This Is What Can Happen.....

[Gerard]

Alex did not hydro test anything, Al (Gippetto) did. Alex may try to take credit for it with grammar but all he did was supply a unit for testing.

[cilami]

Alex did not hydro test anything, Al (Gippetto) did. Alex may try to take credit for it with grammar but all he did was supply a unit for testing.

No problem. In the post I quoted does not say who actually did the test anyway. The important thing is we can trust the results, no?

[Gerard]

Of course I trust Al's results. Problem is, as we're now seeing, there seems to be a fairly wide range of wall thicknesses involved throughout the production of the HiPAC, as yet unspecified in terms of individual units being somewhat random. The unit sent to Al for testing would have bearing only on units with that particular set of dimensions, not applying to any with substantially thinner walls at any point in the tubes.

[rsterne]

If the method used PROVES that the unit is safe, not only when tested, but will remain safe for thousands of fills, then yes, if your unit equals or exceeds the dimensions of the one that is tested, in the critical areas, then if should be as safe as can be assured by the testing of one unit.... There can always be flaws in materials, which are not only beyond your control, but to a large extent beyond the control of the manufacturer.... This is exactly why we use such high safety factors when designing and building PCPs, with 3:1 or greater being typical from a design point of view.... Personally, I would prefer to have a pressure vessel designed (and built) for a 3:1 safety margin, in the worst case scenario, than use one that has been Hydrotested to the typical 5/3 of the working pressure.... In other words, if it is designed to operate at 3000 psi, it would normally be Hydrotested to 5000.... If it was designed with a 3:1 safety margin, it would be designed not to fail at 9000 psi....

One of the problems with stressing metal repeatedly is that every cycle (eg. filling and using a pressure reservoir) makes the metal move a tiny bit, even though you can't measure it.... Metals have a Tensile Strength (the point where they fail) and a Yield Strength (the point where they stretch and don't recover fully).... When steel is stressed beyond 50% of it's yield strength, it loses strength as the number of cycles increases, by a process called "metal fatigue".... That is why having at least a 2:1 safety margin above the yield strength for steel is necessary to insure long cycle life.... If there are notches or sharp corners, such as threads, or the corner of an O-ring groove, you get what is called a "stress riser", where the localized stress is higher than the overall stress.... In such a situation, although the overall safety margin of the part might be 3:1, the stress in that corner could still be more than 50% of the yield strength, and fatigue can still occur.... The greater the overall safety margin of the part, the less likely metal fatigue is to cause an ultimate failure....

The bottom line is, there is no absolute guarantee on any pressure vessel.... Hydrotesting of cylinders on a regular basis, gives you some assurance that for that one cylinder, it was OK when tested.... On tanks larger than 2" in diameter and 2' long, they must, by law, be tested on a regular schedule, or they cannot legally be filled.... To apply the same logic to a series of parts, where one part was tested when new, or near new, may or may not be good enough.... It depends on the quality of material of each individual part, how much of a safety margin was designed into it, quality control to insure that each part meets or exceeds the one that was tested.... and ultimately a bit of luck....

RE the Hydrotest that Al did, since he now has personally a HiPac that has a thinner wall than the one he tested, he has withdrawn the results of his Hydrotest, in this very thread, in post #208....

in the face of changed specs...I've no choice but to consider my hydro test as no longer valid.

Bob

[cilami]

Thanks Bob!

[Gippeto]

Had no choice but to declare the test no longer valid Bob. Specs were changed from what I tested.

Al

[rsterne]

Agree 100% Al, I was just passing that information on, as cilami apparently missed that post....

Bob

[chekbone]

Had no choice but to declare the test no longer valid Bob. Specs were changed from what I tested.

Al

If 2012 was when it was tested, especially if some of the specifications have changed, isn't it time for another? So the .050 i have is good?

And yes, caliper. Hope i don't explode for making that mistake.

