Sonic Choke? Maybe not. 2162 FPS

[Scotchmo]

As I understand it, a CD nozzle is used to create a situation where the velocity of the gas reaches Mach 1 at the narrowest part of the restriction.... That allows the gas to continue to accelerate beyond Mach 1 downstream of the restriction.... By definition, the flow is choked, and you must get a pressure reduction of 1.89:1 across the system.... ie the outlet pressure cannot exceed 53% of the inlet pressure.... and in addition, there is a maximum mass flow that occurs once Mach 1 is reached in the nozzle.... The only way you can increase the amount of gas flowing through the system after choking occurs is to increase the upstream pressure.... Reading through this thread may help refresh your memory....  http://www.gatewaytoairguns.org/GTA/index.php?topic=66737.0

Since the goal is to maintain the greatest force against the bullet, and that requires pressure and air mass.... we should be avoiding restrictions in the system.... However, trying to understand how those restrictions affect the results is a sensible goal, because we use that as part of the tuning process sometimes....

Bob

I could imagine some benefits to the CD nozzle and maybe limit barrel length to about 50 x bore diameter.

If "sonic choking" is not an issue, than no CD nozzle and no limit on barrel length. But I think it may an issue if trying to reliably exceed 1700fps.

I would like to see a reliable configuration the yeilds >1700fps. Just to show that it is not an absolute limit.

My spreadsheet does not take into account "sonic choking". I'd rather that we not need to worry about that phenomenon, but we might have to once we are velocity constrained.

[rsterne]

50 x bore diameter is basically a pistol barrel for an airgun.... 12.5" in .25 cal.... or a naval gun eg. 16"/50 cal on a battleship (puts our long barrels I context, no?).... Artillery pieces are usually only 22-23 calibers long !!!

I played with Lloyd's spreadsheet trying to reduce the pressure by 47% to emulate choking, and had zero success.... Yet how can it NOT be happening when we are shooting well above Mach 1, even at high pressures.... The speed of sound in air at 4500 psi and 20*C is ~ 1550 fps, and we are beyond that.... At 3300 psi it is ~ 1400, and we are also beyond that....

Bob

[lloyd-ss]

   ( Bob, you know how to yank my chain, don't you, LOL.     )

IF sonic choke occurs in PCPs, and I say again, IF it occurs, it has to be a progressive phenomenon, that builds up very slowly after the air velocity exceeds Mach1.  So slowly in fact, that I have yet to see that anyone has shown hard evidence that it requires special consideration in our calculations.  It seems that the normal losses from increasing velocity (which we have not been able to wrap our heads around yet, anyway), will be enough to deal with the subtle effects of sonic choke at normal PCP pressures and velocities, even when we are pushing the limits of their performance.

I will again present this chart from earlier in the thread to back up my argument against sonic choking.

These shots are all .22 cal, 10.2 gn 14.5 cc dump shots.


In particular, look at the last 2 columns regarding the 8" barrel.  In all of these shots, the MV from the 8" barrel is below (or at)  Mach1.  But looking at those identical shots from a 23" barrel several of the shots are well above Mach 1.  But still, all of the shots from the 8" barrel are about 80% (within margin of error for the experiment) of the velocity from the three-times-longer 23" barrel.  My contention is, that if sonic choking WAS occurring,   that the highest velocities from the 23" barrel would be degraded enough that the velocities from the 8" barrel would be a noticeably higher percentage than the 80% measured during the actual experiment.  For example, the 4500 psi shot with 100% port area did 1605fps. That is certainly well above Mach1, but the 80% is still constant.  If sonic choking had started to occur, maybe the 23" velocity would have been 1470fps and the percentage would have been 88%.  But it wasn't. 
For the 4500 psi shot with the 24% port area, you might say that the 83% is an indication of something starting to degrade.  Maybe yes, but I seriously doubt it.  Again, I think that 3% is within the margin of error for the experiment.  If someone wants to present data that actually shows the effects of sonic choking, I would love to see it and be proven wrong.  I am always eager to learn new things, but I appreciate the evidence.
Lloyd

[rsterne]

As you know, I think I see an inflection in the curve below the 50% area TP, with the 25% port velocity depressed somewhat.... but I know you disagree.... I do agree that any solid evidence of sonic choking is lacking and/or hidden within the margin of error.... Further, I have tried introducing the 47% loss in pressure that must occur in sonic choking into your spreadsheet, and it makes no sense, basically that reduction in pressure would result in much lower velocities, as you say....

