Hacking the Crosman Vigilante

[George Schmermund]

I've gone back to doing some bench shooting in the garage. I have three pistol frames that can be switched between different barrels for testing. I noticed that 2 of the frames came from the factory without any lube. The one with some lube had just a swipe of grease over a couple of parts and it was obvious that it had been applied after the gun had been assembled. A friend took his Vigilante apart and found that it had no lube at all, either. My choice for doing a lube job at all of the sites that needed it was to use a drop of type 68 way lube oil that I use on the lathe and mill. The "tackifiers" in the oil help to keep it from migrating away from where it's applied and also has other interesting properties.

One of the frames had been disassembled and was laying in pieces on the bench for a while now. This is the one that I mentioned had a gap between the cylinder plate and detent cylinder. BTW, they all do. While I was considering how to close the gap (~.004") I got distracted with some other measurements. Today I decided to just put the gun back together without any mods and see if there was much difference between a couple of different frames on the same 10" barrel. It was surprising to find that there was a significant fps reduction in the newly assembled gun compared to one that was used quite a bit. This required some head scratching.

After a couple of beers to clear my thoughts I tried looking through the side of the pistol and watching the clip as it was rotated by hand. There were points when the detent cylinder didn't sit squarely into the pellet clip detent. This caused a gap to form on the other side of the clip between it and the breech of the barrel. The frame was disconnected from the barrel and the detent cylinder was then accessible for poking at. Using a pencil, I could rotate and tilt it some. There were points were the cylinder would get stuck and resisted rotation. After applying a drop of Pellgunoil to the recalcitrant part things worked much better.

The frame was put back into service on the same barrel and the improvement in fps was remarkable. Sometimes it only takes a drop of oil in the right place. The improvement may also have been enhanced by the oil having sealed the .004" gap that was spoken of earlier.

[George Schmermund]

I've been thinking about the acoustical measurements that have been posted lately. It seems that I've been getting myself, and possibly others, confused by trying to use crossover terminology and definitions. If one evaluates the properties of a classical N wave produced by a shock cone from a supersonic projectile you'll see a fairly symmetrical trace that indeed looks like the letter N.

The traces from the microphone that are recorded in my experiments are pre-muzzle blast impulse records, but they aren't shock waves from a supersonic source. They are compressed air (or other gas) events that precede the pellet exiting the muzzle. They are shock waves none the less.

To avoid any conflict with firearm nomenclature I'll refer to them, for the time being, as just impulse (Imp) waves when they come from subsonic projectiles leaving an airgun barrel or the new impulse generator. As stated in an earlier post, shock waves can also come from many common subsonic sources.  I don't know if this helps or hinders the cause, so if any of you have a other ideas about this I'd like to hear them.

[George Schmermund]

Stan - Thanks for the missive.

[George Schmermund]

Now that I'm re-acquainting myself with the HP Digital Signal Analyzer (DSA) it's clear that a person can get more information out of these instruments than can be easily digested. The ability to do wideband testing with high resolution is starting to remind me about not being able to see the forest because all of the trees are in the way!

I've decided that the information I'm looking for can best be represented by using fractional octave analysis. This type of measurement represents acoustical energy in a way similar to how the ear and brain process sounds. It's an old method and not without is foibles, but for what I'm after it works well. The impulse nature of this noise is also problematic, but if we discard the notion of potential hearing damage, things are again simplified. I'm going to limit the bandwidth to between 20 Hz and 20 KHz. This is plenty for the testing that I want to do because I'm only interested, at the moment, in evaluating perceived loudness.

The impulse standard I've built has a definite tendency to disproportionately produce higher frequencies. I'll try to redistribute the spectrum by tuning the striker board. This in itself looks like an interesting project.

Where's all this headed? I have no idea!

[WhatUPSbox?]

So, letting Google take me past the weeds, into the rabbit hole and into the weeds in the rabbit hole, I find papers such as http://boson.physics.sc.edu/~kunchur//temporal.pdf which looked at temporal resolution in hearing. An interesting quote: "In one experiment by Leshowitz (1971), listeners were presented with a single pulse or two narrower pulses (with the same total energy) separated by an interval Δt. The click and click-pair could be distinguished down to Δt ≈ 10 μs" Other times were observed for other sound types, but it seems to bound the ear's response time. Gap detection in noise was much slower at 2 ms. Your nominal .22 ms result for the two observed shot peaks is in that range so in my mind, the question remains whether it is the amplitude of the two individual peaks or the combined energy that drives the perceived loudness.

Lots of hits under the topic of temporal summation of loudness

One more beer before I push search again

[George Schmermund]

I agree that there are many paths to explored when it comes to evaluating sound, noise, and impulse signals. The path I'll take at the moment is a pragmatic one. This divergence into acoustics is, at the moment, only to be able to give a reasonable answer to the question about the sound level difference between the factory 6" barrel and the 10" upgraded barrel. I'll make the same comparison using the new 18" barrel.