[rsterne]

There has been something bothering me about the simultaneous failure of the (single) valve screw and the thin part of the HiPac where the O-ring sits.... While reading Walter's comment on another Forum, that the two happened at exactly the same moment, I starting thinking about that.... We have be presuming that the valve screw failed first, sending the valve one way (into his leg) and the gun the other, rocketing across the room to collide with something and break off the HiPac at the weakest point.... There is only one problem with that theory, and that is that by the time the HiPac had travelled any distance at all, say it's own length, most of the pressure would have dissipated.... Think of the air staying still, and expanding, sending the HiPac rocketing across the room.... wouldn't most of the air not accelerate with the HiPac?.... after all, the small end had been cut off.... If that is the case, by the time it hit anything how could it have much pressure left inside it?.... Did it hit with enough force to break the wall of the HiPac from "whiplash" for lack of a better term?.... We don't even know, for sure, if the gun even left his hands.... Determining that would answer a LOT of questions....

How about this for a wayyyyyyyyyyyyy out scenario.... Sound, ie shock waves, travels through steel at 20,000 fps.... For two items only inches apart, the time it would take to move from one point of failure to the other would be maybe 20-30 microseconds (millionths of a second).... ie virtually instantaneous.... When one part failed, could it create a shockwave, like a mini-earthquake (a steel-quake) that was powerful enough to cause a part that was stressed almost to the point of failure (or already cracked) to fail instantly?.... Perhaps the frequency generated by one failure was just the right frequency to cause harmonic resonance in the other part, and it shattered along an already present crack, just like a piece of glass....

There is no way, of course, to test this idea.... but it MAY explain the seemingly simultaneous failure of the two parts.... All the other theories, where one part fails first, have the same problem.... in that the time lag between the two failures allows the air pressure required to cause the second failure to decrease, lessening the chance of it occurring at all.... IF this is indeed what occurred, I would point out that, IMO, which came first, the chicken or the egg, is irrelevant, as both parts would not have failed if they had not been so highly (over)stressed, or already cracked....

Bob

[TKO22]

You NEVER test anything to destruction using air pressure.... you use a fluid that doesn't compress and store energy.... Hence the term "Hydrostatic" testing.... Alternately, for testing things like the shear strength of screws, you can use direct mechanical force, like a hydraulic ram, as used by Lloyd to test the Disco screws....

Bob

Bob,

We always test to destruction (TTD) using the air/gas the unit was designed for...

Hydro to Destruction (HTD) shows failure of components, but ATD or GTD will show cascading failures and the collateral damage that may occur.

All products our client manufacture are tested to destruction in the application and gas in which it will be used ... natural gas tank packs on public transit buses, trunk packs for cars & taxis, accumulators used to stabilize off shore drilling platforms, satellite maneuvering supply tanks, etc.

When the engineers are all done with their engineering and math, the HTD almost always meets their expectations. But then show them the slo-mo video of a ATD or GTD test installed in the final configuration, and the engineers start thinking about engineering the entire assembly for worse case scenarios: containment fixtures, fill rate throttles, angles of fittings, routing of supply lines, etc. Impact directions, puncture angles, etc. Even the location of fittings/lines or orientation of the vessel may change based upon the results of a ATD/GTD... that you would never discover with a HTD.

How does this relate to airgun pressure vessels ...

-- The Air Force air rifle pressure vessels used as butt stock/cheek rests.
-- Airgun valves/hammers high and in line with head/face.
-- Lack of inlet fill throttles in most HPA rifles.

Just as sudden destruction of a vessel using air/gas will show collateral damage, sudden filling will also show failures where a sudden hydro fill wouldn't. A very sudden steel lined vessel fill to 6K air/nitrogen with even a trace of a volatile inside will show pressure spikes above the fill pressure. The same pressure test with hydro won't. That same air/gas test can produce internal heating that would and could deform and melt typical buena and urethane o-rings

2 more points ...

