Hacking the CP-2

[George Schmermund]

The pressure measurements got stalled because I wanted to do the same transit time tests on the .177 cal. unit while the instruments were still set up from the .22 tests. I'm glad I did. The photo shows the results from the .177 gun with its 8" barrel and powerplant. These pellets were Crosman 7.4 gr Destroyers. The temperature was 70º F.

This is definitely a different time sequence from the previous test. This would be expected. The highlights of interest are the differences in how the hammer strike effects the valve timing. The picture of events is much clearer than the .22 screen shots.

My interpretation of the time events, through the optics of a few beers, is that the first peak is when the hammer hits the valve stem and triggers the o'scope. The hammer then loose some energy and starts to slow down as the valve begins to open. The next peak is when the hammer slams into the end of the valve body and deposits the rest of its momentum into the gun using the valve body to transfer it. The valve spring and compressed CO2 then push back against the hammer and cause a spring / mass rebound to start. The hammer again hits the valve stem and begins another cycle. This is a simplified explanation and hopefully leads to a clearer idea of what the measurements are telling us. Keep the beer vision in mind, too.

OK, I know all of this is preaching to the choir, but what I'm finding interesting is how fast the first rebound occurs. The valve appears to have opened and close twice in less than 900 µS which means that the pellet is still in the barrel while that is happening. If I calibrate the force transducer It will determine the true exit velocity with more certainty than a typical chronometer. For now I'm just interested in collecting some numbers.

The pressure tests become a much more compelling measurement now because they can demonstrate that what we see with just timing information can be correlated with a time/pressure measurement.

[WhatUPSbox?]

George,
Looking forward to the pressure timelines.

The hammer/hammer spring should be a fairly low frequency system. Not sure it can respond fast enough.

[George Schmermund]

Stan - You're probably right about the hammer/ hammer spring. I'm going to switch my guess to the valve stem/ valve spring. They're much less massive.  I'll do some tapping with the small hammer and run the signals through the filter and measuring amplifier and see what I can find.

The time/pressure measurement scheme has changed from putting the pressure sensor through the side of the main tube and valve housing. The sensor is now going into the center hole of the breech lock down screws that hold the barrel in place. This change will allow the pressure information to come from the transfer port instead of inside of the valve. I'll drill through the set screw hole and the barrel opposite the transfer port hole and tap it to match the sensor. The pressure info will now come from right behind the pellet. A 10-32 set screw can then replace the sensor when the testing is done.

[George Schmermund]

Some progress was made today toward getting the pressure sensor into the breech and barrel. Next is to make an adapter to screw into the newly threaded hole to act as a height stop to keep the sensor out of the bore. The sensor will now look directly down into the transfer port and see the same pressure at the same time as the pellet. This would have been much harder to accomplish on the Vigilante. I'm hoping that the powerplant is clean enough not to foul or destroy the sensor.

The photo shows the set up in the mill with the .22 cal breech and barrel. I'll do the same procedure on the .177 pistol.

[George Schmermund]

I got to play with the pressure transducer installation today and It's about ready to start making some measurements. The placement over the transfer port was a good decision in the overall scheme of things. These miniature transducers from Endevco are really nice. They're right up there with my favorite B&K instruments.

I'm working on a calibration system for the accelerometers so that they can be used for measuring muzzle displacement when the testing gets back to barrel harmonics. Terry (dv8eod) sent me some interesting info back when the Vigilante barrel projects were alive. I've been integrating his information into the mix. It will be a good place to start the new measurements.

The photos shows how the pressure transducer is installed.

[George Schmermund]

It's easy to get distracted when instruments are sitting around the work area and asking to be put back in harness. I've spent the last couple of days setting up the barrel harmonics experiments. The measurements should be ready to go in another day or so. I'll be back to the pressure/time testing when the instruments for those tests are set up.

Here's the current calibration setup for doing the muzzle displacement measurements. I love these old analog instruments. It's really enjoyable to to turn the knobs and flip the switches on B&K stuff. It makes it very easy to see where the signals are coming from and going to when I'm trying to learn something new. I can break-in anywhere along the line if I think something anomalous is happening. My brain seems to work in the same block diagrams.