However, I am drawn back to the data presented previously by Jim_HBar about what a sonic choke is, and what it does to the pressure across the choke.... I don't think Jim figured that chokes were likely to occur, but at the time, we weren't working with velocities high enough to exceed Mach 1 at the pressures involved.... Now, we are.... Some of your shots are exceeding Mach 1 at 4500 psi, and a few at lower pressures as well.... That being the case, unless the pellet is moving faster than the air in the barrel, must we not have supersonic flow in the barrel, at least right behind the pellet?....

Waitttttttttt a minute.... Look at what I just said (in bold above).... Is that possible?.... Maybe that is the key to the whole "Sonic Joke".... Could it be that he air in the barrel is staying below Mach 1 at the pressure it is at (eg. ~1550 fps @ 4500 psi).... but enough of the molecules of air are moving fast enough (some are way beyond the 1650 RMS velocity), and colliding with the base of the bullet, to maintain the pressure, and hence the force needed to continue the acceleration to the 1605 fps (or beyond) muzzle velocity?....

Now THERE is a way out thought to ponder....  .... Is there a velocity gradient in the air in the barrel, with the air at the breech (or in the valve) essentially stationary, and ONLY the air touching the pellet going the same speed as the pellet (at any given instant)?....

Bob

[Matt15]

This post is very interest. However I do not understand much of it. Hope you guys can solve the mystery. 

[rsterne]

Matt, it's a work in progress.... unknown if it will be solved in my lifetime....

Bob

[Bill G]

My rational for the CD is not for achieving sonic speed of the air but to allow a volume to pass as quickly as possible.  The projectile allows a back pressure due to restriction.  In fact other than looking like a CD nozzle it probably isn't.  Essentially I am taking a vessel of compressed are that is 1"_ in diameter and narrowing it down to bore diameter.  So .22cal with 100% porting is obviously the highest flow we can get but the volume flow can be higher the shorter the pipe.  The "nozzle is to allow a restriction without a restriction if you will.  Since there is no pipe, the volume of flow is higher and that will allow me to use a smaller port, which in turn requires less force to pop it open.  My thinking is that the projectile is going to disallow the super sonic flow until it is likely 50% of the way down the bore.  (Barrel length can have a considerable effect on the flow since it is effectively a pipe)  By the time that the projectile has reached this point, the initial volume has expanded enough that the pressure at that point is inadequate to achieve the affect of choking and the projectile will have been able to maintain the highest pressure possible applied throughout the cycle.  The CD nozzle was originally intended for hot exhaust gas and the pressure would be continuous as long as fuel is being burned.  That isn't us.  We have a diminishing pressure which brings me to the theory of no choking is occurring unless we go really small with the nozzle or have a really large plenum.  Ideally, I would see this being applied to a valve that uses a plenum of 30% of bore volume(if viewed from the aspect of dump chamber.  But instead of being a dump chamber, there is a transfer between the chamber and reservoir that has a port that can dramatically exceed bore diameter.  The idea is to allow, as close as possible, zero force drop at the projectile for the first 3rd of the bore while having the volume flow at the highest speed possible.  The result would be 1/3 bore volume expanding from a decided psi with the projectile already moving at a substantial velocity.  If we are lucky, this could result in a very short lived super sonic flow applied, to an already fast projectile, just before the valve begins the close cycle.  This may already be occurring which could explain why we are getting velocities above sonic from a system that this isn't supposed to be possible.  For all I know an orifice is better but either way not having a pipe flow is even better. 