1. Fill/empty cycles is more important than calendar life of a vessel. The reason vessels of  certain size require re-hydro periodically is based on the "guess" how many cycles a typical vessel in that application would be filled in a certain timeframe. Scuba tanks are "generally" filled/emptied 2 times a day during diving season. Testing showed failure rates increase after the time/cycle expired. A scuba tank that was filled once with dry air, correctly stored for 100 years would be as safe as a new similar tank, but still not legal to fill.
The case of the Anschutz air rifle tubes bursting (and subsequently recalled) was discovered to be "caused" by national shooting teams that were filling their cylinders 4-5 times per day, every day. The duty cycle was reached in short order and a few of them popped.

2. Lastly, hidden mechanical damage can occur to any mechanism. Drop a 2240 with a hi-pac sticking out and it wouldn't be difficult to imagine the tube could fracture at the joint. Now add the fact of many fills of that assembly (50-60 fills ...) and the work hardening of the materials that occur with expanding and rebounding vessels.
One of our main tests of vessel systems is to use a pendulum hammer to knock the fittings from fully filled tanks. Some of the craziest results I have ever seen have  occurred and only visible when the slo-mo video is analyzed.

The only "accident" related to airgun pressure vessels I've witnessed occurred during a fun shoot, when a fellow was airing up his Discovery from a 3K Benjamin bottle. I heard him complaining that the foster fitting was stuck on the gun, and before I could tell him to check that the tank valve was closed and vent the hose, he yanked the foster fitting off with pliers. The 3 ft. microbore hose turned into a slithering, whirling hissing whip and slapped him across the forearm and chest a few times, leaving some serious marks. We couldn't go near that tank til it had bled most of the way down. He tried to corral that hose with a board, but the hose wasn't cooperating. The danger isn't always obvious.

Stay safe and take care!

MIke T
TKO

[TKO22]

Bb,

As I mentioned in #2 above ...

The craziest things occur when observing ATD/GTD in slo-mo.

We saw fractures way down the line from the primary failure point. With no reasonable explanation.

[rsterne]

Mike, thanks for the heads-up regarding industry testing of tanks, etc. to destruction with the working gas, I was not aware that was common practice.... Still, I think my warning to a neophyte about not attempting it on an airgun is valid....  I look forward to your comments when you read the post I was writing while you were.... Ever heard of the idea of a "steel-quake" ?.... Could that explain your "way down the line" failures?....

BTW, I am well aware that fill cycles are far more important than the age of a vessel.... http://www.gatewaytoairguns.org/GTA/index.php?topic=97441.0

Bob

[rsterne]

I have achieved a dialogue on You-Tube with the poster of the second video (AirgunsWorld) RE the incident.... Here are the results so far....

Bob Sterne   13 hours ago   
Randy, I sure you are aware that many airgunners are trying to figure out just what happened.  Could you please answer 2 questions for us.1. When the gun exploded, was it ripped from your hands and thrown across the room?2. If that is the case, where did the gun (tube) end up, and did it hit anything on the way there?Thanks.... Bob

Reply  ·   
AirgunsWorld   11 hours ago   
+Bob Sterne - The HiPac tube went into the ceiling as seen in the first video (), the valve went into Randy's leg and the gun itself just hit the ground below Randy's feet.

Reply  · 
Bob Sterne   1 second ago 
+AirgunsWorld - Thank you for the clarification.  Do you feel that the HiPac, while still connected to the gun, had the opportunity to collide or impact with anything, which might have added to the chance of it breaking through the O-ring groove?.... Bob

It would appear from his answer that the HiPac and valve both exited the gun, which ended up at Randy's feet.... I am hoping my followup question will clarify that completely.... We may well be dealing with a simultaneous failure here, and not a secondary impact....

