GEEK Alert - Sonic Choking!

[Cal]

I agree, for maximum performance, we want to keep the restrictions in the system to a minimum.... The only useful purpose I can see for intentionally inducing choked flow is to limit the velocity to something less than what the gun is capable of..

Bob

Bob

I read that there may be some misinterpretation.  My comments are not in forming choked flow for any purpose,  but rather,  I'm going under the impression that at some point in the valve operation,  choked flow is inevitable.  And so,  best understood.

And on that theme,   A thought passed through (pretty much empty space ;-)

IF the pellet were retained from first movement, until the pressure drop across the valve opening was small enough to inhibit choked flow, then any "disturbance" attributed to the choked flow on opening event would be immaterial.  Thus eliminating 1/2 of the possible difficulty.   Playing with break away factor,  much like adjusting the neck size on a cartridge or throat dimensions on a barrel in order to manipulate chamber pressure.

Perhaps not easily accomplished in a controlled manner,  but it could just be a matter of size and fits.  Smooth twist style barrels may have advantages in testing.

[Cal]

There is a lot of good stuff to consider. It is thoroughly mixed up in the, 'just how big can I make the ports?' How big a port can be squeezed into a .375 OD transfer sleeve, or into a .375 barrel spot face...and then given the divide-by-cosine growth lengthwise as the valve port is angled, how big can that be done...LOL
sheers,
Douglas

Douglas

The Evanix AR6/Renegade design has no port to squeeze into any transfer sleeve....   A design trade off I suppose ;-)

[TimmyMac1]

I agree, for maximum performance, we want to keep the restrictions in the system to a minimum.... The only useful purpose I can see for intentionally inducing choked flow is to limit the velocity to something less than what the gun is capable of.... This can be used to great advantage detuning a bullet shooter for (much lighter) pellets, for example.... I use this in my Disco Double to detune for pellets and achieve stellar efficiency by doing that, while still being able to utilize a full pressure fill.... something you can't do by reducing hammer strike....

Bob

Bingo,
Reduced striker force translates to high variability so we don't like that direction as it makes it shoot like it has a HDD(not accurate). I'd rather stand on nthe spring and fly the hammer fast and back the power down at the Restrictor. A lot of people go for pure FPE output but the level of power that exhibits the best accuracy is normally also where it is efficient. So why choke it!. It is the best way to back the power down to where the system is getting efficient. Where we see the accuracy is optimal is so often the best efficiency as well it is the way we get the best of both worlds. The Really good 12 FPE guns get 150-160 shots per fill from a 13 CI. That is a bunch of FPE per didly. When we see a gun that makes big or little power and doesn't do it with reckless disregard to the practice of matching valve timing with barrel timing, we get what we want. GOLD!
Lots of folks focus on FPE while my practice of focusing on the Accuracy has led me to the practice of choking the flow.
As you turn a port restrictor in (that is the choke point) it will get a loud gun a lot more quiet (and accurate)before it ever drops velocity. If you see that as much as I have you get the picture.
A huge amount of what I do is pragmatic. Mistakes are valuable because as you explore you learn more. I made some big improvements to stuff I did not understand why till later. Why doesn't matter to me. Like 885. Some stuff works so often it is your security blanket. Getting the best use of your air has always worked (short of increased striker drag) ON TARGET! ALL MY GAMES HAVE POWER LIMITS. Shooting harder is not an option unless there is no power limit like the EBR. Even then, efficiency ruled. I think it always will with pellets.
What works for Drag stabilized pellets may not work for SLUGS. That is a whole different discussion(I'm clueless on that). Getting mileage is a lot like getting the best accuracy because you are sipping fluid, not gulpin'. Fluid dynamics is a hugely interesting subject I have involved myself with for my entire life. I've never had the lab to get it done proper so we focus on the ACCURACY. 950 would not be accurate.
Backed down guns WIN! Turning pellets sideways will never work on Target. WE are playing with lead shuttlecocks and she wants to turn when you blow up the girls skirt. Accuracy and BC suffer when the air timing doesn't match the Barrel timing. You audibly sense the extended duration that ruins BC and makes guns shoot 3 times the groups at the Paper.
Loud is Waste!

