Hacking the Crosman Vigilante

[George Schmermund]

High speed photography is moving up the list of experiments. Being a free spirit has its drawbacks. My attention span controls what I end up doing each day. Having several things going at the same time is OK with me, but it drives my wife nuts. And then there's the beer.....

As for the repeatability of the displayed impulse waveforms, they show everything exactly as the microphone sees (hears?) them. If the setup isn't changed the only difference in the display is a difference in what leaves the muzzle.

It might be helpful to examine the top waveform in the last post. What we see is the N wave of the air that was compress between the muzzle and and the breech as the pellet enters the forcing cone. The arrow at the bottom of the grid indicates exactly where the trigger starts to record the event. Each vertical line of the grid is .1 mS (100 microseconds) apart. The space to the left of the arrow shows what happened .1 mS before the waveform trigger said 'go'. There is essentially no ringing behind the N wave.

~ .22 mS behind the N wave the pellet leaves the muzzle and the CO2 blast emerges along with a series of well defined pulses. I'm going to assume that these pulses are the same ones that showed up when the pressure/time waveforms were recorded. I think I'll change barrels and remeasure the p/t curves with the microphone and pressure transducer running simultaneously.

The second (middle) trace is interesting because it indicates that some air reentered after the duster gas was blown into the barrel and before the pistol was fired. The small first pulse is the compressed air followed by the larger peak caused by the compressed duster gas followed by a series of smaller pulses. The duster gas is heavy, which makes it slow to compress and that the speed of sound in the gas will be slower. Are these smaller pulses shock waves or standing waves? What's also of interest to me is that the air and duster gas didn't mix. they both have distinct and well defined peaks. The barrel appears to be acting like large diameter capillary column as used in gas chromatography separations. This needs further investigation.

Since all three traces are triggered by the air N wave, the time delay and shape of the pellet's muzzle blast have a lot of information content in just 0.9 milliseconds of recording.

What's happening in the third (bottom) trace should be clearer now, so I'll leave it as a pop quiz!

[George Schmermund]

As an aid to understanding what's going on in the previous scope traces I thought it might be helpful to convert the screen grid into distance instead of time. I'm confident that most of you following this thread can switch between the time and distance relationship in your heads, but there may be some that don't have that faculty. Therefore, I'll post the scope display photo again (now that we're on a new page) and state the time vs distance conversion.

The speed of sound in air at 74º F (when the test was done) is 1132.17 f/s. That's the same as 13,586 in/s. Now, if we convert the display time interval for each vertical line increment into distance traveled by just the air N wave in trace 1 we will get 1.356" or ~ 1-5/16". That makes the distance between each dot 1/5th of the distance between vertical lines equal to .272" or a little over 1/4". That put's the pellet's instantaneous position as it leaves the muzzle at ~ 3" behind the N wave .

I know all of this has a big "so what?" factor attached to it, but this really helps me to do things like set up the high speed shadowgraph experiments.

Again, I'd like to thank all of you who are following my musings and allowing this thread to continue into the unknown.

[dv8eod]

You don't have to thank us, just tell us who is going to publish all your findings into an encyclopedia for airguns. 

[WhatUPSbox?]

George,
I was thinking the position lag of the pellet would be based on pellet velocity (~500 ft/s?) not the speed of sound. So for the .22ms this would be around 1.3". That leads me down the rabbit hole of what pellet velocity does it take to form the N wave?....did I go down the wrong rabbit hole?

Thanks for data and insight

[George Schmermund]

Stan - I think that the N wave is probably new to everyone as far as airgun measurements go. I'm not sure (yet) what the pellet velocity in the barrel needs to be to produce a sharp N wave. My excursion into time vs distance conversions is only to determine where to place the microphone to capture the best muzzle images for the shadowgraphs. I didn't even know that the N wave existed at these low velocities. It's a boon to my thinking and very useful for things down the road. These experiments will continue and I'll report my findings for others to consider.

[George Schmermund]

I've been doing some research on N waves. As was previously asked, it would be interesting to know how fast the pellet would have to be going to produce them. The closest that I've come to what's going on in an airgun barrel is in references about "open-end shock tubes". These most closely resemble a gun barrel. Many of these devices are driven by compressed air. Apparently these waves are common in many everyday events like hand claps and popping balloons. The wave doesn't have to be supersonic, just loud enough to form a strong pressure wave. This requirement would allow the pressure wave to remain at Mach 1 at the muzzle and beyond.

