psi MAPP) within a fraction of a second. I have already worked out the issue of the pressure difference between the MAPP line and the Oxy line, I just need to know if a standard oxy regulator/MAPP regulator will be able to fill the small chamber within, say 1/4 of a second. If not, are there some other MAPP and Oxygen regulators that will?

Unread postPosted: Sun Dec 13, 2009 1:49 pm
Author: rcman50166
The regulator itself, by design, has a limited flow rate. It takes exponentially longer time per unit pressure as it approaches equalization between desired pressure and regulated pressure. To overcome this you need a holding tank of sorts after the regulator to have a "capacitor" effect if you will. The air coming from the regulator will continuously fill the holding tank while the hybrid takes up-to-pressure air from the tank.

Unread postPosted: Sun Dec 13, 2009 1:55 pm
Author: kjjohn
A pressure accumulator sounds like an excellent idea!! How big would the holding tank have to be? I am planning on having this be a full auto hybrid (using a type of blowback bolt with a piston valve). It will have a fairly low rate of fire - 5 rounds per second max), but it will still have to require a very high flow rate.

edit: changed capacitor to accumulator, D_hall

Unread postPosted: Sun Dec 13, 2009 1:56 pm
Author: D_Hall
@RC: What model did you use to generate that curve? The initial pressure ramp for a regulator should be pretty linear as flow is choked. Depending upon the supply pressure and reg set point, it may decay as you show (if supply and set are reasonably close so flow is no longer choked). It may decay linearly. It may do all sorts of things depending on the specifics of the system.

Admittedly, this is not a problem I've studied in depth, but... Well, just looking for more explanation of where that plot came from.

edit: Oh, and when you're dealing with pneumatic systems, you don't call them capacitors. You call them "accumulators."

Unread postPosted: Sun Dec 13, 2009 2:11 pm
Author: rcman50166
The plot itself is just a negative inverse plot. I did no modelling to generate it. It is just a generic chart. (Hence no units) Logically it is the way a pressure regulator would work. I've also noticed this trend when I used a regulated pressure compressor for my T-shirt gun. Admittedly, the model better represents a regulator filling a pressure vessel. I could model it. However seeing as I am taking finals at the moment I'd rather invest my time studying.

edit: Oh, and when you're dealing with pneumatic systems, you don't call them capacitors. You call them "accumulators."

Note taken :D

Unread postPosted: Sun Dec 13, 2009 6:17 pm
you could use fluidsim to model it... it's not as useful for us as GGDT but still you can simulate quite a lot of things

though you would need to know the exact flow rate of your reg and valves for its results to be meaningful

accumulator will help... but I guess that you would have to set the pressure a little bit higher than what you want...

pressure you need to set * (volume of the accumulator + the meter) = volume of the accumulator * pressure you want to reach

how do you plan to control each cycle ? with the use of electronics, right ? you can also build and test some simple electronic circuits with fluidsim, you know... nothing advanced but enough for your needs

Unread postPosted: Sun Dec 13, 2009 7:15 pm
Author: kjjohn
Actually, yes, a lot of the fuel injection system will be automated with some simple sensors that monitor the position of the bolt, as well as a main control PCB. I am planning on using some kind of IC to make it selective fire, and the ignition will be from a high voltage pulsed coil.

In terms of meters, there will be no fuel meter since the mix is so high and the chamber is so small. The fuel injection system will be setup like this: MAPP or Oxy tank > Regulator > Ball Valve (to turn in on or off) > Accumulator > Check Valve > Chamber Inlet

This means that the pressure in the chamber will just equalize to the reg/accumulator pressure.

Unread postPosted: Sun Dec 13, 2009 8:17 pm
This means that the pressure in the chamber will just equalize to the reg/accumulator pressure
normally it would... but you can't just set the reg at your desired pressure (36psi).. with a setup like this the reg has to add more air to keep the same set pressure when the check valve opens (so when volume increases)normally the reg could do that if it had enough time or flow but that's not an option here...

but you can take the chamber + accumulator volume into consideration and set the reg so that the pressure would be exactly 36 psi once the two equalise (of course you can't forget that reg adds some air to the chamber while the two equalise)

Unread postPosted: Sun Dec 13, 2009 8:36 pm
Author: kjjohn
So basically, to be precise, I will need to take into account the flow rate of the reg, the volume of the accumulator, the volume of the chamber, and the flow rate of the check valve/chamber inlet.

Maybe if I model the accumulator as a metering pipe, I could figure it out easier...

edit: what if I made the accumulator have the same volume as the chamber? Then could I just set the regulator to twice the 36psi? (72psi) Of course, the reg would have to be able to bring the accumulator up to twice pressure again within a fraction of a second.

Unread postPosted: Sun Dec 13, 2009 9:25 pm
Author: Technician1002
The simple solution is to consider how the air is delivered to purge and replenish the cannon. It may be possible to simply take a propane regulator carburetor setup from a small gas engine and adapt it to the air delivery into the cannon. These should be able to handle the air delivery rate just fine.

