Hacking the CP-2

[WhatUPSbox?]

George,
For the scope trace in post 429, Do you have the lbs/V for channel 2, the stimulus trace?

[George Schmermund]

The posted graph image isn't meant to be analytical by any means. It's just to illustrate a concept in physics. The change of momentum produced by an impulse is numerically equal to the impulse. Or, the other way around, an impulse is equal to the change of momentum: Ft = m(V terminal - V original). V terminal will always be equal to zero when it's captured by the KZQ. V original will always be the velocity at the moment of impact. We end up with negative acceleration for V. It's F = ma from there out.

As far as plot 1 and plot 2 in the image go, we could switch the the hammer experiment for a KZ of either a vital organ shot or for a head shot. One is hide + flesh and the other is hide + bone (mostly). These forms of target can be replicated, to some level, by changing the KZQ's 'cookie'.

"Myth Busters" gets a "No comment". TV is what it is.

I have no 'real' numbers for the DSO traces. The charge amps were set to produce a 1 for the 8200 and the KZQ balanced at .685. That would be the gain ratio for these sensors in early tests. I'm only interested in relative measurements at this point so that the impact surface's sensitivity can be plotted for further tuning.

[George Schmermund]

As previously mentioned, an aspect of the KZQ front cookie is that it can be changed to represent differences in potential targets. Pellets of the same caliber and weight can have different geometries and materials at their tips. Comparative impulse profiles may be instructive for different applications. An impulse is generally considered to be a specific quantity measured as a time variant of momentum. It might also be viewable as an informative shape.

Much more testing will have to be done to make this device a reliable source for data, but some interesting things are beginning to show up in the DSO traces.   

[George Schmermund]

The KZQ project has reached a point in the design and measurements stage that allows a certain level of predictability as to what the data window will look like. What happens within that window is what I'm sorting out now.

The measurements are such that the amplitudes and half sine frequencies are pretty well bunched into a very easily evaluated group of shapes that represent a pellet's impulse profile. This will allow the relevant data to be captured and stored on a much simpler acquisition system than my earlier ideas lead me to believe.

To that end I've chosen to continue the prototyping with just a very simple and inexpensive 1 channel DSO that is about the size an early smartphone. It can be connected to the output of the force transducer's charge amplifier and will be capable of storing many screen shots along with their data. This approach will eliminate the need for developing specific software and WIFI connections on an entry level system.

There was a time not too long ago when the idea of an average airgunner having a measurement system at the muzzle of their gun that would measure things like velocity, kinetic energy, and also be capable of doing a statistical analysis of exiting pellet groups would have been laughable. Now it's commonplace. Being able to measure and evaluate a projectile's impulse force at the POI is not that much of a stretch. It's just something new and different for the same airgunners.

The mini DSO that I'm starting with is really no more complicated to understand and operate than a stereo receiver's control panel or a modern car's dashboard controls. I've included a link to a YouTube video that gives a good introduction to this DSO. It's done by a guy that's sort of a 'Mr. Rogers' with and accent. 

I'd like to add that this instrument is not far removed from the mini DSO that Stan referred to back when we were both considering getting new o'scopes. Now I can see where these might fit into doing some level of ballistic testing!

[George Schmermund]

I've been giving some thought as to the best way to represent the impulse signals from the KZQ. Since the signal is, for the most part, a half sine wave and the impulse is numerically equal to the integral of the momentum there may be a better way to separate the two. My approach so far has been to inspect the maximum force signal as a time variant of the total momentum. I now think that there's an easier way to look at the data other than just the impulse's integral.

An alternative approach would be to consider the FWHM (full width half max) value. This method would only require measuring the impulse's maximum signal voltage and than measuring the time interval at 1/2 of that maximum value. There are many types of measurements that use the FWHM metric for device or circuit appraisal. Usually this metric is expressed as the "Q" value of what is being measured. This is a handy signifier since I've already got a "Q" in the new instrument's brand that describes the instrument's purpose as a Quantifier.

Any thoughts out there?

[WhatUPSbox?]

George, I'm curious how repeatable the traces from actual pellet impacts are. The putty is a non-linear material and may have different thickness. There may also be small voids. While the area under the curve is dependent on the pellet, the shape of the curve is influenced by the putty and could vary. Do you have an example of a pellet strike?

[George Schmermund]

The ballistics test bench has been rearranged to make it easier to position the KZQ for testing and making measurements. I did some test shots with the muzzle to POI set at 14". The Combro chrony is installed on the carbine version of the CP-2. Having muzzle numbers for comparison with the KZQ data at different distances can finally be done.

These DSO images are just a couple of screen shots to make sure that everything is working. There are some numbers here that can be seen as samples of where the current testing is headed using the FWHM approach. I think this will be a good method for POI measurements if they compare favorably with the muzzle data. The 2 shots were done using CPHP 22 cal pellets aimed at different areas of the cookie. 1 image is of a shot aimed at the center of the cookie where the calibrated KZ is 3/4" in diameter. The second shot is closer to the flexure's edge on the face of the load transducer. The sensitivity drops near the edge because the cookie tile's base attaches to the very stiff body of the sensor.

