This Is What Can Happen.....

[chekbone]

Seems likely to me that for whatever reason (the blind damaged by the HiPAC hitting it, for example) they hung the green sheet over the blinded window behind the work station after the event

Seems likely that he could have thrown the gun into the window and needed a blanket to cover it up, but why not show all the damage if it's part of the event.

[rsterne]

What a coincidence! I suggested that in post 273 on page 14. Guess i should have called it My "steel-quake" theory.
Sorry Bob, but I did too d-mn much searching and reading for you to claim the idea as being originally yours. With the, "my steel-quake idea" no problem, it'just the same question I asked that went unanswered. Isn't that like "Yellow Journalism", by definition.

Please refer to my post #268 on page 14 where I coined the term "steel-quake".... five above where you suggest a "shockwave", not that it matters.... I used the term "tongue-in-cheek" to convey the idea that energy travels in earthquakes and can do damage some distance away.... and also to differentiate it from a "shockwave", which can be in any medium, particularly air, which is what I thought you meant in post #273, because you used the term "back pressure".... I mean you couldn't have meant what I said 5 posts earlier, without referring to it, could you?.... Thanks for the insult, accusing ME of stealing YOUR idea, BTW.... 

Bob

[Gippeto]

Slow it down guys...you're both talking about harmonic vibration...a (somewhat slim) chance there could be something to it as well...from launching the valve as a projectile.



Al

[chekbone]

Thanks for the insult, accusing ME of stealing YOUR idea, BTW.... 

Bob

I stand humbled and corrected. To that end i withdraw my post and offer my sincerest apology.

[rsterne]

I agree, Al, there is a VERY slim chance that is to blame.... However, we are left with VERY few options IF the Valve and HiPac BOTH departed the gun BEFORE the pressure dropped below the point where it drove the Valve into his leg AND the HiPac into the ceiling.... That is why I suggested something so far out....

Bob

[ChrisK]

This part of the statement....

massive rapid change in air hit the air valve like a hammer

doesn't make a lot of sense, but I can't disprove it.... It is possible that a shockwave from the first part that failed triggered the second one to fail.... and that shockwave could have been transmitted through air or steel (my "steel-quake" idea).... It is also possible that the gun "rang like a bell" vibrating violently in several directions, and a harmonic did in the second part.... I suppose it's even possible, if the valve failed first, that the sudden lack of hoop stress in the O-ring groove in the HiPac caused it to reverse or flatten it's curvature and fracture before the air pressure dropped low enough to not blow it off the end of the gun.... The point I have been trying to get across is that it is POSSIBLE for both parts to have failed at so close to the same instant in time that the air pressure had not dropped to the point it could not occur.... and that is the story we have to work on....

While I understand the "sleuthing" of trying to dissect the video, the layout of the building and furniture, etc.etc.... I still think it only valid if you don't believe the basic statement that the HiPac departed the gun before rocketing off to end in the ceiling.... If it departed the gun during the event, and not as a result of a secondary collision (which they say didn't occur), then its trajectory after the fact doesn't matter....

Bob

This is where I agree with Bob. I believe that the harmonic fractured the o-ring groove. If you look at my post #274, you will a good visual of the harmonic in the pressure tube.

If one were to think of the pressure tube as a shotgun barrel there is a heavy slug being sent down the bore and an opposite reaction is travelling the other direction. If you look at my post #274, you can see a visual of almost instantaneous reaction leaving the rest of the pressure tube in the user's hand.

I am not sure how far the second component traveled while the slug was in the barrel but as far as the mind is concerned to the user, it was in the same time frame.

Chris 

[ChrisK]

While I am doing visual reconstruction why does a spring operated pneumatic rifle kills an optic scope that is not rated? There are two recoil functions acting, a reverse pulse and the a forward pulse of energy. To the operator, it is all in the same time frame.

When I shoot my 300 WinMag, the time frame, in my mind, is instantaneous but I know there are vibration occurring in the barrel and the stock. The whole rifle is warping in the sudden release of energy.

In my mind, the recoil and harmonics in the barrel that failed are not that much separated in a same time frame. 

Chris

[rsterne]

Assuming the valve screw failed first, before the valve exited the tube and dumped the pressure, the tube would have recoiled forward, and likely have started to vibrate sideways like a whip as well.... After all, the (single) screw was on the bottom of the tube, so the stresses in the steel tube before release were asymmetric.... If the HiPac was already cracked at the O-ring groove, or even just had a weak spot there, that vibration could easily have broken it and allowed the HiPac to jettison from the gun under almost the same pressure as the valve saw before the screw broke.... The question then remains whether or not the HiPac in question had a thinner wall than designed, or indeed would have failed if the material was thicker?.... Since we don't know which failed first, of course, the reverse sequence of events could also have take place....