I can always revisit these measurements when they can provide information about other aspects of impulses. I'm still very interested in alternative methods of attenuating muzzle noise.

[Tomg]

George, this is massively interresting, you are doing and finding the same issues I have found on sound level testing. In case you are unaware, there are a milspec (actually a couple) standards for testing impulse noise from guns. It requires a 20microsec rise time capable device such as tge B&K or the Larson & Davis LTX. I got one of the latter and have their 90 to 190db micon it. After many hours of living inside the manual and its user interface I know how to use it. To my suprise ag impulse levels are much louder than I expected. The milspec standard I have employed is 160cm above ground, 100cm distance to the muzzle at 90 degrees and 10m away from reflective surfaces (walls etc). Since my mic dont allow for anything less than 90db I have placed it at 50cm distance from the muzzle which is half distance which increase the spl by 6db. So its a simple method to take readings and deduct 6db from the result.
My device can messure both impulse and averages at the same time. Infact it does 2 averages one being an average over a fairly fast cycle and one at the typical average. The other thing it let you do which is immensly helpful is that it can use different weighting for the different messures. So for impulse i can use linear (no weight) and a or c weighting for the others.
Anyway, not hijacking your thread, just filling in with more knowledge.

[Tomg]

Sorry...my phone double posted

[George Schmermund]

It's good to know that others are trying to make measurements on airgun loudness levels. Larson Davis makes some fine SLMs.

Also, you're not hijacking the thread, you're adding information.

I am aware of the various attempts to standardize the measurement methods by various agencies and organizations around the world. None of these efforts are directed towards airguns as far as I'm aware of. Some interesting links have been provided in previous posts in this thread. All relate to firearms.

A few things to keep in mind while trying to make sense of all this info include the chain of events that go along with the final numbers. It's important to know the rise and decay times of the microphone, preamplifier, filters, and measuring amplifier. The best and most reliable method to evaluate the total system performance is to capture the system's output signal on a storage type oscilloscope. Without well defined timing and amplitude information it's difficult to know where the numbers are really coming from and where any bottlenecks are. A type 1 SLM has builtin constraints in order to conform to agreed upon specifications and tolerances.

A simple example and case in point is the type 1 SLM's Impulse response time window. The standard requirement is that the window be 35 mS wide. Now, to a first approximation sound at 70º F will travel 1.13 ft/mS. This distance times 35 mS is about 40 ft. That means that any reflection less than 20 feet away can become part of the measurement. The amplitude will be diminished, of course, but consider the ground (grass, dirt, concrete, blacktop, etc.?) being only about 5 feet away in a typical measurement geometry. How far away is the shooter from the microphone? All of these reflections are integrated within the impulse window.

There are other settings that can be used on some SLMs to look at impulses, but without a fast stored image on a scope it's still guess work.

[Tomg]

35 micro seconds is a cycle lasting only 0.000035, so I doubt the reflections any closer than 1m would affect the impulse reading (peak). The Larson & Davis have a cycle time of 25 micro sec.
However if any averages are taken in the millisec range could be colored by reflections if they are sufficiently close.

[George Schmermund]

For clarity I'd like to bring the time units into perspective.

1 µS = 1 microsecond = 1/1,000,000 of a second = 0.000001 seconds.

1 mS = 1 millisecond = 1/1,000 of a second = 0.001 seconds.

Type 1 SLM specifications require an Impulse time window to be 35 mS which is equal to 35/1,000 of a second (0.035 seconds.)

I hope this information is useful.

[Tomg]

For clarity I'd like to bring the time units into perspective.

1 µS = 1 microsecond = 1/1,000,000 of a second = 0.000001 seconds.

1 mS = 1 millisecond = 1/1,000 of a second = 0.001 seconds.

Type 1 SLM specifications require an Impulse time window to be 35 mS which is equal to 35/1,000 of a second (0.035 seconds.)

I hope this information is useful.

Very good George, most over the counter db meters are in the mS range, which absolutely will be picking up some reflections. Most of them are not doing peaks either, only averages over said range.
When you step into the µS space, you start paying for your equipment by a huge margin. I am trying to take mine and doing some testing across a wide array of AG with the help of a fellow AG fan.
I've have so far tested break barrels, .22 and .177, gas/nitro piston break barrels .22 & .177. PCP's .22 & .25. The peaks using the milspec standard netted 148db peak impulse, and 120.2db avg over 0.025sec on a short barrel 38fpe gun with some hammer bounce.

Ps. The current milspec standard for gun/explosion impulse metering is 25 µS, A-weighting.

[George Schmermund]

The quest for better measurements using a type 1 SLM is still evolving. I found a copy of the MIL-STD here:  http://everyspec.com/MIL-STD/MIL-STD-1400-1499/MIL-STD-1474E_52224/

It appears to be the latest version. It looks as though the bar has been raised again. The 10 µs requirement is making things more interesting.

