Barrel Vibration testing

[mks]

Mike I want to try and ask this in such a way so as not to allow the thread to be hijacked.
There have been several threads in the past where recoil, time and distance was discussed.
However I never once saw anything other than math for answers. For a basic lets say ppc or pick any other caliber.
1. How much time elapses from primer ignition and the bullet leaving an average barrel.
2. How much has the gun moved in a rearward direction due to recoil before the bullet leaves the barrel.
Rather than everyone start throwing in "depends on the variations, caliber etc etc.
Give me some kind of generalized answer..... appprooooox ? milisecs approooooxxxxx 1/4" recoil,,,2" recoil just some small idea is all I ask.
Thanks

These are some of the right questions to ask before a system can be designed for studying barrel motion. Here are some other important criteria:

1. Since the time the bullet spends in the barrel is roughly a millisecond, the sampling rate of the data acquisition system must be much faster, say 1 MHz.

2. During that millisecond, the rifle travels rearward about 0.1", while the muzzle moves vertically a few thousandths. Since we are most interested in controlling vertical motion to tune a rifle, we need some measurement mechanism that can focus on the small vertical measurement, in spite of the much larger horizontal recoil motion.

3. The vertical position of the muzzle has little influence on where the bullet strikes the target. What is most important is muzzle angle. One inch difference in elevation at a 100 yard target results from a less than 0.02 degree change in muzzle angle. Therefore, we need to measure the angle with greater accuracy, say +/- 0.0001 degree. Or if we approximated the angle by measuring two points separated by an inch near the muzzle, the difference in the heights of the two points would be less than 0.0003". Therefore, we would need to measure the two heights with accuracy much greater than 0.0003", say millionths of an inch.

Meeting criterion 1 is not a problem, there are plenty of systems with high sampling rate.

Criterion 2 is more difficult. In a previous thread, I think it was Vibe who suggested that initial experiments be done with a rimfire rifle clamped to a huge bench, to eliminate recoil motion. While there could obviously be some differences in the response of such a gun and that of a more typical BR rifle, it would be a place to start.

Criterion 3 is probably the most difficult. Measuring millionths of an inch can be done under a microscope, or maybe by capacitance. Small angles might be measured by bouncing a laser off a mirror attached to the muzzle and measuring the displacement of the laser beam at some distance from the mirror with a high speed camera. The laser/camera system might be best because it measures angle directly, but a 1 MHz camera is probably in the $10k-100k range. Capacitance would likely be cheapest and could produce good results. Whichever system is chosen, it would need to be isolated from the vibration of the rifle.

Acceleration is not worth the effort IMHO. It just doesn't measure what is really important. Acceleration must be integrated twice to approximate displacement, which causes inaccuracy. And one would need two accelerometers to find muzzle angle, which would involve even more inaccuracy.

Cheers,
Keith

 

[Vibe]

These are some of the right questions to ask before a system can be designed for studying barrel motion. Here are some other important criteria:

1. Since the time the bullet spends in the barrel is roughly a millisecond, the sampling rate of the data acquisition system must be much faster, say 1 MHz.

2. During that millisecond, the rifle travels rearward about 0.1", while the muzzle moves vertically a few thousandths. Since we are most interested in controlling vertical motion to tune a rifle, we need some measurement mechanism that can focus on the small vertical measurement, in spite of the much larger horizontal recoil motion.

3. The vertical position of the muzzle has little influence on where the bullet strikes the target. What is most important is muzzle angle. One inch difference in elevation at a 100 yard target results from a less than 0.02 degree change in muzzle angle. Therefore, we need to measure the angle with greater accuracy, say +/- 0.0001 degree. Or if we approximated the angle by measuring two points separated by an inch near the muzzle, the difference in the heights of the two points would be less than 0.0003". Therefore, we would need to measure the two heights with accuracy much greater than 0.0003", say millionths of an inch.