Bob

[chekbone]

The collar onmy hipac stops about .10" above the end of the tube. The one in question was dead tight to the tube. In a photo of the parts the collar shows what appear to be tool marks that look like a vise of some sort. I bet he locked in a vise  and screwed the tube on super-snug. The reverse pressure of the tube pushing up under the collar would constantly push the collar away from the tube. Kinda like screwing in a dolt til the head breaks off. Isn't it the reverse force being applied under the head of a bolt/screw what causes it to break or strip the inner threads.

[Gerard]

My understanding now is that the HiPAC broke on the outward side of that 'collar', not the side which would have been stressed by over-tightening it into the tube. That wasn't the impression I got from frame-by-frame viewing of the video, but that's the expert consensus as I see it in the more recent pages of this thread.

[rsterne]

The HiPac in this incident was shortened to clear the (longer) Disco valve and increase the airflow into the valve.... This pressurized the tube, and to seal it, they epoxied the HiPac into the Crosman tube.... Since it was shortened, I would imagine it could be screwed in all the way to the shoulder without requiring extra force, and the epoxy, once hard, would seal it.... They may or may not have used a vice to tighten it in, and that may or may not have stressed the area outboard of the collar and below the O-ring groove.... However, even if the HiPac was "overtightened" during the installation, once that force was removed, the O-ring area, which is on the outside of the collar, would no longer be under any loads except by the air pressure contained in it.... or an external force, such as a side force, put on it during the fateful disassembly.... Unless the HiPac failed the first time it was filled after gluing in place (and they said it had been used for a year and a half), then I would suggest that it likely was not overtightened to the point of near failure when installed and glued in place.... Certainly, however, a year and a half of use could have contributed to metal fatigue in the area that failed.... If the safety margin in that area was over 2:1 to yield (above the stress caused by stress risers of corners, etc.) then fatigue should not have been an issue....

Bob

[ChrisK]

Assuming the valve screw failed first, before the valve exited the tube and dumped the pressure, the tube would have recoiled forward, and likely have started to vibrate sideways like a whip as well.... After all, the (single) screw was on the bottom of the tube, so the stresses in the steel tube before release were asymmetric.... If the HiPac was already cracked at the O-ring groove, or even just had a weak spot there, that vibration could easily have broken it and allowed the HiPac to jettison from the gun under almost the same pressure as the valve saw before the screw broke.... The question then remains whether or not the HiPac in question had a thinner wall than designed, or indeed would have failed if the material was thicker?.... Since we don't know which failed first, of course, the reverse sequence of events could also have take place....

Bob

Bob,

I agree with your assumptions

One may never know, this is where the math guys come in.

Even with a thinner sidewall for the o-ring, the tube may have never failed under {normal usage}, not the designer's fault

But with the sudden collapse of the pressure, there was a lot of energy moving through the system.

So assuming the screw was release first as the trigger, the ping with the slug would have traveled to the HiPac unit. Think of a string getting longer with would produce lower vibration and harmonics.

To me, this an interesting subject for future designer of systems for failure analysis.

Let us take the 1970s with no video. I think the users of this forum have put to much into the video. The video is to slow for failure analysis. I have not watched the video and do not want to. In the 1970s failure analysis could be taken to a part failure.

But I think it is a good idea to let future people that want to move into high pressure air or above to understand what they are getting into.

This not like tying your shoe laces and people need to think of the outcome.

Thanks to you, Walter and Al plus a name that I do not remember for moving this forward for all to understand to keep their fingers,

Thank You!
Chris

[Tater]

The video is to slow for failure analysis. I have not watched the video and do not want to.

For what it's worth, if you're squeamish about watching it the video does not show the actual incident. It shows what was going on before it happened, and then guys leg with the injury right after it (at least the versions I saw). It also shows some of the parts.

If not squeamish, please disregard this.

[rsterne]

IMO, the designer and manufacturer of HPA parts sold to the public have a responsibility, not only to insure that their product is safe when used as intended when sold, but continues to be so for the reasonable life of the product, however many cycles that might be.... In addition, again IMO, they should to look at common incidents that can happen during use, such as people tripping and falling while hunting, a gun getting knocked over or falling off a table, etc.etc. and make sure that their product can stand not only constant use, but the occasional misuse as well.... That could well include having a sufficient safety margin to stand at least a single accidental overfill with the supply systems commonly in use today, which includes filling directly from SCBA tanks adapted for Airgun use that may contain 4500 psi.... I always make sure that all my PCPs have a sufficient safety margin to allow for such an event, because anyone can make a mistake, be it a slip of the hand on the control valve, or a moments inattention or distraction.... If a gun is intended for use at 2000 psi, like a Disco, and has a 3:1 safety margin to yield, that means it could stand a 4500 psi fill, once, without incident.... It also means that since in general a safety margin over 2:1, for steel, insures that metal fatigue is not an issue, that the service life of the product should be high.... In addition, it should allow for stressing the tube, such as would occur in a fall, at least to some extent, without it rupturing.... Lastly, it even allows for a slight amount of corrosion thinning the wall of the tube without it failing in normal use.... Now, stack all those items together, an old gun, some corrosion, many cycles, the occasional overfill, and then falling on the gun with such force to bend it, and it still might go bang.... but at that point it would be pretty hard to blame the manufacturer, wouldn't it?.... On the other hand, if the designed in safety margin was low, and then during manufacturing the designed dimension were not followed, or a cheaper material was substituted, the amount of liability of the manufacturer would surely increase.... Please note, I am not referring to any specific manufacturer or product with these comments.... but with the recent failures of HPA products from different manufacturers it does raise a talking point if nothing else....