What I'd like to point out is that the meter specification calls for a 'peak' measurement which is different than the type 1 'impulse' measurement.

MIL-STD-1474E   15 April 2015   SUPERSEDING    MIL-STD-1474D 12 February 1997

4.7.4.3   Peak sound pressure level measurements.  The peak sound pressure level measurements shall be made using either a peak reading sound level meter or a combination of instruments with equivalent characteristics, such as a sound level meter and an impact meter, or a sound level meter and a calibrated oscilloscope or other appropriate calibrated display instrumentation.  The measurements shall be made with a sound level meter capable of making peak sound pressure level measurements with the total rise time of the instrumentation not exceeding 200 microseconds (µs).  The rise time of the peak detector shall be such that a single pulse of 200 µs duration produces a meter indication within 2 dB of the indication produced by a pulse having a duration of 10 µs and equal peak amplitude.  The amplitude of the 10 µs reference pulse shall be such as to produce a meter deflection 1 dB below full scale.


The requirement of  "The amplitude of the 10 µs reference pulse shall be such as to produce a meter deflection 1 dB below full scale" places a real demand on the crest factor for any SLM.

Fortunately, I'm only interested in making up my own rules for the relative loudness of airguns while using a type 0 measuring system.

[Tomg]

Yes, they are also looking at changing the location of the measurement to be close to the shooters ear. I think it was 12" from mic to earlobe.
But at any rate, it's good to know what the standards are so that when you want to compare apples to apples, you can.
I should clarify, it's highly useful when you are trying to make sense of how efficient a sound moderator/technology is vs. another...which I am.

[George Schmermund]

I don't think that a comparison of firearm measurements to airgun measurements could possibly be an "apples to apples" situation. They each release their muzzle energy in different ways. If you're going to make devices for just airguns it might be better to compare device 'A' to device 'B' and then compare both to no device at all. This can all be done in your own dedicated and well thought out testing environment. If everything is fixtured and standardized for one specific measurement your numbers will only relate to any difference between the devices.

At that point you'll know if your making progress or not. No measurements in other venues with different guns, devices, or measuring equipment operated by other people can add anything but confusion. Stick with what you can control.

This is only the beer talking. Choose your own path.

[Tomg]

George, 
You might have misunderstood my intention.
I am not suggesting comparing a firearm against a Airgun energy wise.
But the milspec standard is to meter the SPL (Sound Pressure Levels) And it's a log scale, so while we can debate the total energy output of them is not apples to apples, we can surely agree that let's say 130db at 1m at 90 degrees, is 130db at 1m @ 90 degrees regardless of the sound source as long as they are metered the same.

The reason why this is important is that when let's say you got brand X silencer for brand Y airgun and it reduce the DB level by 25db on airgun Y, you can now compare brand Z silencer and see how it compares to X.

[George Schmermund]

Now that I've chosen my path for making loudness measurements relating to the difference between the 6", 10", and 18" Vigilante barrel modifications, it's time to actually build the 18" model. I've been working on the design features that I wanted to Incorporated into this new carbine. I bought some parts from Crosman and started to do the machining to rough out the overall details.

Previously, I had been machining the rails for mounting a scope onto the barrel. This was time consuming and the part sill had to be painted or anodized. I finally came to my senses and looked for a weaver rail on the internet. UTG makes a nice one that fits perfectly onto the barrel housing rail and can be screwed in.

The photo shows the parts as they now are and the final machining and fit testing can begin. I'm sure that this effort will result in the best preforming gun yet.

[Rob M]

I'm sure the muzzle velocity will skyrocket.

[George Schmermund]

My curiosity got the best of me today so I decided to see how much each of the Vigilantes weighs. The scale was calibrated with a 5# Toledo weight. I was surprised at how light the 18" carbine build will be when it's assembled. A powerlet was in each pistol frame.

[George Schmermund]

I've been reading a lot about muzzle crowns lately. Almost everything that is out there refers to firearm muzzles. There seems to be considerable disagreement about whether or not the accuracy of any firearm is really improved using various crown and chamfer treatments.

There are some odds and ends written concerning the efficacy of these treatments as they apply to airguns, but they're few and far between. Some of the earlier tests that I've done on the Vigilante barrels leads me to believe that there can be a marked improvement in accuracy if the factory muzzle is re-machined and carefully crowned and chamfered. I attribute this fairly consistent improvement to the ballistic differences between pellets and bullets as they leave the muzzle.

Rather than start an opinion war here, I'll set up a few experiments that will demonstrate some interesting things that can be tested and measured using just the straw barrels from the Vigilantes I've been working on. I encourage those with contrary ideas about the experiments and their results to please step forward if I'm missing anything.

The first photo is of a Vigilante barrel as it comes from the factory. The second photo is of a barrel that has been given the lathe and Dremel treatment described in an earlier post.