The point here was just to capture the pellets without over penetration or rebound. The goal was to see if there were any noticeable distortions in the signal's shape or other perturbations that might give some merit to Stan's mention of putty nonlinearity. These were simple tests to confirm that things are ready for real testing and I'm satisfied that no problems have cropped up yet.

The 3rd generation of the load transducer is close to being finished and some important improvements have been made. The area of flat signal response of the KZ has been increased along with other improvements that should make the system fairly easy to use.   

[George Schmermund]

The gen 2 KZQ force transducer has a central area that has the highest sensitivity to impacts. Now that I'm starting to standardize the testing and measurements using the gen 2 transducer until the gen 3 can be installed I thought it might be interesting to show the testings variables evolve. The confectioneries have changed from being an Oreo without the top cookie to a single cookie (ballistic tile) with an attached Milk Dud (putty) in the center. See photo.

The aim (pun) here is to try evaluating the putty in small amounts where the force transducer is most sensitive and reproducible. It's sort of a shotgun approach to getting initial values, but the testing will scale down to simple parameters quickly.

The photos are self explanatory to me, but may leave many readers nonplussed. I'm counting on Stan and Lloyd and anyone else out there to jump in and pry more information out of me. That would avoid a lot of possibly unnecessary details on my part.

[WhatUPSbox?]

George,
Thank you for the photos. The test setup looks great, looks like the Mild Dud approach works well.

I'm still trying to understand the shape of the curves you are getting. I didn't expect them to be so symmetric. Earlier you said you had a 10 Khz filter on the amplifier. With most of the curve falling within .2 msec is 10 Khz high enough? Could it be smoothing the curve?

[George Schmermund]

Stan - Thanks for the input. The symmetry of the impulse's shape is do to the filtering I'm using to reduce signal noise and DC offset. I've chosen to filter the low end at 2 Hz and the high end at 3KHz. If you look back to post #412 in this thread you'll see how an impulse can generate continuous frequencies and amplitudes within selectable bandwidths. With the KZQ materials I've chosen for capturing the pellets along with the force transducer's design it's possible to carve out a half sine wave shape that will hopefully be compatible with using the DS211 DSO as an acquisition system.

Keep in mind that a square wave with frequencies broader than the described filter, when run through that filter, will have something (on one half of the wave) which will be similar to the shape of an impulse when it's viewed in the time domain. If you evaluate the same filtered square wave with a spectrum analyzer you'll see a series of odd harmonics that the signal is constructed from.

On the other hand, a true impulse signal is constructed from an infinite number of frequencies (in theory). The amplitudes within the filtered frequency band I've chosen should provide a uniform and reproducible signal representing the impact. This should hold as long as the force transducer is linear within that band and the impacted material stays consistent within the dynamic range of the system.

 

[WhatUPSbox?]

I"ll have to think about that 3 Khz filter. Also, I thought the only assumption or feature of the impacted material was that it absorbs all the pellet momentum and there is no rebound. Adding a consistency assumption for the Milk Dud from shot to shot may be hard.

Perhaps it will all be clearer once you tie it together.

Thanks for the update.

[George Schmermund]

Here are a couple of photos to further explain the signal filtering along with the attempt to find the best size for a Milk Dud (MD) within selected energy ranges at the POI. The filter in use is set at 3KHz and is second order in effect. As can be seen it's not an axe like a digital brick wall filter would behave on this spectrum. The cursor is set at 2 KHz  to indicate the approximate area where just a single peak would be found if the impulse were a typical sine wave. The impulse has generated ALL of the frequencies shown in the shaded area of the spectrum.

The other photo is of the MD being pressed to a defined thickness (.5") using a platen. The MD is first formed by rolling a predetermined amount of kneaded putty into a uniform ball between the palms of the hands. The size and thickness of the MD has an influence on the impulse signal.

The previous reference to consistency was regarding the geometry of the MD. So far there has been no sign of nonlinearity in the makeup of the putty itself. The kneading and palm rolling seems to be reliable at this point. The thickness will have to be adjusted to accommodate different energy ranges. This will be a trivial matter once these thickness vs energy ranges are determined.   

[George Schmermund]

The gen 3 KZQ has gone through the preliminary tests with flying colors and is now ready to be installed into the trap. The clamping back plate just needs 2 holes for the bolts and a new round of testing can begin. The early tests on this new and improved design show a much higher sensitivity than the gen 2 and better linearity over a larger impact area.

When finished as a product the goal is to have an instrument that will give reliable POI information in the kill zone diameters for rats, squirrels, and rabbits. I suppose this could be referred to as the "Small Mammal Pest" model for humane kills or just a POI test instrument for those airgunners that want to raise the bar above spreadsheet guesses.