Let me ask you all this.... If you bought a car, used it for 10 years, and then found that the fuel tank was improperly designed, manufactured, or installed and a loved one died in a resulting explosion and fire.... after an accident that was their own fault.... who would you blame for their death?....

Bob

[chekbone]

Let me ask you all this.... If you bought a car, used it for 10 years, and then found that the fuel tank was improperly designed, manufactured, or installed and a loved one died in a resulting explosion and fire.... after an accident that was their own fault.... who would you blame for their death?....

Bob

That's pretty much the GM issue with the ignition switch. For this that's more like an orange in an apple tree, since improper design, and knowing about hasn't been established. If you screwed with the fuel tank - which you shouldn't have - and your car blows up, possibly as a result or not, and there hadn't been any prior reports of problems, it would be safe to assume that screwing with the fuel tank caused the issue. JMHO. I agree that he should have it tested to be sure. If it fails and he doesn't let us know, then it'll be the situation you're bringing up. I don't think he would sell them if he felt there was a problem.

[GoneShootn]

Yeah, I had seen that. Makes you think and also wonder a bit about that bomb that you routinely and often put up to your face.

[lloyd-ss]

Introduction
I am coming in on the tail end of this discussion and only want to add my observations about the actual failure, and my thoughts on the  sequence of the failure events, supported by an  incident and  some testing that I did long before this particular failure occurred.   If I re-state things that have already been said, it is unintentional, and I apologize in advance.  My goal is to present a concise analysis of what I think happened, in the hope that it might improve the overall safety of the airgun community.

[Disclaimer- I am not an authority or expert in failure analysis, nor am I a PE, but I am a retired manufacturing engineer who has done a lot of mathematical analysis of airgun components, particularly PCPs, and also designed and performed some controlled testing-to-failure experiments of some  critical PCP components.  The information I present is for information only and is not to be used for design or engineering purposes. ]

First of all, I want to thank the gentlemen who posted the original videos for sharing the incident as  they did in the interest of safety for ALL of us.

The Failures
There were two  major failures in this “sequence of failures,” (1) the launching of the valve out of the back of the tube, and (2) the breaking off of the HiPac adapter and front tube, along with other failures leading up to, and in between these two major failures. 

Sequence of Events
1) During one of the most recent air fill-ups of the gun in question, I believe that the single 8-32 screw securing the Disco valve in the tube started to bend and shear off.  The valve started to move toward the back of the tube, but was stopped when the groove in its O.D. bumped into the stock/trigger screw that protruded inside the tube.
2) At this point, the valve was being retained by a partially failed 8-32 pan head screw, and the single trigger/stock screw. Because the movement in the valve had been so slight, the failure was not obvious.
3) Disassembly started. The bands and the barrel/breech assy were removed, along with the rear cap and hammer. No problems at this point.
4) The gun was turned over and held, in some fashion, and the trigger/stock screw was unscrewed. At this point, the single 8-32 valve retention screw finished shearing off, and the valve was launched out the rear of the tube. Because the valve was snagging and shearing, it did not leave the rear of the tube in a smooth, coaxial fashion. Instead, its movement was irregular and imparted both axial and lateral reactions in the remaining, hand-held,  tube and HiPac assy. Assuming only 1,500 psi of air in the gun at this time, the valve was launched at close to 400fps with over 200fpe.
5)  The reaction of the remaining hand-held tube and HiPac assy was also the same 200fpe, but much slower because of the heavier weight (equal and opposite reaction), maybe about 60mph (changing units to help us visualize).
6) The tube and HiPac assy flies out of the hand, but because of the way it was gripped, and the offset nature of the  force from the valve, the tube doesn’t go exactly straight but instead some of the energy starts to flip the tube like a boomerang. Visualize muzzle flip in a pistol, or slinging a stick across the yard.
7)  The moving and spinning 22xx tube and HiPac adapter and HiPac tube impacts something, maybe a table, or wall or appliance.  Maybe it has only spun an eight of a turn, but its impact is angular, and no matter how slight the angle, the HiPac adapter snaps at its weak point where the o-ring groove is. Think of holding the back end of the tube and whacking it at 60mph against the surface of a table. A solid tube would stand up fine, but there is a weak point in this assy.
8] Finally, the three parts: the valve, the 22xxr tube, and the front HiPac tube,  all end up in different places.