Bob

[chekbone]

I'm headed into the woods next weekend to do some rabbit punching with my 2240 hipacker. I'm taking my .45 ruger just in case a bunny runs under a bush and come running out seconds later morphed into a 200 lb hog. I took both to the range today. It been so long since i shot the .45 that i forgot about the Kick-My-A recoil. I was wondering if that could have been what happened with the hipac. First off, you have no idea how many searches i had to do to find a few smart, what i hope are relevant questions. Anyway. Could back pressure have created a shockwave, thereby creating just the right resonance and shocked it apart? Wouldn't there have been less of a chance of vibration if the barrel bands had stayed on? I saw the photo of what was pointed out to be tool marks, assumed from a vise. If the part was somehow deformed in a vise i assume that would be a contributing factor. Seems impossible for them to go at the same time. My new book hasn't arrived so i don't know physics, yet.

[ChrisK]

The only thing that I can add to this discussion is free recoil.

Search for the below subjects

This brings to mind Gun Recoil from S
saami.org/PubResources/GunRecoilFormulae.pdf

Now add barrel harmonics:
varmintal.com/amode.htm

Chris

[rsterne]

Further dialogue with AirgunsWorld in the comments on You-Tube to the 2nd video, a followup to thread #272 above....

AirgunsWorld   1 hour ago  
The HiPac was not bumped, touch and impacted anything when it blew up. We took our second HiPac gun apart and noticed how thin it was at the point of breakage with an unbroken one.

Reply  · 
Bob Sterne   1 second ago   
+AirgunsWorld  - Thanks again for the clarification.... We are then faced with a very strange situation where two failures apparently occurred simultaneously, not one and then the other.... If one part had failed first, then the pressure release would have reduced the likelihood of the second failure....That means we have to look at each failure separately.... Thanks again for your help.... Bob

Since the answer was that the HiPac did not launch into anything until after it departed from the gun.... and the gun landed at Randy's feet after the explosion.... IMO we must consider the two failures as separate incidents and assess them accordingly.... I am not saying that there is not some mechanism, involving a shockwave or vibration that transmitted a force from one failure to the other, acting as a trigger.... However, it would appear that the valve screw failure, which I think we have successfully diagnosed (only one screw where there should have been three), did not launch the gun and HiPac and cause a secondary collision which broke the HiPac off the gun.... We may never know the exact mechanism by which one failure may have triggered the other, or even which came first.... but it would appear prudent to assess the HiPac failure by itself and further investigate the thin tube wall below the O-ring groove at the point of failure.... I hope that Walter weighs in with any new information about the specifications of that area if he has it....

Bob

[chekbone]

This is going to make for brain-farting research. They say that if you can think about it, somebody's thought about it already. The imformation i seek is somewhere on the web. The wall is thin - based on my "caliper", thank you - but isn't it just as likely that the action caused a reaction? I can't say for certain, but feeling myself in that instance the whole gun would have kissed a wall, maybe gotten tossed into the ceiling. If for no other reason than knowing better and still having done something that careless. I'm going to work on mine parallel to my bodyfrom now on.

[rsterne]

A final comment has been added to the dialogue reported in posts 272 & 275 by AirgunsWorld to the Video on You-Tube....

AirgunsWorld   12 hours ago   
After looking over our broken air gun, the owner of the Ashland Air Gun Range with over 30 years of air-gunning, has determined that it was the HiPacs that failed first and the massive rapid change in air hit the air valve like a hammer and then it let go. This is the same as we think, that it was the Hipacs that failed first. I recommend not getting them.

I can neither confirm or deny his conclusion, but IMO it doesn't matter which failed first.... I still don't see how a sudden reduction in air pressure could cause the valve screw to fail (or the HiPac), but there could be some other mechanism (eg. shockwave, "steel-quake", vibration, "whiplash")  that occurred before the pressure dropped too much and there was still enough pressure left to cause the second part to fail.... A cascade effect, in either direction, is still of great concern, IMO, and could indicate that the "as built" safety margins were insufficient in both cases....