TimmyMac1

[rsterne]

Cal, I think that having choked flow on both ends of the shot (flow) cycle is beneficial.... I was aware of the sonic choking at the closing part of the cycle, which "squares off" the pressure pulse at the end as the poppet approaches the seat.... What I was not aware of until Jim brought it up was that there was a "pressure choking" (if I may use that term) which occurs at the beginning of the shot cycle as well, until the pressure difference across the seat drops below 1.893:1.... I had read that there was little flow that occurred until the valve was open a few thou, and perhaps that is the reason for that?.... In any case, the theoretical lift to dwell curve, which is a parabola, can not only get clipped across the top, from the "curtain limit" (when the lift exceeds 1/4 the throat diameter flow no longer increases), but also on both ends, on opening from pressure differential, and on closing from sonic choking.... This effect helps us approach the ideal "square wave" pressure pulse that we are striving for....

Having said that, any restriction in the flow path downstream of the valve (but ahead of the pellet) MAY cause sonic choking to occur prematurely.... If the goal is maximum power, IMO this is a bad thing.... However, at Tim@Mac1 points out, it can be used to great benefit to tune a PCP to a desired velocity while allowing full pressure fills.... The mechanism is that of sonic choking limiting the mass flow once Mach 1 is reached at the port restriction....

Bob

[rsterne]

I did up some diagrams that may help you visualize the flow through a "strike-open" valve in a PCP.... I started with the Lift to Dwell curve which from Physics should be a parabola (assuming a constant closing force).... Then I limited the flow to that allowed by a curtain height of 1/4 the throat diameter.... Next I added choking from the pressure differential as the valve opens and from sonic choking as the valve closes.... Lastly I showed what would happen if the port downstream of the valve was restricted so that sonic choking occurred to limit the mass flow....




I used a dwell of 2 milliseconds for the horizontal axis, which is typical.... The vertical axis numbers are only relative, with 100 being the maximum flow through the throat diameter at the pressure being used.... These are not necessarily to scale, but just presented as a visual aid to display what we have been discussing above....

Bob

[Cal]

The article you linked to talks about a pressure LOSS due to the constriction which accompanies the increase in velocity.... good ol' Bernoulli.... How can we use a decease in pressure to our advantage?.... The force on the base of the pellet is area (caliber) times pressure....

Bob

That raises a good question.

When does the dynamic pressure of density X velocity X are become static pressure that is based on reservoir pressure and is used in the F=ma quantification? 

The stagnation point being the base of the pell,  which is under constant deceleration with respect to the air flow. 

Hmmm.  makes a fellow wonder.

[rsterne]

I think Jim would be a good guy to ask that.... I assume you are asking if there is a difference (and how much) between the pressure seen on the walls of the barrel compared to on the base of the (moving) pellet?.... Since pressure is nothing more than the collision of air molecules with the container (barrel), and the random motion of said molecules (in every direction) at room temperature is ~ 1650 fps RMS (with some much faster and some slower, near zero), I would think that much below that velocity there would be no (or little) difference.... at least until you approach it very closely.... During the valve open phase, you are continually adding air from the reservoir, of course.... only after the valve shuts is the air expanding to any noticeable degree.... so my guess is those two conditions are quite different as well....

The airflow is accelerating along with the pellet all the time the valve is open, as I see it.... That is why the FPE lost in accelerating the air mass is such a large percentage of the losses we see.... and one reason why Helium has less such loss, and hence develops more power....

Bob

[TimmyMac1]

Air has mass so when it expands in a barrel it is directional and has momentum. Consequently it acts as an unrestrained spring loaded energy force than can generate much greater flow speed when suddenly constricted. The mass flow momentum is modulated/controlled @ restrictor/restriction to create the moment of valve closure via Sonic Choke. It can be done at the Bolt probe OR it will be choked at the redirection if that is the Max Restriction. There will always be some point where it is the bottleneck. If that point has adjustability it is best. That way you can set all your Striker settings to Optimize the best pressure map and power adjustments DON'T affect your map other than to cut the top of your curve off to benefit shot count. All good.  Air Flow is very dynamic and you must consider both the flow as well as the pressure as the closing forces you can manipulate to get the best valve flow profile.
Sonic choking takes the violence out of the valve closure and the resulting Pressure Pulse created with the air mass reversing direction. The Pressure/mass wave moves things and often it moves things too much. Most structure shakes unless it is adequate. The Ping says the air was what you could move. Structurally it means to Me I have achieved adequate or better. We build guns with massive Billets for structural integrity and then still try to minimize the Jolt. I think it is one of the reasons Wide Open never shows well in Competition even when there are no limits. Physical limits will always exist whether you can see them or not. Team Wide Open will always exist but it is a frustrating path. I like watching people venture there. It is where you get experience meaning you don't get what you want downrange.