Google small gas engine carburetion and you will learn of the primary regulator, the vaporizer, the low pressure regulator and venturi needed to do this.

It will fuel with the air replenishment cycle.

How do you plan on rapidly venting the chamber and drawing in fresh air?

Unread postPosted: Sun Dec 13, 2009 10:11 pm
Author: kjjohn
The chamber will not be vented (yes, I know that that is inefficient and cuts performance, but it doesn't really matter to me.) Fresh air will not be drawn in. As I mentioned in an earlier post, it will be fueled by MAPP and disposable pure oxy cylinders. This pure oxygen will have its own separate regulator (obviously) and injection system. That means that the carb will work for the MAPP and regular air, but I need another reg with a high enough flow rate that can handle pure oxygen (unless carbs are safe for oxygen use).

edit: Instead of a regulator, would it be possible to use a fixed orifice to reduce pressure without significantly reducing flow?

Unread postPosted: Mon Dec 14, 2009 12:22 am
Author: Technician1002
kjjohn wrote:
edit: Instead of a regulator, would it be possible to use a fixed orifice to reduce pressure without significantly reducing flow?

I think you are on the right track, but a regulator is still required. Liquid fuel such as propane maintains the vapor pressure over use and it is directly related to the liquid temperature. Oxygen on the other hand is not delivered at liquid vapor pressure (unless deliverd in a dewar as a cryogenic liquid). Oxygen pressure will drop in pressure directly with the amount remaining. This means the volume deliverd through an orifice will decrease with the pressure.

As for oxygen delivery rate, any regulator suitable for welding should deliver enough volume as they are designed to supply volume at pressure for cutting torches. Don't use the little brazing regulator that comes with the brazing set. It is not high volume. In the long run a small welding cylinder instead of the disposable ones are a much cheaper way to buy Oxygen.

I would recommend getting a pair of regulators and setting both for the same pressure, such as 15 PSI. Then a flow meter can be used on both to set delivery rates for both gasses. Both can be deliverd together through the same injection port to fill and purge the chamber in one cycle. Use a pair of welding back flash arrestors like is required on torches and a pair of ball valves or solonoid valves to shut off the gas during ignition.

A quick question.. What is the chamber made from and how big is it? The mix you are trying for is beginning to look like a bomb. It needs to be made to handle very high peak pressures and high temperatures. A short run will reach brazing temperatures. Anything that can blow out will. Your ignition will need to be able to withstand repeated detonations. This will destroy a spark plug. :shock:

Unread postPosted: Mon Dec 14, 2009 1:37 am
Author: D_Hall
For what it's worth, what you're asking for is very similar to the controls I built for an igniter system once upon a time (Google for "WSL gas igniter", should be the first hit). The way I did it was basically....

1) Set up oxygen and acetylene (yes, I was using acetylene...but that doesn't affect this conversation) tanks with regulators set to the same pressure.

2) Put in restrictor plates immediately before my combustion chamber. One for each gas. The orifaces were such that the relative areas corresponded to my desired fuel/O2 mixture.

Voila. Let 'em rip.

My mixture wasn't perfect, but more experimentation with restrictor plates would have made it so. I didn't really need perfect, I just needed a reliable deflagration and I had it so I ran with the system even if it wasn't quite optimized.

edit: Or yeah, what Tech said! :lol:

Unread postPosted: Mon Dec 14, 2009 6:01 am
could he just ditch the reg and use the dwell time of the valve to meter the right amount of MAPP ?

I guess he would have to add a pressure sensor on the MAPP tank as well as on the OXY tank to automatically adjust the dwell time as the pressure falls

this method would make sense if you really needed high flow

Unread postPosted: Mon Dec 14, 2009 6:54 pm
Author: kjjohn
The chamber will be constructed from sch 80 6061 t6 aluminum sleeved inside of sch 80 stainless 304. The whole setup will be encased in a low tensile strength aluminum structure. I modeled the whole setup with a pressure calculator, and figured out that this setup will withstand a working pressure of about 4000 psi. The ignition will cause a pressure spike of about 3000. I didn't take into account temperatures, however. Would some kind of cooling system help? I want to be able to hold on to this when it fires! (And I don't want to blow myself up) :)

The volume of the chamber is about 21 cubic inches (1ft of 1.5" dia). With the bolt fully compressed, it will be about 10-18ci.

Also, while on the topic of pressure, a few days ago I was wondering how gun barrels can withstand such extreme pressures produced by powder ignition. They are usually made out of regular old stainless (not sure which alloy, however), and do not have extremely thick walls, but can withstand up to 30000 psi!! How come nothing like this is available to spudgunners like us?

And in terms of the regulator, are you basically saying that a welding reg would have a flow rate capable of continually filling my chamber in a fraction of a second?