Criterion 3 is probably the most difficult. Measuring millionths of an inch can be done under a microscope, or maybe by capacitance. Small angles might be measured by bouncing a laser off a mirror attached to the muzzle and measuring the displacement of the laser beam at some distance from the mirror with a high speed camera. The laser/camera system might be best because it measures angle directly, but a 1 MHz camera is probably in the $10k-100k range. Capacitance would likely be cheapest and could produce good results. Whichever system is chosen, it would need to be isolated from the vibration of the rifle.

Cheers,
Keith

They are pricey, but Keyence makes such a device.

• Fastest in the World: 392kHz sampling rate
• Highest Accuracy in its Class: ±0.02% of full scale (Full scale being as small as 1mm or 0.04")
• Highest Repeatability in its Class: 0.005μm (.005 * .001mm = .000005mm = .0000002")


http://www.keyence.com/products/measure/laser/lkg5000/lkg5000.php?pb=LT.01.LK
http://www.keyence.com/products/measure/laser/lkg5000/lkg5000_specifications_1.php

 

[4Mesh]

Vibe,


Their repeatability is down to very tight tolerance (i'm skeptical), but the accuracy is more in line with what most of us would believe. Something tells me they have the wrong suffix in the .005. Yes, they clearly say .005um, but I've been around non-contact measuring for a pretty long time and as a rule, it doesn't split things down quite like that. Maybe that's possible but, that seems a bit optimistic for that sort of device. The accuracy part is not hard to believe at .02%. The repeatability is a bit much to swallow, expecially with repeatability so high and accuracy so relatively low. Lasers are kewl and all but they're not magic. I mean, they graphic on the page shows it looking at a casting...

If they mean .005 Millimeters (more likely), that would still be .005m .25.4 = about .0002". This I could buy. It would not be the first time I saw specs listed with optimistically errant numbers.

Even still, that device seems like it would be easy enough to make with components at home. Remember in this application, you don't have to have the accuracy, it just needs to repeat.

PS. The same pic chip listed above has the timer/counters for that job too...

 

[TRA]

That .005 is listed only for a measurement range of +/-.5mm.

Trying to isolate and measure only the barrel vibrations seems like an exercise in futility. It should keep you occupied.

 

[alinwa]

I feel that one of the best ways to map barrel vibration/launch angle is to shoot over a good chronograph and map the sine wave on the target.

al

 

[Don]

Acceleration is not worth the effort IMHO. It just doesn't measure what is really important. Acceleration must be integrated twice to approximate displacement, which causes inaccuracy. And one would need two accelerometers to find muzzle angle, which would involve even more inaccuracy.

Cheers,
Keith

Depends upon what you are trying to analyse, say if it were to distinguish between a .5 grain powder load difference in a gun I would agree, but if you were trying to distinguish the damping affect of a synthetic collar in a railgun block vs a metallic collar you would see the difference..................how do know? Ive observed it many times, much easier to distinguish large amplitude variations than slight frequency changes...............Don

 

[JerrySharrett]

My suggestion would be to duplicate some of Vaughns' equipment/experiments
Once you have acquired this basic level of knowledge and proficiency you will be able ask specific questions about equipment upgrades and vibration analysis that can be answered on these messageboards..................Don

Hey DJ, if we could get Greg Walley at Kelblys interested.....He came from the electronics world and has done some complex electronic testing on the Kelbly trigger timing and a few other items. Besides, he has a great range just outside his door and a great shop just inside!!!!

 

[4Mesh]

I feel that one of the best ways to map barrel vibration/launch angle is to shoot over a good chronograph and map the sine wave on the target.

al

If it were the biggest error component (by far) then I'd agree. But otherwise, there's too many other variables clouding the results. In other words, if that would work, then there's some really basic mechanicals troubles that need fixed first.