I've looked at some high speed shadowgraphs of everyday events and the claim is that the sound pressure level for good images needs to be on the order of 110-120 dB to get enough contrast. This would explain why the scope trace N waves we're seeing near the airgun muzzle are so clear and prominent. The microphone in these experiments is recording ~150 dB at its position. I don't know yet how slowly the pellet would have to be going down the barrel to not produce an N wave.

This gets us back to the pellets velocity relative to the N wave being generated. This wave will always be propagating at Mach 1 in CO2 powered guns. That info is a very useful tool to have.

If I've missed something here please let me know.

[WhatUPSbox?]

I found an interesting comparison of .22 starter pistol shooting blanks and bullet rounds. In that short barrel pistol it should be subsonic, though faster than the pellet.
http://c.ymcdn.com/sites/www.hearingconservation.org/resource/resmgr/imported/Sondergaard_Jacob_022511.pdf

Here they added a .32 pistol

http://c.ymcdn.com/sites/www.hearingconservation.org/resource/resmgr/imported/Soendergaard_Jacob_022512.pdf

The same group described their instrumentation and methodology here for a high velocity rifle. I would guess they used something similar for the pistol tests.
http://www.sandv.com/downloads/0908rasm.pdf

Kind of fun trying to diagnose the results

[George Schmermund]

Thank you for posting these resources. I've read them earlier and it's where I became aware of N waves. There is also a good deal of info from US military testing and NATO. The problem I'm having now is to try and relate these published findings with what the airgun signals look like. With airguns, in general, there is no extreme barrel pressure and high temperature to contend with. This make the airgun measurements easier and less complicated, but there's still a lot of speculation, interpretation, and understanding that's left to do. That's where the beer comes in!

[George Schmermund]

To add to the confusion I found this information today.

Acoustical Characterization of Gunshots   Robert C. Maher Department of Electrical and Computer Engineering Montana State University Bozeman, MT

"A subsonic rifle or handgun without a suppressor will produce a muzzle blast acoustic signal, but the subsonic projectile will not create a shock wave or any other appreciable acoustic signal as it propagates through the air."

What's wrong with this picture?

After doing a lot more reading and attempting to get a better understanding of acoustic measurements relating to airguns a clearer picture is beginning to form. I'll have to admit that I misinterpreted the above statement about shock waves. After parsing the sentence I realized that I didn't pay attention to the part about "as it propagates through the air." I was still thinking about the impulse wave generated by a subsonic pellet "propagating through the barrel". So, it is true that a subsonic projectile won't produce an "appreciable acoustic signal" on its own.

[WhatUPSbox?]

In the picture of your test (6/29), it looks like the pistol and the microphone share the same support stand. Is there any chance that there is a coupling through the supports?

[George Schmermund]

Yes, there is definitely coupling between the two, but that would show up as a ringing like a damped tuning fork.  Also the time window for the measurement is too short for the ringing to effect the measurement.

I did the test anyway just to make sure that nothing was overlook in that area. The test was done by using the same settings that the original measurements were made with and just tapped the stand and gun with a small hammer. Even with some aggressive blows (far more coupling than the muzzle blast could impart) I couldn't get the system to even trigger the scope.

[WhatUPSbox?]

Thanks for checking that. I thought maybe the initial hammer strike was registering. But you are right there are enough joints in that support path to quickly attenuate that.
For a shot without a pellet would there be just the muzzle blast signature since there is nothing to generate the N-wave?

I did find a few more papers  by Maher from around the same time frame...Looked like they used the same data.

Another paper showed a .45 cal handgun (presumably subsonic) response that had a leading N-wave

[George Schmermund]

Here's a quick test that was done this morning. The photo shows the top trace of a shot with a pellet in the barrel. The bottom trace is without a pellet in the barrel. As can be seen there's a marked difference in the amplitude of the two traces. The top trace is set to 1 volt/division. The bottom trace is at .5 volts/division. The muzzle N wave with no pellet is still distinct, but the CO2 has changed shape and is moving faster because it isn't pushing a pellet. There is still a time lag with the CO2 which indicates that the air in the barrel is acting like a plug of gas. This sort of effect is very apparent in the duster gas experiment shown in an earlier post.