With regard to flow, 1800 psi, 23.3' barrel, 14.5cc plenum and 100% bore porting.  The possible flow under those conditions is .6099ci (9cc).  That would indicate the remainder of the 14.5cc is being unused.  Under the said conditions, my calculator returns a velocity of 1706.9fps.  But when I adjust the "used" volume to match what possible, I get 1553.27fps.  that's pretty close to Lloyds results.. Next, I will look at the other test pressures comparatively to see if this was a coincidence, I hope not.  I did some comparison with the port data and it seems that the flow becomes more efficient but tops out at .30. what I'm seeing based on the presented data is that, although the velocity is increasing with increased pressure, the efficiency of flow is decreasing as the port tightens. This makes me think that if we are using 100% bore porting, the efficiency of flow will increase as the caliber increases.  I did a regression for this and it indicates that 100% flow efficiency is reached once the caliber reaches .30.  This only raises the question of how does the plenum size has an affect on this flow efficiency. It also make me think that there is an optimal volume dump chamber for the caliber and barrel length, which makes since really.

Yea, this is bordering on a hijack. I'm just very interested in the information that you collect as it helps me and just thought that it may be useful to see my line of thought and how it flirts with the data. 

Bill                 

[Bill G]

As you know, I think I see an inflection in the curve below the 50% area TP, with the 25% port velocity depressed somewhat.... but I know you disagree.... I do agree that any solid evidence of sonic choking is lacking and/or hidden within the margin of error.... Further, I have tried introducing the 47% loss in pressure that must occur in sonic choking into your spreadsheet, and it makes no sense, basically that reduction in pressure would result in much lower velocities, as you say....

However, I am drawn back to the data presented previously by Jim_HBar about what a sonic choke is, and what it does to the pressure across the choke.... I don't think Jim figured that chokes were likely to occur, but at the time, we weren't working with velocities high enough to exceed Mach 1 at the pressures involved.... Now, we are.... Some of your shots are exceeding Mach 1 at 4500 psi, and a few at lower pressures as well.... That being the case, unless the pellet is moving faster than the air in the barrel, must we not have supersonic flow in the barrel, at least right behind the pellet?....

Waitttttttttt a minute.... Look at what I just said (in bold above).... Is that possible?.... Maybe that is the key to the whole "Sonic Joke".... Could it be that he air in the barrel is staying below Mach 1 at the pressure it is at (eg. ~1550 fps @ 4500 psi).... but enough of the molecules of air are moving fast enough (some are way beyond the 1650 RMS velocity), and colliding with the base of the bullet, to maintain the pressure, and hence the force needed to continue the acceleration to the 1605 fps (or beyond) muzzle velocity?....

Now THERE is a way out thought to ponder....  .... Is there a velocity gradient in the air in the barrel, with the air at the breech (or in the valve) essentially stationary, and ONLY the air touching the pellet going the same speed as the pellet (at any given instant)?....

Bob

Yes I think that is possible Bob.  that is what I was trying to covey in my previous post.  I think that if there is a sonic choke occurring, it is very short lived and that it is enough of a kinetic punch to get us above sonic.  And beyond that, there could be a fine line of sonic flow at the center.  This is evident in the exhaust gas of jet and rocket engines.  (shock cones). Which is what we may be able to achieve with CD valve.  Using that shock could result in the use of less volume which gets us more efficiency.

[rsterne]

Bill, if the idea of a velocity gradient is correct, then there can never be sonic choking in the transfer port unless it is TINY.... When the pellet is just starting out, there is no significant air velocity.... As the pellet moves down the barrel, the air behind it is constantly expanding, but is also being replace/augmented by additional air exiting the valve.... If there was truly a significant velocity gradient, the air at the transfer port, which is near the valve, would never reach Mach 1, unless the port was very small....

The idea that we could get more power by restricting the flow is counter-intuitive to me.... Rocket engines are PRODUCING large quantiies of hot gasses under enormous pressures, a far cry from the meager expansion of already compressed air in a PCP.... I'm sorry, but I see no similarities at all....

The possible flow under those conditions is .6099ci (9cc).

I have no idea how you arrived at that conclusion, so can't comment....