 

[Roy Allain]

Here is a relatively inexpensive "Vibrations for Dummies" universal laboratory kit (complete reference and support information data included) for around $180 and along with a $300 multi channel oscilliscope will get an individual well into the analysis of gun barrel vibrations, without having to build alot of the sensors, circuits, accelerometers, and other components that Vaughn had to develope in his efforts.............Don

see pg. 22/23 "Vibrations/Accelerometer design kit"


http://apps.meas-spec.com/myMeas/MEAS_download/Catalogs/Piezo/Piezo_Product_Guide.pdf


http://parts.digikey.com/1/parts/955190-sensor-piezo-film-design-kit-0-1004308-0.html

Don, the first http doesn't work. Is there another way?

Roy

 

[mks]

Depends upon what you are trying to analyse, say if it were to distinguish between a .5 grain powder load difference in a gun I would agree, but if you were trying to distinguish the damping affect of a synthetic collar in a railgun block vs a metallic collar you would see the difference..................how do know? Ive observed it many times, much easier to distinguish large amplitude variations than slight frequency changes...............Don

Don,
I'm not saying that acceleration is worthless, just not worth the effort. It is easy enough to glue an accelerometer to a barrel and get data. But the trouble is that the correlation between acceleration and barrel tune is not clear. It may seem intuitive that reducing acceleration would improve accuracy, but the opposite is more likely the case. From VarmitAl's simulations, we see that greater muzzle angular velocity (in the vertical plane) at the time of bullet exit is needed to compensate for variations in bullet exit time for all barrel contours tested. Further, increasing vertical acceleration doesn't necessarily mean that muzzle angular velocity at the time of bullet exit will increase.

Now transverse acceleration is another story. We want to eliminate all transverse muzzle motion. The correlation between transverse acceleration and transverse muzzle angular velocity is similarly imperfect, but is probably good enough if the goal is zero. The exception would be when the accelerometer is located on a node, in which case it would read zero regardless of the angular motion.

Cheers,
Keith

 

[mks]

I feel that one of the best ways to map barrel vibration/launch angle is to shoot over a good chronograph and map the sine wave on the target.

al

Good point, Al. Throw in some intentional variations in powder charge/muzzle velocity, and this is probably the easiest, and arguably the most relevant, way of getting an indication of muzzle angle.

 

[Boyd Allen]

At the risk of being pummeled, I think that we have sufficient information as to how the parts work, but the subtleties and unpredictability of all the various combinations' interactions may prevent useful prediction of outcome, and so, to a large extent, we are mostly left with cut and try. For example, I know of no barrel maker that claims that he can, by inspection, pick out a hummer barrel. Some things can be done "by the numbers" some cannot. We are like the weather man who says that it rained, and here is why it did. Things are a lot easier to explain than predict, and many times the explanations are no more than experienced guesses. Ya...I know; you knew all that.

 

[Don]

Don, the first http doesn't work. Is there another way?

Roy

Roy, this should get you there, see last page of the pdf guide...........Don


http://www.meas-spec.com/downloads/Piezo_Film_Product_Guide.pdf

 

[Greg Walley]

Hey DJ, if we could get Greg Walley at Kelblys interested.....He came from the electronics world and has done some complex electronic testing on the Kelbly trigger timing and a few other items. Besides, he has a great range just outside his door and a great shop just inside!!!!

Jerry,

I looked into barrel vibration measurement at one time...and it was Don Jackson that gave me some pointers on sensor selection and mounting. That was several years ago, and measurement technology has come a long way since then. There are very low mass devices that would probably work well for our application and the barrel tapers we use on BR rifles.

I never bothered with the project due to all of the work involved for obtaining data that probably won't be relevant to what we are looking for. However, as 4Mesh has pointed out - there are many new ways for cheap data acquisition as long as one has a modern laptop. There are some freeware programs that are user friendly. I did a lot of research and planning a few years ago on this topic, but unfortunately my hard drive crashed two weeks ago, so I don't have ready access to that information at this time.