The photos show the gen 2 next to the gen 3 model.

[George Schmermund]

Lloyd mentioned something a few posts back about an analogy relating to punch presses and hydraulic presses. These presses use force differently. The punch press uses momentum which is a vector quantity and the hydraulic press uses pressure which is a scaler quantity. I didn't want to leave that concept just dangling out there.

[Nvreloader]

George
Thank you for the information you (and others) have provided.

I have read your first post, Hacking the Crosman Vigilante on another site and then here on GTA,
I have just finished reading your second post, Hacking the CP-2 , Great information, what little bit I understood,
I understood some of the information presented, even thou 99% was over my head.

In both of these post, 2 different sites, you have dangled a carrot, showed some photo's, in front of several of us,
and kept leading us down the rabbit hole path etc, are we going to get at least a nibble of the carrot?

I am sitting behind LLoyd and in the corner, as I know nothing..............

There is several of us that would like the info, and I for one am hoping you can provide it for this carrot:

The photo shows the Co2 cylinder heaters, Lithium-ion battery pack and the temperature controller, as they will be installed in the stock.
I'll make a mounting plate for these devices and do the simple wiring needed to power the heaters.
This has turned out to be much easier than what I thought it would be,
when the original heaters were installed into the pistol frame and powered using a bench power supply.

* 1-P1050159-001.JPG (92.91 kB, 1024x470

I am hoping that you planning to revisit this section? after all it is hunting season and too COLD to operate Co2 reliably,
at least please provide the proper names of the above equipment and the wiring diagram etc.
It would be greatly appreciated by us carrot nibblers, bring up the rear on this trail............

Thank you,
Don

[George Schmermund]

Don - That was a long time ago. I bought the parts on eBay and they've since been spirited away into who knows what other projects. The parts should still be available is my guess. I don't have a list or wiring diagrams. Never did.

If you and your friends are thinking about heating powerlets I advise great caution. The old adage "Don't step where you can't see the bottom" is applicable here.

[Nvreloader]

Thanks George

That is the thoughts I have, 2 different Co2 airguns to set up, A Crosman 160 and 150, 22 cals.
Both have massive heat sinks, with their tubes+barrels etc.

I am thinking about starting with the 85*F heater, and testing both, empty cylinders, and then an empty Co2 (fired) cylinder,
then a live cylinder in that order.

I have a very accurate  temp meter with an 18" long probe so I can get temp reading right over the heater/Co2 cylinder/bbls etc.

What is the proper name for the  temp controllers?

I for one will use extreme caution, I have had enough Kabooms in my life............Lol

Any and all info you can supply would be a great help.........

Thank you for your reply.
You can send a PM or email, whichever you feel is the best.................

Don

[WhatUPSbox?]

Don, as said be careful. Having an accurate temperature meter can provide a false sense of security. The placement of the sensor relative to both the heater and what is being controlled, as well as the thermal response time of each, all go into the design of the control system.

There are self regulating heaters that could help but unfortunately the ceramic PTC type all control to higher temperatures. There is a PTC rubber heater that controls lower but I'm not sure they are available at the retail level.

Not sure if you are planning to control the individual powerlets like George suggested in the Vigilante application or the tubes. The tubes are probably easier but could lead to a bigger Kaboom.

[Nvreloader]

Hi Stan

I have some question as to the heating of the cylinders/tubes and the Co2 gas once released, at this time.

I would believe that, in this order, that the heating to xyz temps would be,

1. Co2 gas in the tube, easiest heating/fastest response time
2. Steel cylinder/tube and Co2 gas, would be slower to get to temp, but hold the temp longer etc

There is No way to heat just the Co2 cylinder in these type of airguns, 150 pistol and 160 rifles,
(like George did with the cylinders in the pistol grips) as the wires stop that process etc.

That leaves wrapping the heaters (depending on which size used) around the Co2 cylinder and,
the bbl right directly above where the Co2 cylinder sets.

I am not sure if heating the bbl has any advantage as compared to the Co2 cylinder with gas only etc.
I am also thinking of machining a Co2 cap that has a 1800 psi burst disc in the end,
I would like to find a 1000 psi burst disc, LOL.

Lots of item to be considered to avoid a Kaboom...........

Your thoughts or suggestions would be greatly appreciated and not held liable..........

Tia,
Don

[WhatUPSbox?]

Don, excuse my ignorance of the classics. Do both the 160 and the 150 have a pressurized tube (like my QB78) and don't just press the powerlet against a seal like the 2240?

I think this would be easiest to implement on a bulk fill, then the 160, and the 150 being a little more challenging.

A custom cap would be my first choice. I don't think you need to heat the barrel. On the 160, I'm not sure you need to wrap wires around the tube (unless you are sneaking in a cozy hand warmer in the fore stock).