Bob

[match]

not sure if this will help but I found a very interesting paper on this

http://www.aft.com/documents/AFT-CE-Gasflow-Reprint.pdf

[Bill G]

Bob,I suppose I'm not articulating it as I intended.  The point is that even though the port is restricted it has nearly no length and that will allow a higher flow rate.  With a typical PCP set up, there is likely an 1" or better of piping from the poppet to the base of projectile. I was interrupted earlier and mis-stated the .6099(9cc) volume.  It was an incomplete statement.  9cc is what can flow through a 1" long porting under those conditions.  I suspect that it the longest porting/ barrel length combination that can achieve what Lloyd is achieving. With a dump configuration like Lloyd is using, I have to show a .334ci(5.47cc) volume to get 1254fps. The barrel, at 23.3" looks to be able to flow up to .6099ci (9cc) with his projectile and a 1" long porting.  Looking at my acceleration profile, it looks like the highest point of acceleration occurs at ~ .279" down the bore.  The rate of acceleration decreases from there.  What I'm wondering is, Is the projectile getting a boost of acceleration due to a shockwave forming for the briefest of moments at some distance down the bore?  Based on the force applied over the distance of the barrel the velocity should be much higher.  If a shockwave is forming and giving acceleration beyond what force is applying, if only for the briefest of moments, then  the projectile would essentially outrun the volume of expanding air, thus resulting in a diminished velocity.  The .6099 volume through a 1" long porting when compared to the .335ci volume that I purpose is fairly close to 50% rule (choked air at 50%pressure drop).  If this is true, then there may be a shockwave formation at ~5.7" down the bore. Perhaps pushing the projectile to sonic speed just to end up decelerating just until the force from pressure catches up and gives a final push resulting in a muzzle velocity of just over sonic in this case.  This could explain why smaller calibers are so much less efficient. Hard to tell when they are using different projectile weights but I think if we were to analyze it based on the projectile SD it would likely scale pretty well.  The smaller the porting / bore the more likely we are to experience this choking if it exists at all.  According the the regression models that I toyed with, the efficiency of flow increases as the caliber increases until .30.  The larger the caliber the less likely we are to have any choking.  Of course this hasn'r even covered the other pressures used. .    This could work in our favor with short barrels and smaller volumes.  In theory, we could achieve near sonic velocities if the barrel length and volume were matched to produce the shock wave say 1" from exiting the bore.

This is becoming compulsive at times.   

Bill G
               

     

[rsterne]

The barrel, at 23.3" looks to be able to flow up to .6099ci (9cc) with his projectile and a 1" long porting.

Sorry, I still don't understand this.... Over what period of time, is the first question (of many) that comes to mind (and that is assuming the "9cc" is at 1800 psi).... The barrel volume is the same as the valve (14.5 cc), so at the point the pellet leaves the muzzle, of the 14.5 cc of 1800 psi air that was in the valve, half (or more) is still in the valve, and half (or less) is now in the barrel, and the maximum pressure at the base of the pellet will be 900 psi.... It won't matter to the pellet if the valve closes at that instant, or continues to dump all the rest of the air.... so in reality the most air that can be "used" is 7.25 cc (of 1800 psi air), the rest is "wasted".... although without it being in the valve to start with, the pressure drop would be greater, so it still contributes to the FPE of the shot.... However, the fact remains that you could close the valve at that instant, trapping 14.5 cc at 900 psi, and double the efficiency....

I won't even pretend to understand shock waves, and the effect they may or may not have on a pellet preceeding them down the bore.... They definitely cannot exceed the local Mach speed at the pressure of their location, precisely why you see a supersonic jet pass overhead before the shock wave hits you.... My gut tells me that anything to do with restricting the flow, and/or choking it, and/or creating a shock wave, uses up energy that could be used in accelerating the pellet if that didn't occur.... It would take hard evidence to the contrary to convince me otherwise....

My guess is that you are trying to explain something way above my head, and it's simply not soaking in....

Bob

[Scotchmo]

.... but enough of the molecules of air are moving fast enough (some are way beyond the 1650 RMS velocity), and colliding with the base of the bullet, to maintain the pressure, and hence the force needed to continue the acceleration to the 1605 fps (or beyond) muzzle velocity?....

Now THERE is a way out thought to ponder....  .... Is there a velocity gradient in the air in the barrel, with the air at the breech (or in the valve) essentially stationary, and ONLY the air touching the pellet going the same speed as the pellet (at any given instant)?....

