Hacking the CP-2

[WhatUPSbox?]

George, the sights look great. Now I don't have to order those glow in the dark jelly fish tentacles to glue onto the front sight. I guess you have several options for the rear blade.

[cobalt327]

There are tritium vials available that are suitable for gun sights. I got a few to replace the night sights of my G22- cost for the trit vials and UV cure adhesive was about $15, new night sights are well over $100.

[George Schmermund]

Looks like you may be close to the point of no return on making AG measurements, Stan. The microphone approach is a good starter exercise. they're cheap enough to be expendable. When you start upgrading your transducers you'll be glad the you laid the ground work with the cheap stuff.

I noticed that some of your signals are clipped on the peaks. This can really overload the amplifiers and their recovery time in some measurements. It may be worthwhile to check each amp in the string including the front end of the scope. Phase shift is also worth looking at. A tunable filter is also good to have when you're trying to control noise levels.

Here's a link to a nifty little hammer: https://www.harborfreight.com/watchmakers-hammer-with-6-heads-99895.html . You can use this for many test setups. The interchangeable heads offer lots of creative opportunities. One would be to attach a microphone to one end and use the copper tape trick that you applied to the snapper on the impact end. With the hammer set up as described you can trigger the scope with the tape and look at rise and delay timing and phase shift of the mike (I'll stick with the old abbreviation). With the interchangeable heads you can make a very useful 'instrumented hammer' (worth Googling). You can't beat the price.

[WhatUPSbox?]

Thanks George, I'm intrigued with how far one can push very low cost instrumentation if you set up the conditions appropriately. Yes, I'm sure I'll get sucked into the better transducer cycle.

I saw the clipping on the traces, I postponed tweaking that since I was looking for the initial response. The scope has some filter capabilities built in that I also need to explore.

Thanks for reminding me of the little HF hammer. I was planning to make a poor man's smart hammer from some delrin rod I have but the HF approach is more flexible...I don't need much of an excuse to go there.   

[George Schmermund]

I was day dreaming between beers this afternoon and had another one of my induced epiphanies. I was quite impressed with these micro LEDs and thought about how easy it would be to EDM some 0.038" - 0.040" holes through a barrel at intervials. A series of light gates could then be made by putting an LED on each side of the bore to get really good acceleration numbers to overlay onto the P/T curves. Don't calculate what you can measure. The photos show an LED.

Of course the measurements would be best served if I had a slick 4 channel scope like Stan has. I feel as though I'm being inexorably 'channeled' toward the edge of another hardware precipice.

[Finn]

Wouldn't those 16 channel cheap logic analyzers do the job just fine.

[George Schmermund]

After giving more thought to using the LED's for pellet acceleration measurements I realized that they can be made using analog methods. I've collected a substantial number of nuclear instrumentation  modules (NIM) over the years and the bins that they operate with. It will be easy to use an Ortec 533 Dual Sum and Invert Amplifier and then run the summed LED output to a  Canberra 2025 AFT Research Amplifier. The conditioned signal can then be routed to an Ortec 551 Timing Single Channel Analyzer (TSCA). The TSCA output pulses would then go to one channel on the scope. The LED timing pulses could then fall into the same time window as the P/T curve. This arrangement would allow me to put off the new scope decision for a while longer.

This is actually an easy setup to put together. The NIM system is standardized across the industry so the you can mix and match different manufactures. The instruments are like Legos.

I'll do some simple scope tests to make sure that the LEDs will work in pairs and talk to each other. This may be the answer to my transit time/acceleration questions. The data may also be of interest to some of the spread sheet folks out there.....

[George Schmermund]

Further testing on the LED timing experiment was an enjoyable distraction to confirm that the devices will talk to each other. The polarity of the listener needs to be reversed in this experiment. The talker was driven directly from the sine wave output of the 3310B @ 25 Hz. I didn't measure the voltage, but the 50 ohm output had no trouble lighting things up.

The photos show the scope trace of the listener LED and the lashed up "optical cavity" used for the test. The Al foil was folded and used to shield most of the wiring. This experiment indicates that the barrel light gates should be easy to implement.

I'll probably use one of the Vigilante barrels to set up the first gating system. The CP2 barrels are at a premium right now, so I hesitate to start making holes in them until this measurement is proven to be headed in the right direction.

I'll have to start pedaling backwards now to get back to the barrel harmonics experiments.

[WhatUPSbox?]

OK, a little bit more tinkering with instrumentation. I set up the 2240 again and added a probe to try to detect hammer release (image 1). The probe is an insulated rod (barbecue skewer) with an electrode (brass plated pin) in the tip (image 2). The probe contact is a little finicky to set (should have left in a bit of compliance) but seems to trigger cleanly and repeatably. I mounted the capped microphone onto the barrel band and suspended the bare mic near (1.5 inches) the muzzle.

Image 3 shows a timeline trace for the 2240 without a pellet. Channel 1 (yellow) is the capped mic, channel 2 (blue) is the bare mic, and channel 3 (purple) is the probe. Looks like the overall timeline from hammer release to muzzle blast is around 13 milliseconds with most of that in hammer travel. I'll have to weigh the hammer and measure the spring to see if a prediction comes anywhere close.

Image 4 is more detail around the time of valve impact. Not sure how to read the activity on channel 1. I'm trying to think of an explicit way of detecting hammer impact. Maybe drill a hole in the side and put in one of George's fireflies.

Fun stuff, though George's nuclear instrumentation modules sound way cooler than a barbecue skewer.

[George Schmermund]

You're really moving along with the testing, Stan. That new scope sure collects a nice ensemble of inscrutable data. Don't forget the phono pickups. The stylus cantilever is very small and can easily be extended into an orifice.