The instrumentation photos include:

HP 3310B Function Generator

B&K 2706  Power Amplifier

B&K 4810 Mini Shaker

B&K 4384 Calibration Accelerator

B&K 2511 Vibration Meter

B&K 1617 Band Pass Filter

B&K 2636 Measuring Amplifier

Owon Scope

[George Schmermund]

For clarity concerning the accelerometer calibration I want to add that the 4384 device is being used with the 4810 shaker to produce a traceable displacement signal. The photo shows how the Endevco 2250 low mass (0.4 grams) accelerometer is being calibrated. The 18" .22 cal barrel, including the thread protector, weighs 304 grams.

I'm confident that the added accelerometer mass won't interfere with the barrel harmonics measurement, but I'll still do a stimulus/response measurement with the non-contact VRT devices to make sure that the resonant frequency doesn't shift.

[WhatUPSbox?]

That's a pretty nice collection, looking forward to the data

[George Schmermund]

I was sorting through some boxes in the attic and found one the Schlieren optics sets along with assorted other stuff. These mirrors are 10" f/1.6 type. This makes them very fast as mirrors go. Their short focal length will be an asset for setting them up on a reasonable size optical table.

I wanted to see what the arrangement would look like so I set them up on the dining room table. My wife came in and asked what the equipment was for. I informed her that it was for doing some ballistics experiments. She then informed me that if I thought I was going to do any airgun testing in HER dining room I'd be looking for more than just a new hobby! Something about her grandmother's china. Wives and physics are not always the best mix in this household.

Anyhow, now I have a feel for the footprint of this experiment and can plan accordingly. I'll have to make some mounting hardware and arrange some precision translation instruments to get things going, but the project is becoming more compelling as time goes on. I'm already a year behind on this project.

The photos shows the main hardware needed to get started. This includes the mirrors and a high intensity light source with an adjustable aperture and flexible fiber optic cable. When the system is all adjusted the Micro Flash system will replace the conventional light source.

[AmBraCol]

Funny how wives are not always as enthusiastic as their husbands about experiments in the dining room, kitchen, etc.

 

[WhatUPSbox?]

Dabbling in air pistol timing .... without George's toolbox

I wanted to try some of the timing measurements with amateur sensors. I set up a 2240 just because it was convenient...and happened to have a powerlet in it. I ended up using a pair of small microphones and an amplifier from a Velleman kit (MK136 <$10). I fed this to a scope, but a soundcard app may work as well.
Initially, I compared the impulse (yes, I used George's mousetrap clapper) response of the amplified microphones to raw response output from a couple of small (headset) speakers and there didn't appear to be significant lag through the amplifier so I went with the convenience of the microphones.

The first picture shows the 2240 with one microphone stuck to the barrel band and the other loosely hanging to decouple it from the barrel. The second shows the trace for a shot without a pellet. The channel 1 (yellow) is the suspended microphone its lag is consistent with speed of sound at 74 micro-sec/inch. The second spike in the yellow trace is the echo from the table top 8" away. The channel 2 (blue trace) is the mounted mic and has both acoustic and mechanical responses. The speed of sound lag helps identify the acoustic ones.

The third picture just shows the clapper board test result. The interesting part is the very front where the amplified microphone and the small speaker raw responses (channels 3&4) happen within  about 20 micro-sec of each other ( the yellow trace is the suspended mic and the lag is generally consistent with speed of sound).

None of this replaces results from quality lab equipment, but it does give me a little better understanding of the results George is providing us...and gives me motivation to hunt down an accel or 2.

Now back to the real data in this thread.

[George Schmermund]

Stan - It's nice to see someone else taking the plunge into the mysteries of ballistic tests and measurements. Starting with things that are readily at hand and re-purposing them to do your bidding warms my heart. We need more experimentalists contributing to these "Engineering - Research & Development" threads.

I envy your 4 channel scope. That opens extra doors when making simultaneous measurements in a triggered time window. I've been on the fence about choosing one. How did you make the decision on the one you chose? There are a few questions I have about some of the data you've presented, but I'll consider them to be irreverent at this point in the measurements.