Bob

[BigTinBoat]

A final comment has been added to the dialogue reported in posts 272 & 275 by AirgunsWorld to the Video on You-Tube....

AirgunsWorld   12 hours ago   
After looking over our broken air gun, the owner of the Ashland Air Gun Range with over 30 years of air-gunning, has determined that it was the HiPacs that failed first and the massive rapid change in air hit the air valve like a hammer and then it let go. This is the same as we think, that it was the Hipacs that failed first. I recommend not getting them.

I can neither confirm or deny his conclusion, but IMO it doesn't matter which failed first.... I still don't see how a sudden reduction in air pressure could cause the valve screw to fail (or the HiPac), but there could be some other mechanism (eg. shockwave, "steel-quake", vibration, "whiplash")  that occurred before the pressure dropped too much and there was still enough pressure left to cause the second part to fail.... A cascade effect, in either direction, is still of great concern, IMO, and could indicate that the "as built" safety margins were insufficient in both cases....

Bob

This times 10.

I'd sure like the see the area directly under the ceiling where the HiPac entered it. Looking at about 5:32 in the first video it seems to me like he would have had to have been standing several feet off the ground and right on the wall in order for it to have entered the ceiling at that angle, without first hitting something.

Of course maybe you are on to something with the sudden drop in pressure. Maybe the sudden drop in pressure from the Hipac failing actually created a "suction" then pulled the valve out the front of the tube? This would explain how the valve went into his leg face first, right?

IMHO - these guys, and their 30yr Expert have no clue....

[Gerard]

So if I have followed this correctly... first he says in the video that we should never use a 'Hipacs' (guy still can't be bothered to spell nor pronounce the product name correctly), then he comes out with a second video where he states several times that this isn't the 'Hipacs' fault and sort of apologises for suggesting that... then he comes out with a comment under the YouTube video saying he recommends no one buy the HiPAC. Should we wait another few days and watch him flip-flop yet again? Is this the 'final' statement?

His range friend's 30 years of airgunning experience notwithstanding (don't know the guy, no idea how expert he actually might be - lots of folks shoot airguns for decades but very few are what could be called 'expert' where the mechanical aspects are concerned), I'm just not buying any of this. Smacks of building on a lie. The initial, highly emotionally charged reporting, complete with missing video of the actual incident and then waggly hand-held video focused more on the bloody valve than on trying to elucidate the actual events, was all a bad start. These two guys could have tried to mend that with some clear thinking and proper investigation. Instead they've chosen to focus on the 'can't we all get along' follow-up tone with no real specifics as regards the airgun failure itself, and now they're slamming the product.

I have, much to the chagrin of chekbone and one or two others, an obvious distaste for some elements of HiPAC quality. But it's a good idea, which could be executed much better with properly sharpened tooling for inside boring and better consistency in production to make sure the wall thickness was at some safe minimum at the thinnest points. Those two changes would make a vast difference, rendering most forum critiques of the product no longer relevant. Instead Alex has persisted in vaguely defending his product as it stands. And these two guys aren't helping matters by blatantly, it seems to me, lying about what happened. Unless they're truly that bewildered that they imagine a sudden REDUCTION in air pressure could cause a bolt to shear in the opposite direction to the escaping pressure! Physics be darn, eh? Who needs it when we have redneck physics to explain it away! Or magical thinking...

The valve bolt sheared. Bob, you're wasting your good brain on this. They either modified the HiPAC in such a way that it was borderline failing while under 2,000psi, and the flight and impact snapped it in half... or they deliberately snapped it in half after the fact to point the finger of blame at the product rather than their own stupidity in not restraining the valve correctly in spite of HUNDREDS of forum posts spread over a handful of forums for YEARS stating that extra reinforcement is necessary with HPA conversions, whether a HiPAC or other vessel is used. I don't think we're going to get a lot more information out of these guys. They're not interested in sharing information. They just want the spotlight off them.