I like the blood on the business end of the skewer. Nice touch.

[George Schmermund]

I was getting back to the Pressure/Time measurements on the .22 pistol today. When the pellet loading was tested after reassembly, the force needed to insert a pellet with the probe was excessive.  I haven't played with this gun (.22) for a while since most of my time has been spent with the .177. The same problem with hard loading with the .177 was cured by doing the Dremel treatment with the green grind/polish burr.

I had previously done a light Dremel treatment to the .22, but not as extensively as with the .177. I dissembled the .22 again and inspected it. The breech entry had the sharp factory edge broken, but the corner was still pretty much square to the bore. Being late in the afternoon I was able to focus my attention more closely through the usual agent. Now it became apparent that the large force needed to seat the pellet into breech was due to the skirt pressing into the o'ring. This caused the o'ring to then compress onto it's front sealing surface with no place to relieve the extruding pressures.

The barrel was placed back into the lathe for another Dremel treatment with the larger burr. This time the the corner that was previously just broken was now ground and polished to form a chamfer. The chamfer allows the o'ring to extrude into it and dramatically reduce to pressure need to load a pellet. The o'ring is now less traumatized by the loading exertion and recovers itself to still make a good seal against the probe.



 

[huklbery]

The barrel was placed back into the lathe for another Dremel treatment with the larger burr. This time the the corner that was previously just broken was now ground and polished to form a chamfer. The chamfer allows the o'ring to extrude into it and dramatically reduce to pressure need to load a pellet. The o'ring is now less traumatized by the loading exertion and recovers itself to still make a good seal against the probe.

Would it be possible to see a pic of that?  I lightly rubbed my entry for a touch smoother entry the idea it may allow the oring to jam was a background worry as well. 

[George Schmermund]

Here's a shot of the chamfered breech. This is the .177 barrel. The treatment is about the same as the .22 breech got. I have no dimensions because nothing was measured. Just guess work until it looked about right. This process can't be easily reversed, so you're on your own.

[George Schmermund]

Here's a screen shot from of the first Pressure/Time tests. The red trace is without a pellet. The aqua trace is with a pellet. There's some interesting stuff in here already. Pressure scale is ~100 PSI/vertical division. the time scale is 2.5 mS/horizontal division. Nothing is calibrated yet (except the scope's time base),  but these numbers are close to real. The traces were offset a little for clarity.

Next will be to add the force transducer to the muzzle to get the pellet's transit time in the barrel. This is starting to get interesting. I'll probably move the measurements into the electronics area since the footprint of the instruments is comparatively small and I prefer to use the Lecroy scope for these experiments.

[George Schmermund]

I've still got the instruments from the earlier CP2 measurements in the machine shop area, so a few more experiments were in order before things go back to the electronics room. The force transducer, used back in the Vigilante era, was again employed this evening to see what the transit time would look like with the simple Owon scope. This is starting to be a compelling set of measurements even with these quickie tests. The pellet transit time is ~ 3 mS with some gas left over. Here's where tuning would be helpful.

Things are starting to gel.

[George Schmermund]

The pressure in the power plant is starting to drop and the pellet's exit time is starting to increase. There are some interesting pressure bumps that occur after the pellet exits that can be examined in more detail with this simple setup. I'll put another pressure transducer at the muzzle to get much more detail from the bumps.

[George Schmermund]

With the gun setup in the vise it was easy to swap the .177 modified piston cover over to the .22 main tube. The screw can now be used to adjust the hammer spring tension and direct measurements of the effects. I did a couple of quick tests and the results were very interesting. I'll do some more testing this evening and report the findings.

[George Schmermund]

Some testing was done today to see what tightening the power adjustment spring would do. The photos show the preliminary results. The screw was just touching the spring at 1/2". This would be the factory setting. The second photo is a screen shot of the 1/2" results. The third photo is a screen shot with the 1/2", 1/4", and ~ 3/4" settings (where the spring bottomed out) overlay of the 3 tests.

The pressure is ~500 psi for each shot. The transit times changes from 2.88 mS @ 1/2"(no compression) to 2.76" mS @ 1/4" to 2.72 ms @ ~3/4" (full compression). This is just a first look at things with the limited Owon scope. I'll switch the experiments over to the Lecroy scope and have it do the integral calculus for me while I drink a few beers. The numbers that it generates will be very enlightening to me for sure and others, maybe.

[George Schmermund]

Now that the P/T curves and transit time markers can be measured I'd like to add an accelerometer closer to the valve stem and another pressure transducer at the muzzle. I could just use the ext. trigger as a third channel using the accelerometer output, but that won't let me see the hammer/valve stem impact waveform.

I'm designing this more extensive experiment in order to force myself to get a 4 channel scope. After viewing Stan's post about the EEVblog video review of the Rigol DS1054Z I checked out the same reviewer's assessment of the GW Instek GDS-1054B 4 channel scope. I found myself enamored with most of what the reviewer said he didn't like about it.

I like a larger front panel. There's more room between knobs and switches. These are assets when you're getting old and feeble. Beer vision is a bonus when you're creating experiments in your head, but in the real world of making measurements a little extra room on the front panel is very helpful when you're fumbling with adjustments. I want a knob for each channel to reduce the amount of thinking required to set things up. Also, I'm not inclined to make the choice of an instrument based upon what color it is. Maybe a camo front panel would be more universally accepted nowadays.

At this entry level of scope hardware it mostly boils down to choosing a Ford or a Chevy. My personal choice is the GDS-1054B.

[WhatUPSbox?]

George,
Not sure I understood the 3/4" setting in your earlier post. I thought you started at 1/2" and moved in.

I'm designing this more extensive experiment in order to force myself to get a 4 channel scope.

I understand that process well