[George Schmermund]

Now that we know that the N wave exists in even low velocity airguns and is easily recorded, it would be nice to know what the frequency content of a muzzle blast looks like. I've searched for any information about the airgun spectrum, but have found nothing of interest yet. There is some info about small firearm spectral plots, but they aren't in good agreement with each other, so it looks like another can of worms to sort through.

Rather than rely on what has been published for firearms it would make sense at this point to just do the measurements and get the truly relevant numbers for our airguns. The LeCroy scope has an FFT mode that can easily do the job, but I've also got an HP 35665A Dynamic Signal Analyzer sitting in the corner that's not earning it's keep. It's a fine instrument that's enjoyable to work with and may be up to the task.

I'm starting to think that when the dust settles there will be more than just measurements of the differences between Vigilante barrel upgrade noise (the original reason for doing these tests). It may be possible to design a barrel housing with a new type of Impulse damping. 

[George Schmermund]

Now that the experiments are going in the direction of making frequency spectrum N wave and muzzle blast measurements it's going to require a standard mechanical impulse generator. Rather than using the gun itself for these test I've decided to build a simple device to at least make N waves. If this works I'll be able to do the first tests many times without having to use up powerlets and pellets. The device is also more compact than the pistol arrangement.

As the photo shows, it was a simple task to re-purpose a mousetrap into a new laboratory instrument (the proverbial better mousetrap). The snap is loud and will probably make useful N waves. We'll see.

[George Schmermund]

This mousetrap snapper is looking very interesting as a noise source for doing impulse testing. At 1 foot from the microphone the sound level is reproducible at 137dB +/- ~ 1 dB.

As can be seen in the photos the time to peak is 9.5 µS and the ramp back down to the baseline is 16 µS. That's not bad for such a simple contraption. Also, one of the hallmarks of an N wave is a fast rise and a slower ramp down. Now, I'm not going to say that this first pulse is 'for sure' an N wave, so that leaves room for the rest of you to reassign it as something else. In the mean time I'll just claim the device to be a good reference for something loud and fast.

The third photo shows the full impulse decay envelope when the scope is slowed down. It looks pretty classical to me.

[WhatUPSbox?]

Love the impulse standard.

When you mentioned getting the frequency content of the response, how would the result be evaluated?
I don't have an acoustics background so some web search helped.
For continuous tones, the the equal-loudness contours are used https://en.wikipedia.org/wiki/Equal-loudness_contour but these don't apply to an impulse.

For short duration (<200ms), there is a significant reduction based on the response time of the ear. http://cjur.ca/wp-content/uploads/2017/03/Identifying-Perceived-Loudness-in-Audio-Signals.pdf

I suspect the answer is also different if one is interested in detection or in damage.

[George Schmermund]

As a reminder to myself I'll comment that the purpose of these impulse measurements is to test the different loudness levels of the various Vigilante barrel modifications. Under normal circumstances hearing loss wouldn't be a consideration unless you were to get shot through the ear. I'm talking about typical CO2 guns, of course.

Most often shooters are actually concerned about airgun noise leaking over into the neighbor's property. A quieter gun would also be desirable for target practice indoors. Rather than sticking with just comparative levels of attenuation with barrel length, it now seems to me that all of these impulse experiments might steer things toward quieting techniques that can be incorporated into whole barrel assemblies.

My propensity to stray far from the original course is what's driving this latest boondoggle. I think that some good information can be squeezed out of all this testing and that some other folks might benefit from it, too. I have to confess that I'm learning a lot more than I had anticipated.

Evaluation of the results is up to what one chooses to take away from all this. I'm writing my own standards for in-house use as comparisons on my own projects.

It's all here for the taking.

[WhatUPSbox?]

Yep, at my age, additional hearing damage from airguns may be a low risk...and I hope shooting my self in the ear is as well (though the foot is often at risk)

For kids though, the exposure thresholds are lower (120db per WHO) so some consideration may be appropriate.

Data for different airguns here https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4896309/ I was surprised that there wasn't a big difference between the lead and alloy pellets. I would expect the alloy to be supersonic. Their instrumentation (1/2" mic) may not be good enough to catch the peak.

Also shows the comparison of muzzle plane and ear location mic response. would have been nice if they had a time domain response chart.

[George Schmermund]

Thanks for the new reference article. I'm adding it to my expanding list of information that is now flowing into a growing pool of uncertainty. It's all grist for the mill.