TimmyMac1

[Jim_Hbar]

I assume you are asking if there is a difference (and how much) between the pressure seen on the walls of the barrel compared to on the base of the (moving) pellet?

In a non-dynamic, constant cross-section application, we all know that there has to be a pressure gradient down the barrel, otherwise the air would not flow.  But make it the dynamic situation that we are interested in, and it very quickly becomes exceedingly complex, and I suspect, virtually impossible to generalize across the wide range of applications we are visualizing.  Of course Bernoulli is ever present, but once it get's beyond the First Law of Thermodynamics, all the rest of the mumbo-jumbo comes into play.   And that is exactly where this discussion is heading! That's the deep end of the pool, and over my head.  I'll sit on the edge of the pool and dangle my feet in the water.

To model what happens at the level being discussed here, could be done by a CFD geek, I would expect - If the software was up to the challenge.  And the model would still need to be validated by comparing to actual data, and tweaking it to match actual results.

For practical purposes, the way I understand Loyd's model works, by qualifying the model to match field data through the use of an "efficiency" factor, is exactly the way I would do it.  And is effectively what happens in a model of any complex system, regardless of the modeling method.

A graphical isothermal  portrayal of the regions around choked flow would be fun if anyone comes across such tid-bits"

I've been looking,

Me too. Thus far coming up empty handed.

A sonic choke is anything but isothermal - adiabatic is closer, but I fear the 2nd and 3rd law apply. 
Unless this was a joke on the newby, then Ha Ha Ha..

[Cal]

A sonic choke is anything but isothermal - adiabatic is closer, but I fear the 2nd and 3rd law apply. 
Unless this was a joke on the newby, then Ha Ha Ha..
[/quote]

Jim

I agree,  real computation would be overly tedious.  Fudge factors answer for most purposes, based on observations.  I'm impressed that the presented algorithms perform so well!

As to "iso thermal",  it is perhaps a confusion of terms in common usage.  My intent is "lines of equal temperature" .  i.e.     iso- thermals.    A map.
http://dictionary.reference.com/browse/isothermal+line

adiabatic has nothing in common

ps

I would bet there are some interesting avenues of development along these lines.
http://en.wikipedia.org/wiki/Ludwieg_tube

I've always wondered what performance would be obtained if launching a projectile with a fixed volume of compressed air filling the entire system, and a rupture disk at the muzzle.   

Not practical,  but may give insight into harnessing  shock wave energy for certain purposes.

[Jim_Hbar]

Cal:

Isothermal and Adiabatic processes have very precise and specific meanings in thermodynamics/engineering.  Check Wiki for "Isothermal Process".  The first paragraph of that page does a fair job.

Isothermal and Adiabatic processes describe the two ends of a possible range of real processes.  Isothermal processes (constant temperature) are slow, and adiabatic processes (constant energy) very fast.  Reality is usually somewhere in between.

"Isotherm" is the word for used for the lines of constant temperature in this area of study.

[rsterne]

When I'm doing my calculations, I use Isothermal while the valve is open, and usually afterwards as well, even though it makes sense for the expansion phase (after the valve closes) to be Adiabatic, or nearly so.... Isothermal has an exponent of 1.0, Adiabatic of 1.4, and as Jim says, you can use something in between, such as 1.2 for after the valve closes.... However, using Lloyd's spreadsheet all that happens if you use Adiabatic is that you have to change the Efficiency "fudge factor" to a higher value.... I find his Model matches reality better using Isothermal, so I use that as my default.... If you used Adiabatic for the entire shot cycle the Efficiency factor gets too close to 100% to be realistic, IMO....

Bob

[Cal]

Cal:

Isothermal and Adiabatic processes have very precise and specific meanings in thermodynamics/engineering.  Check Wiki for "Isothermal Process".  The first paragraph of that page does a fair job.

Isothermal and Adiabatic processes describe the two ends of a possible range of real processes.  Isothermal processes (constant temperature) are slow, and adiabatic processes (constant energy) very fast.  Reality is usually somewhere in between.

"Isotherm" is the word for used for the lines of constant temperature in this area of study.

For your reference,  I attach the link again
http://dictionary.reference.com/browse/isothermal+line

And restate.  adiabatic nor isothermal processes have anything to do with my interest in an isothermal line graph of a gas flow in a choked state.