Adding accelerometers to the mix is certainly a leap forward in these tests. A cheap and expendable intermediate device can be fashioned from an old crystal/ceramic phono pickup. The old mono ones were usually designed to detect displacement in only the vertical direction. Their output is generally high so they don't need a preamp to produce line level signals. There are lots of them on eBay. Even lots of lots. I'd stay clear of the oldest ones. Something in the era of 45 RPM stuff should be useful. Try to get the ones that still have the cantilever visible. The condition of the actual needle itself doesn't matter.

With some creative fixturing you can arrange a moveable contact point that is very sensitive to vibration, but virtually immune to the acoustical output of the DUT. The mass loading is negligible. These are basically "throw away" sensors so they can even be used as targets. Raw piezoelectric discs are also available. These would also make good targets. A simple design could make the discs survivable for many tests.

Keep up the good work and report your findings!

[WhatUPSbox?]

George, the 4 channel DSO is a new addition, my old analog scope wasn't useful for this. I blame Bargain Gate here because of the $50 off ebay coupon a couple of weeks back.

It is the Rigol 1054z. https://www.rigolna.com/products/digital-oscilloscopes/1000z/ They also have a promo through the end of the month that adds a bunch of software functionality. I got it through one of their distributors but they also have their store...I was tempted by their open box items.

I'm not a EE so I relied on a decent (enthusiastic) online review

 .

 Basically I was looking for 4 channels and the ability to store data to USB...The rest of the features I'll have to learn as I go along. I've also been poking my nose into the arduino ecosystem and its wealth of dirt cheap sensor modules so that helped justify the purchase....but mostly it was an (another) impulse buy because of bargain gate here.....

The scope is very pleasant and intuitive to use, even for a non-EE

[Cindi]

Hi George,
 I'm enjoying your  thread, of course it has my brain hamster running amok. 

The pressure sensor looks like an Endevco 8510B series strain gage based transducer.
   Which Endevco transducer are you using?
   What is the cost range for the ones you have?

Have you been able to take any pressure measurements on your CP-2 project ?


Why I'm interested:
I'm trying to determine the cost of putting together a pressure measurement setup for a PCP air gun suitable for measuring the air pressure rise and fall time behind the pellet after the hammer strike hits the discharge valve.  Given the pressure levels and valve timings, I think that an 8511A series transducer will be needed. It is unclear what I will need to capture the measurements. Dataq has some low cost USB data loggers that can cycle up to 160KHz, but  will need an instrument amp to boost the signal level.

-Cindi

[George Schmermund]

Thanks for the scope info. Your choice looks like a good one. Lots of capability there for the price. I'm still vacillating about what to get. In the mean time I've fallen prey to a B&K 2515 Vibration Analyzer I found on eBay. I'm setting it up now for an enhancement of the barrel harmonics displacement measurement tests. There's a point at which I'll come to my senses and switch everything over to the HP DSA and be done with it, but like they say " The journey is the reward!". I'm in no rush to finish this project.

As far as what to use as a pressure transducer, there are quite a few choices out there. Right now I'm using an Endevco 8510B - 500. It has a conservative rating of 500 PSI, but it specs out to 3X pressure. I won't be using it at more than 2X with CO2. The 8511 series would be good for higher pressures but their form factor is considerably larger than the 8510 series. The higher pressure transducers I've collected are Kistler 601 series 1500 PSI and 5000 PSI.   I did some preliminary P/T measurements in the "Hacking the Crosman Vigilante" thread. The CP2 in coming up next.

A signal conditioner is a good device to have so that you can adjust the signals from the transducer to calibrate them. Check out the transducers' websites and study the brochures and tech notes to find really good information.

I've got a couple of the Dataq acquisition devices and they have their places in some experiments, but I find their software to be somewhat clunky and not all that friendly. The software I'm using came with the units so I can't really complain about the price. A data logger is another animal in many respects so be sure of what and how you want to use them.

A simple digital scope would be a necessary instrument for these measurements in my opinion. If you're already familiar with scopes you're well on your way to making useful measurements.

All of the above hardware, and lots more, is available on eBay for very low prices if you are a good shopper.