Backup Documentation

Valve Screw Failure
In the normal 22xx valve installation, the small snout of a CO2 cartridge seals against the front of the valve.  There is only a 1/8” dia of pressurized interface between the valve and the cartridge.  Even at 2000 psi, that is only 25 pounds of force trying to force the valve out the back of the tube and shear the single valve attachment screw off.  That screw has at least a 20 to 1 safety factor in this OEM application.

But this installation now had a Discovery valve installed.  Therefore, the entire ¾” dia front face of the valve was seeing full pressure.  At just 1500 psi, that would be  663 pounds of shearing force on that single 8-32 screw.  That is stressing that screw in the area of a 1 to 1 safety factor, in other words, absolutely NO safety factor. So when that screw was repeatedly loaded and unloaded by filling and shooting, it weakened and eventually failed. This part of the failure is obvious.

For reference, here are two videos  testing the shear strength of the Discovery valve screws. I made them in August of 2014, and they are directly applicable to this failure.  Please note that the video is testing the valve with 3 screws, but the failed valve only used a single screw. 
The two videos are the same, but the shorter version has some of the engineering details edited out. 

Screw failure video - short version       

Screw failure video, long version           


HiPac Failure at the O-Ring groove
First, a little background regarding my very limited experience with HiPacs. In March of 2105, I received 2 HiPac adapters and tubes, and a 22xx tube and valve from a customer who wanted me to make aluminum front tubes that would fit the HiPac adapter.  I agreed to make the tubes using 2024T3 aircraft aluminum.  I did the relevant engineering strength calculations and 2024T3 aluminum tube would work safely with a 3 to 1 safety margin, including in the threaded areas.  After the aluminum tubes were completed, I screwed them onto the adapters and the 22xx tube and noticed that one assembly was not straight.  I wanted to see if a little side pressure would straighten the assembly.  I chucked the 22xx tube into the lathe, gripping it such that the exposed land of the HiPac adapter was just outside the lathe chuck.  The lathe was stationary, and not turning.  I applied some hand pressure at the tip of the tube to see if I could fix the alignment.  I pushed it, and it sprang back. I pushed a little harder and it sprang back.  Finally, I pushed with maybe 35 pounds of side force at the tip of the tube and to my surprise, the HiPac adapter snapped off at the o-ring groove.  I almost never break things by accidentally overpowering them, so this was a real shock.  Because this was a customer’s part, I took pictures for documentation. Glad I did!

I broke the HiPac with 35 pounds of sideload force at about a 16” radius (the length of the new tube).  That is about 47 ft-lbs of bending torque at that joint. Therefore, when the tube and HiPac assy launched at 60mph in the accident, striking a solid object with a glancing blow, that o-ring groove would have seen many times more than that 47 ft-lbs of bending torque that I had applied.  That is why I think this secondary impact snapped off the front end of the HiPac. Of course, everything happened so fast that the two witnesses don’t really know what happened to the tubes. 

For reference, here are pictures that I took in March and April 2015 of the HiPacs I worked with.



 
Conclusion
Based on what I saw in the videos about the accident, and what I read in the posts, and from own personal experiences and testing, I contend that my description of what happened is probably correct.

Lloyd Sikes

[nervoustrigger]

Awesome, very well explained, Lloyd!  Thank you for taking the time to detail it all out for us. 

[rsterne]

Lloyd, thank you for sharing your experience with your customer's HiPac, and confirming that in some cases, at least, the depth of material under the O-ring groove is so thin as to create a structural weakness, especially when that area is subjected not only to a leverage force (as you did), but also pressure (as in the gun in the Video).... I think your conclusions about the sequence of events are spot-on.... The valve came out the back first, and the resulting torgue (collision or just him holding onto it) broke the HiPac through the O-ring groove.... causing it to rocket across the room and imbed in the ceiling....

Bob

[lloyd-ss]

Thanks for the comments gents. This is a very serious topic and certainly deserves everyone's attention.  Glad that I can offer something to the discussion. 
I think we all take safety, that we are aware of, seriously.  But sometimes we honestly don't know what is a potential problem, and what is not.  If you have the least little bit of doubt, please, please, ask.  The answer is right here among our members.

Considering the hundreds of thousands of pellets that are shot each year, I bet that airgunning has a pretty splendid record.  That said, there is no way in the world I am going to let my wife see that video! That is the scariest wad cutter that I have ever seen!!
Lloyd