If you come across such an image,  please let me know of it.  I'm curious.

cheers

[K.O.]

hey Bob have you ever heard of modeling eddies with fractal dimensions...

I have heard of renormalizing and it seems it would be similar

I have not the concentration for math any more myself...

[rsterne]

Heard of it, know nothing about it....

Bob

[Cal]

Cal, I think that having choked flow on both ends of the shot (flow) cycle is beneficial.... I was aware of the sonic choking at the closing part of the cycle, which "squares off" the pressure pulse at the end as the poppet approaches the seat.... What I was not aware of until Jim brought it up was that there was a "pressure choking" (if I may use that term) which occurs at the beginning of the shot cycle as well, until the pressure difference across the seat drops below 1.893:1.... I had read that there was little flow that occurred until the valve was open a few thou, and perhaps that is the reason for that?.... In any case, the theoretical lift to dwell curve, which is a parabola, can not only get clipped across the top, from the "curtain limit" (when the lift exceeds 1/4 the throat diameter flow no longer increases), but also on both ends, on opening from pressure differential, and on closing from sonic choking.... This effect helps us approach the ideal "square wave" pressure pulse that we are striving for....

Having said that, any restriction in the flow path downstream of the valve (but ahead of the pellet) MAY cause sonic choking to occur prematurely.... If the goal is maximum power, IMO this is a bad thing.... However, at Tim@Mac1 points out, it can be used to great benefit to tune a PCP to a desired velocity while allowing full pressure fills.... The mechanism is that of sonic choking limiting the mass flow once Mach 1 is reached at the port restriction....

Bob

I was just reading through past threads/ posts on this forum.  ('sort of forgot how recently it started ;-)
In reading,  one post by you, Bob, mentioned a detail that is very likely true for most considerations.  That is,   during the period of time the valve is open,  the entire event is pretty much a constant pressure event. (Considering a usefully large reservoir and not a "big bore" with two shots per fill ;-)

With  valve closing while the projectile remains in the beginning half of the barrel volume,  this detail may very likely be the case.  After the valve closes, expansion rules.

For this discussion,  IF we assume an "almost constant pressure", though polytropic  event,  sonic choking on valve closing is highly unlikely when  applying the 1.9 : 1 pressure ratio rule of thumb.  Can we envision the propelling pressure drooping that low with the pell not yet half way out the barrel?  Perhaps....., in a spacial isolation, such as at the valve sealing face.  But the effect must be of very short duration.

  Perhaps this is included in the data that produced displayed plots.  I did not catch it if so.


With that,  If the choke effect might be minimized on valve opening by sufficient projectile break away requirements,  and choking may be suppressed on valve closing due to insufficient pressure differential,  this topic may reduce to only  a thought problem.

Perhaps it comes down to a question ,  "what pressure differential projections and data were used to describe choking conditions?".

  It would be beneficial to know the pressure profile in our air guns under any and all conditions ;-)

Regards

Cal

[rsterne]

As I said, there can be no choking due to exceeding Mach 1 on the opening part of the cycle, that would be due to the pressure difference, hence why I coined the term "pressure choking"..... On closing, the velocity of the airstream is at it's highest, and as the valve closes it speeds up locally through the (ever reducing) gap (curtain area) under the poppet, until reaching Mach 1 when the curtain area is small enough.... That is true "sonic choking", and limits the mass flow through the valve after it occurs.... and it WILL occur at some point as the poppet approaches the seat....

Steve in NC did an excellent analysis on Sonic Choking and the port sizes it occurs at, but he used muzzle velocity, not the velocity of the pellet (and airflow) at the instant before the valve closes.... Muzzle velocity has nothing to do with it, otherwise his chart was right on, just substitute the velocity at valve closure.... The velocity through the ports at that instant is the inverse ratio of the areas.... Go right back to the beginning of this thread for that argument, modified later in the thread by the increase in Mach 1 at high pressures.... However, even if the ports are big enough not to suffer sonic choking the throat (curtain area) of the valve MUST just before it shuts....

The best model I have seen of the pressure profile is the one produced by Lloyd's spreadsheet on Internal Ballistics....