Bob

There may also be a radial component to the velocity gradient. The purpose of the CD nozzle would be to accentuate that. The center of the stream could be >>>mach 1, while closer to the barrel inner surface, velocity is much less. More of the "molecules of air" pushing the pellet to a higher velocity, without increasing the average velocity.



I'm just speculating, not stating any of this as fact.

[rsterne]

It makes perfect sense that there would be a radial distribution to the velocity of the air down the barrel, due to friction with the barrel itself.... at least in the subsonic region, where flow is incompressible.... That is why pipes have to be oversize.... I have no idea what happens to that profile, however, when you approach Mach 1, and the flow is now compressible.... Does it get more "pointed" in the middle, or the opposite, flattening out until you get a shock wave forming at Mach 1?.... I have no idea....

Bob

[Bill G]

.... but enough of the molecules of air are moving fast enough (some are way beyond the 1650 RMS velocity), and colliding with the base of the bullet, to maintain the pressure, and hence the force needed to continue the acceleration to the 1605 fps (or beyond) muzzle velocity?....

Now THERE is a way out thought to ponder....  .... Is there a velocity gradient in the air in the barrel, with the air at the breech (or in the valve) essentially stationary, and ONLY the air touching the pellet going the same speed as the pellet (at any given instant)?....

Bob

There may also be a radial component to the velocity gradient. The purpose of the CD nozzle would be to accentuate that. The center of the stream could be >>>mach 1, while closer to the barrel inner surface, velocity is much less. More of the "molecules of air" pushing the pellet to a higher velocity, without increasing the average velocity.



I'm just speculating, not stating any of this as fact.

This is How I imagine it in my mind.  The question to me is does the center region cause an unexpected acceleration?

[rsterne]

Thinking about the idea of a velocity profile once more.... That diagram is for the free flow of a fluid in a pipe.... without a bullet plugging it.... Once you put a bullet in the airflow, and the molecules of air are bouncing of it (which is what creates the pressure -ie force- to accelerate the bullet).... surely the velocity profile, at best, would be something in between the first two diagrams.... ie no profile, to one with just a slight lag in velocity along the edges.... no?....

Bob

[lloyd-ss]

Sorry guys, I totally missed this last series of replies.  A lot of food for thought, obsessive and otherwise.  The diagram that Scott introduced brings up a number of thoughts.  Even the wording at the left hand entrance end is provocative.  Hydrodynamic..... air is compressible and velocity AND pressure gradients are always changing, both radially and axially in the barrel.  And Bob, yes, I tend to agree that the air velocity in most of the barrel might be lower than the velocity of the projectile.  After all, it is the pressure/force on the bullet that accelerates it, not the velocity of the air that is behind it.
But, back to Scott's diagram, keeping in mind that that diagram is for a system with an open exit, I think.  A pellet barrel is a closed system, with a pellet acting like a cork in a pipe.  The first profile at the left: the flat fronted one, looks like what I'd expect from a big fat 500gn .458 lumbering through the barrel at 600fps.  That 500gn bullet really is a cork in the barrel.  Lighter projectiles must certainly behave differently.

But thinking about the velocity and pressure profiles of a compressible fluid within the barrel, it seems like many different things could possibly happen.  Bill G, you may have already said this, but I am struggling with the concepts and might totally miss the point on some of the things that are said.   So, with that slow moving bullet, the pressure on the back end of the bullet might remain almost constant all the way down the barrel.  And the velocity might also.  But with light weight projectiles things change.
The profiles in those diagrams are for velocity, as it increases in the pipe.  But we are trying to push a cork out of the barrel.  Isn't pressure what is important, not velocity? I realize that they are related, but air can stack up behind a projectile and the pressure can increase (or at least not decrease).  And what about the density of the air? That ties the relationship between velocity and pressure/force together, doesn't it?  With that same heavy bullet, you've got low velocity, but high pressure and high density.  With a light,fast pellet, at the end of the barrel you might have high velocity but low pressure (after all, air will only move from a high pressure area to a lower pressure area).  I know Bill G has mentioned this continuum of changes of velocity and pressure in the barrel before.  So maybe we can maintain high pressure against the back of the bullet, with actually having all of the air at the same high velocity.  This is difficult.  I wish I had actually "learned " this stuff, instead of having just "passed" it. 
Lloyd

[rsterne]

I see our friend Steve in NC is still hyping his "Sonic Horizon Theory".... he put up a link to it tonight....  http://www.network54.com/Forum/79537/message/1188078593

Here is the chart he uses....



along with this statement....