[Cindi]

Thank you George,

I download the tech info and contacted Meggitt Sensing Systems with some detail questions. I now believe that an Endevco 8510B - 2000 is a good candidate for what I want to do, as you noted, they test them to 3x the rated measurement range, they don't guaranty linearity beyond 2K psi (for that model), but they stated the data is still sensible. The response time looks good for a PCP pressure curve, unfortunately the list price made my jaw drop a bit, equivalent to a very nice PCP air rifle purchase.  I've set up some ebay searches to try and mitigate the cost.
 Its been multiple decades since I've used a scope and don't have digital scope currently (although they have come down in price significantly), I'm comfortable hacking software and communication protocols (which Dataq publishes) didn't look terrible to work with. This darn PCP rabbit hole is consuming lots of dollars, I guess I'll need to fire up the CNC machines and see if I can make some bucks to afford new test equipment and sensors.     

Also, I'm part way through the  "Hacking the Crosman Vigilante" thread, my complements to you, you are having the kind of fun I'm interested in. I'm also great at self induced exploration side trips.

[WhatUPSbox?]

Before I made the impulse buy on the Rigol DSO, I was looking at the DS212 handheld scope. 2 channel, 1 mhz bandwidth and very portable. all for around $100.

Here is a review

[George Schmermund]

Thanks for the post, Stan. I enjoy your sense of humor!

[George Schmermund]

A few posts back I mentioned that the CP2 sights were pretty much useless in low light conditions. I suggested that a micro lighting system incorporating an LED and a piece of retroreflective tape might solve the problem. Rob thought that the idea was amusing. His comment was "hahaha, electronic open sights , cant wait". After his comment sank in I got a good laugh out of it myself. What was I thinking?

Well here it is. Electronic Open Sights! I went out to the shop this evening after a few beers and was going to play with the newly acquired B&K 2515 and do some calibration. I inadvertently started sorting through some of the composting piles of project material that had finally arrived from China some weeks ago. In the middle of the heap were the vanishing small LEDs that were ordered. Then I found the strip of retroreflective tape under some pellet tins and the intended course of the evening got derailed.

Being Saturday night by this time I relied on a pint of locally brewed strong ale to see me through this unexpected detour. The photos show what the basic scheme is. I didn't spend a lot of time doing the camera work because the focusing seemed to become erratic. Strangely, the camera's performance always gets much better by the next morning.

The battery was from a collection of worn out ones that were in red dot scopes that I forgot to turn off. The battery didn't last long, but got through this session as a proof of concept test. I think I'll pursue this protect.

[WhatUPSbox?]

I did a little more tinkering with the approach of using miniature microphones for timing dynamic events. I wanted to attenuate the acoustic response and leave the shock response through contact. I made a thick aluminum cap that fit snugly on the front of one of the mics (I used a cap because for now I wanted it removable, otherwise bonding on a thick disk would serve the same function). The other mic is bare. In this case they are both suspended in air about 3" from impact. The first image shows the response to the clapper impact. The blue trace is the bare mic and responds to the acoustic signal. The capped mic (yellow) is quiet...so far so good.

I had some copper tape so I upgraded the clapper with a digital trigger (image 2). This seems to work well. It has bounce as expected but triggers on the initial impact.

So I mounted the capped mic to the clapper board and left the other mic suspended (image 3) and ran a trace (image 4). The clapper (purple) triggers the sequence, about 80 micro-sec later the mounted, capped mic (yellow) responds through contact, and at around 268 micro-sec from trigger the suspended mic (light blue) responds. The suspended mic is about 3.5" from impact so at 74 micro-sec/inch (speed of sound) there should be a 259 micro-sec delay. Pretty close. This seems to indicate that the basic microphone and amplifier does not generate a large delay. The delay for the mounted mic is harder to predict since it goes through some bonded joints.

I'm encouraged enough to invest the $5 it will take to get a fistful of 6mm mics and modify some of them with the cap/plate. The 6mm size should be more convenient for sticking onto airguns. Also get some amplifier boards  ($1-5) to get more channels.

Yes, having fun in the instrumentation rabbit hole