Bob

[Cal]

Bob\

I'm in complete agreement with the flow velocity  on a closing valve. Velocity should be at it's highest unless the crest of the flow wave has passed, and molecules are stacking up.  What is less certain is the required 2:1 pressure ratio.  The sonic flow must pass into a region of low pressure. (1.89:1)  That is not logical to me.  With the pell still in the barrel and accelerating, there must still be high pressure all the way back to the valve.

Where do you feel the pressure difference would come from.  Is there a way to justify the projectile acceleration and velocity using only 1/2 of the reservoir pressure at witnessed volumes?

On valve opening,  we do have the required 2:1  pressure differential.   I feel Mach speed is pretty much instantaneous.  But can it be described as flow?    The gas molecules are already moving with that average velocity, they just need someplace to stretch their legs.  The opening valve is the gate.  (the disparity may be only terms ;-)
This may lead back to the dynamic vs static pressure determination.
 
Just to be sure we are discussing the same subject,  I C&P the description of choked flow from wiki.  I'm not clear on your coined "pressure choking".
 clip
Choked flow is a fluid dynamic condition associated with the Venturi effect. When a flowing fluid at a given pressure and temperature passes through a restriction (such as the throat of a convergent-divergent nozzle or a valve in a pipe) into a lower pressure environment the fluid velocity increases. At initially subsonic upstream conditions, the conservation of mass principle requires the fluid velocity to increase as it flows through the smaller cross-sectional area of the restriction. At the same time, the Venturi effect causes the static pressure, and therefore the density, to decrease downstream beyond the restriction. Choked flow is a limiting condition where the mass flow rate will not increase with a further decrease in the downstream pressure environment while upstream pressure is fixed. end clip

I'll summarize my views on the details here in a few lines.
1) The valve port of our airguns is surely the venturi.  There may be others throughout the port geometry.
2) The establishment of a "flowing fluid" might need clarification.  At least in the early portions of the valve event.
3) The "flow" in the reservoir upstream of the valve is subsonic prior to valve opening, (there is none),   with the gas in the region adjacent to the valve developing flow as the valve event matures.
4)IF the flow into and out of any region between the valve port and the pell   establishes the required 1.9:1 pressure differential,  we may see the static pressure/ density decrease that is the defining condition for choked flow.  i.e. constant mass flow across a restriction.

Help me understand the graph models you have posted.  The sections of the plots associated with the valve in the near closed position?  Are the curve profiles of "choked flow" based on open area X constant mass/unit area flow?  As opposed to open area X (what?)   

Your description
clip,   Next I added choking from the pressure differential as the valve opens and from sonic choking as the valve closes.... Lastly I showed what would happen if the port downstream of the valve was restricted so that sonic choking occurred to limit the mass flow....end clip.

If I have it correctly, the last  plots project  models of flow across the open valve WHEN the absolute mass flow of the system is being determined by flow conditions determined by elements down stream of the valve and prior to the pellet base. 


An added comment

I suppose one could use the MV for estimation if one were wedded to the sonic horizon theory.  For all the same reasons.  Namely,  events happening "back there" no longer matter.   That is a different topic.

eta

The plot on post #14 of this thread would suggest full pressure behind the pell at the get go,  with no significant let off until the valve closes .   
http://www.gatewaytoairguns.org/GTA/index.php?topic=75341.msg717558#msg717558

With that,  if the acceleration force follows Bernoulli's equation  Both Static and dynamic forces need be summed.  Ps+ (density X Velocity^2 X 1/2) to determine the total pressure.

There seem to be ample guess-timation in this investigation ;-)

[Cal]

I think the point is that we should use ports larger enough to avoid the problem, IMO....

Bob

Data point

The transfer port drillings for the Evanix AR6 in .22 measures  4.55-4.6mm diameter ,   the same for the .177 caliber rifle (likely only the barrel and mag differ)

So,  "full bore"  in the one example, and 80% in the other.  There are two 90 degree corners however ;-)

[rsterne]

The sonic flow must pass into a region of low pressure. (1.89:1)

Where are you getting that from?.... Jim said that choking will occur at that pressure ratio, regardless of velocity, I didn't see where he stated that the reverse is required.... Nowhere do I see a "requirement" for a 1.9:1 pressure differential for choking to occur.... In point of fact, if the venturi restriction is only 10% in area and the flow velocity pre-choke is Mach 0.95, the flow will choke at the venturi, but the pressure will only drop ~10% through the venturi....

Jim, is there a REQUIREMENT for a 1.9:1 pressure differential for choking to occur?.... if so, what am I missing here?....

Bob