I became more interested in measuring strain, but I was a little discouraged with finding a way to control for temperature with the sensors available at the time.

There was a paper published on measuring gun barrel vibration using fiber Bragg grating sensors coupled to a spectrum analyzer. I think it was published in Measurement Science Technology. I'll see if I saved this article elsewhere and post it here, since it had a lot of infomation pertaining to what we're looking for. If I recall correctly, the article addressed the complexity of temperature compensation. The above technique would be expensive, so basic analog I/O with standard strain and vibration sensors would be the way for the hobbiest.

It might be possible to use a freeware audio analysis program with FFT, like Steinberg Wavelab and read the sensors through the high impedence microphone inputs on the laptop soundcard, and do all of the filtering through the software.

 

[Vern]

Serious question.....
why is it I only see for the most part vertical vibration addressed when there can be horizontal.

In that respect what about an elliptical tuner that could address that issue?

 

[mks]

Serious question.....
why is it I only see for the most part vertical vibration addressed when there can be horizontal.

In that respect what about an elliptical tuner that could address that issue?

Vern,
Don't forget axial and torsional vibration, too. The reason vertical vibration is discussed most is because it is normally the largest component, and the only component that we want some of. All the rest we want to be zero. Vertical vibration is greater than horizontal vibration because bolt thrust is above the center of gravity of the rifle, but is nearly aligned with it side to side. Pure axial and torsional vibration don't contribute significantly to shot dispersion.

Keith

 

[mks]

...I think that we have sufficient information as to how the parts work, but the subtleties and unpredictability of all the various combinations' interactions may prevent useful prediction of outcome...

Boyd,
I am flabbergasted by your logic. How can one not need more information on something that is currently unpredictable? Consider yourself pummeled.

Cheers,
Keith

 

[David Merrill]

From the point of view of accurately measuring the other components of lateral barrel motion, don't forget that there is a dialational transient as the bullet and gas pressure pass the measurement location.

Also, there is a bit more to the effects of transverse barrel muzzle motion than simply the pointing direction of the muzzle at bullet exit. Again, research the meaning of 'muzzle jump' and 'aerodynamic jump'. Finally, don't lose track of the bottom line objective that it is the shot-to-shot consistency that contributes to target accuracy, not just the magnitude of the barrel motion.

David Merrill

 

[Vibe]

Serious question.....
why is it I only see for the most part vertical vibration addressed when there can be horizontal.

In that respect what about an elliptical tuner that could address that issue?

Vern,
Don't forget axial and torsional vibration, too. The reason vertical vibration is discussed most is because it is normally the largest component, and the only component that we want some of. All the rest we want to be zero. Vertical vibration is greater than horizontal vibration because bolt thrust is above the center of gravity of the rifle, but is nearly aligned with it side to side. Pure axial and torsional vibration don't contribute significantly to shot dispersion.

Keith

You guys are starting to hit upon one of the reasons I've been objecting to the "parallel node" theory and terminology. We have to get past that mental picture (or theoretical model) to start to understand that mass at the muzzle only has an impact on one aspect of barrel behavior, the distribution of that mass - IE centroid in relation to the muzzle and barrel stiffness affects other aspects, and the overall mass moment of inertia of the tuner itself would impact aspects not even being discussed yet, such as these issues that you bring up.

 

[Boyd Allen]

If we imagine that a thing works a certain way, we only have a theory, until testing verifies it, and even if the result is an improvement in performance, there is a good chance that we may not totally understand the reason for the change, because in making the intended change, it is highly likely that we have changed something else, that we were not looking at. I love imagining how things work, with an eye on making them work better, but without the capability of building what we imagine, so that we can test as we go, all that we have is a pleasant, albeit harmless distraction. This is why my next goal is not a new rifle, but a lathe, so that I can build and test ideas....cut and try. It is hard to write this as I sit recovering from my recent pummeling, but I somehow manage to carry on, despite my injuries.
Boyd