Any propellant that leaves the valve after the pellet crosses the Sonic Horizon can make no contribution whatsoever to muzzle energy - and is therefore totally wasted.

So according to that, with Lloyd's shot at 1600 fps, any air that exits the valve after the pellet has moved 5% of the way down the barrel (about 1") cannot affect the muzzle velocity.... Does anybody have any idea how we can work that into our spreadsheets?.... No, seriously?....

Bob

[Buldawg76]

Bob
I am by no means at or near any level of knowledge that you and Lloyd are at in the theory of HPA and its effects in a barrel or gun, But Steve's post on the 54 and graph of the full dump system just makes no sense to me in that the valve is charged and fires, then the poppet closes and stops the pressure discharge into the barrel.

Then he shows the high pressure air following the pellet down the barrel with a invisible wall behind it at the point the poppet closed shutting off discharge of HPA behind the pellet so would the pressurized expanding air not only follow behind the pellet pushing it down the barrel, but also start to expand rearwards back towards the valve poppet seat so that in effect the energy pushing on the pellet would be diminished more so than if the full charge of air in the valve had been released to fully expand behind the pellet and leave the valve empty ready for the next charging cycle.

I just cannot see where that invisible wall that traps the pressure expanding behind the pellet comes from much less how it can even be considered to exist at all.

But I am only a retired auto/cycle technician that can fix anything that rolls down a road and no college graduate but what he says and shows makes no sense to me at all.

Mike   

[rsterne]

Bingo!.... In the case of Lloyd's 1600 fps shot in a (let's call it a 24") barrel, according to Steve you could close the valve when the bullet has travelled 5% (ie 1.2 inch) and nothing after that matters.... As I see it, that 1.2" long volume of 4500 psi air now has to expand to 20 times that, with the pressure dropping to just (4500/20) = 225 psi by the time the bullet reaches the muzzle.... The bullet, therefore, sees a continually decreasing force as it travels down the barrel.... If we use .22 cal, at 4500 psi the force starts at 170 lbs.... The way I see it, the force decreases like this, if you close the valve at the 5% point of bullet travel down the barrel....

Bullet travel and force:
1.2" = 170 lbs.
2.4" = 85 lbs.
6" = 34 lbs.
12" = 17 lbs.
24" = 8.5 lbs.

If we leave the valve open, air continues to flow from the reservoir (let's assume for a moment that is infinite volume), and keeps the pressure at 4500 psi, and the force at (nearly) 170 lbs., for the entire 24" of travel.... Tell me again how those could possibly have the same result?.... It may be possible that the air exiting the valve can't reach the bullet after the 5% point.... but each increment of air being released can effect the increment of air released previously, pushing it along the barrel and making it exert the same pressure on the one in front, and ultimately on the bullet.... You can think of it as 20 "chunks" of air, each one pushing the one in front of it if you wish.... but the net result is that the bullet sees a (fairly) constant 170 lbs. of force all the way to the muzzle....

Incidently, if you run the numbers through Lloyd's spreadsheet for a 10.2 gr. bullet in a 23.3" barrel, using 14.5 cc of 4500 psi air, and close the valve at 5% of the barrel length, even with setting the pellet drag and breakaway at zero, not including the air mass, and the efficiency at 100%, you only get about 1300 fps as a result.... Add 1 lb. of pellet drag, 2 lbs. of breakaway, and include the mass of air and that drops to about 1150 fps.... Drop the efficiency down to 70%, which is more realistic, and you get under 1000 fps.... However, increase the dwell to a dump shot, where the valve is still open when the pellet leaves the barrel, and the predicted velocity jumps to 1609 fps.... Lloyd got 1605 in his test.... Note that with a 14.5 cc reservoir, the muzzle pressure ends up 2250 psi, and the force drops to 85 lbs. (or less), because the barrel volume is the same size.... A larger reservoir would produce a higher velocity....

Bob