Pressure Testing / rupture analysis

Thanks Keith - but we are trying to avoid the use of liquids, especially water due to drying of cylinders and prevention of rusting. Thanks for the comment.

Compressed gas vessels can be potential bombs.

I'll 2nd the previous recommendation for DI water; i'd fill tank w/ water, pressurise w/ your N₂, ... ,drain, rinse w/ anhydrous alcohol, then purge w/ dry air

Have you seen this previous thread?

http://cr4.globalspec.com/thread/25943

Yes, it is an excellent thread.

But it doesn't state the test differences between water or nitrogen.

In regards to the explosive force and containing the blast due to a rupture, if inserted within a tank of water - what is the end result, after the rupture of the cylinder, to the the media within the safe tank? Does it just dampen the vibration sending water somewhere / everywher?.

If we where to use a secured polycarbonate screen, preference towards a cylinder with an open end allowing all pressure to force to atmosphere or have the cylinder with a rupture disk in one end - how thick would the polycarbonate screen have to be?

In essence, we are playing with a 2 litre pop bottle - inserting the steel cylinder within the pop-bottle. Pressure testing the cylinder, upon rupture the pop-bottle controls the blast and sends the excess nitrogen to atmosphere. At 600 psi, with such a short cannister, how much damage would result and can it be controlled?

You could use a bladder to apply pressure to the inside of the cylinder, but if there was a pinhole leak, you would not know it. If there was a larger leak, or a rupture of the cylinder, the bladder would fail and release the energy anyway.

If you do not want rust concerns, use an antifreeze mix, and contain the runoff.

Use mineral base oil, and purge the environment with the nitrogen.

For a slow controlled hydro test, use a grease gun, and fill the cylinder with grease. be sure to bleed the trapped air out before the pressure begins.

Still drawn to nitrogen, hydro completed quickly with no contaminants or questions on how to remove the product thoroughly.

Any suggestions on how to protect testing staff?

Put it inside a piece of heavy wall steel pipe with steel flanges and bolted caps on each end. Calculate the energy content of the cylinder with N2 at 600 psi select the pipe wall thickness to contain the worst case moving fragment of the test piece presumably with the full energy of the compressed gas converted into kinetic energy. BTW anchor the test chamber to a substantial foundation, perhaps a heavy weighted table bolted to the floor.

Consider whether to make the test cylinder pressure tight or fully vented (such as with many small holes drilled in the end caps.)

Another important consideration is how you will detect and qualify a text failure. If the detection is visual and your calculations show that a polycarbonate window won't work then miniature cctv cameras inside may be needed. (this is common and cheap hardware)

So you may have some engineering calculations to do if you like this approach. Since it's a serious safety related issue you may do well to hire a qualified licensed PE to do the calculations and chamber structural design for you once you've determined all the other design features that are not affected by the safety issue. It would be money well spent.

Ed Weldon

Thanks Ed, How much energy would be exerted on a bursting cylinder? Ideally we want a portable chamber - either permanently mounted on a truck or with a length of flexible hose secured to the test item isolated in a cage.

Quoting Foursome: "How much energy would be exerted on a bursting cylinder?"

Foursome -- It's been 20 years since I did these calcs.; so I'd have to spend and hour or two reviewing the subject in old reference books. Besides if I do this calculation for you I could be named in a future lawsuit. I don't have liability insurance for that.

This is serious stuff. If your business and even your personal situation has any real monetary value you need to have your ducks in a row and your posterior well covered in this matter.

If you plan to stick with N2 or any other gas get competent outside technical as well as legal help on this one. You could even find yourself facing criminal charges if an explosion on that truck of yours injures an innocent bystander when it's away from your facility.

Ed Weldon

The difference would be the weight of the water which could cause a break in weak welds that using nitrogen or even pulling a vaccum would miss. Weak welds on tanks can be dangerous and weight and pressure are different from just pressure alone.

You should be certain of what you are testing for. Rupture analysis is not the same as yield strength, and neither of them is the same as leak testing. Safety testing for fire extinguishers and scuba tanks involves hydro-testing. In this test, the tank is normally pressurized while submerged in a tank of water. After releasing the pressure, the change in displaced water indicates yielding that occurred in the tank. This test is safer than a rupture test and provides important structural information. The pressurizing fluid can be a fluorinert like 3M-FC77. It evaporates, leaves no trace, and is non-corrosive.

Trying to answer the basic question,

A metal construction of any shape will react differently to gradually increased pressure applied by the compressing gas as opposed to the rapidly non gradual (almost) increasing pressure of non compressible liquid. If you cause the pressure gradient to be exactly the same in both cases, their shouldn't be any practical differences between the two. I personally prefer the gas test, as if there is a flaw, the rapture maybe be less violent, as you would be able to stop the test earlier by detecting the smallest leaks (if you are not on a hurry...) Look at it through a video camera, listen to it with ultrasonics detector. You will always see more and hear more than with the naked eye or ear. And the best, you can document it...

But did I get it all wrong is this a destructive test?. if so my 5 cents, not worthing even that...

Hope it helps.

Wangito.

You are correct, in that by using a gas, you can detect smaller leaks. This is because of the lower viscosity of gas relative to a fluid. Gas, however is compressible, and in the event of a sudden rupture you will have created a bomb. Fluid is relatively uncompressible, so when a rupture occurrs, almost no energy is released.

Not a destructive test, a presure test equivalent to a hydrotest.

In a controlled portable location.

Good morning foursome. The difference is: hydrotest - preferred. N2 test - blow your head off.

foursome,

I don't know your location, but if you are close to San Antoino, Texas.

Check out Southwest Research Institute, they have several test chambers set up for doing nitrogen or water pressure testing up to 15,000 psi and higher. Just go to their website and look for pressure testing. Its been a few years since I've used them so I personally don't know who to call or talk to. Their not to too expensive to use.

4GSG

If you are determined to use nitrogen, and on a vehicle mounted design, consider this. Self contained breathing apparatus is routinely filled to higher pressures on vehicle mounted fill stations. These fill stations are constructed of thick wall steel, with locking devices on the cylinder entry door to contain explosions. Do a search for air compressors and accessories used by Fire departments.

What is your criteria for fail, or pass? How will you know what goes on inside that closed fill station? You will only know if there is a catastrophic failure.

Similar to a hydrostatic testing with water, slow pressure rise, hold for 30 seconds and release.

All cylinders have been tested prior our receipt, we just recertify them 12 years after their initial manufacture.

All cyclinders are pre-inspected prior hydrotesting. Any signs of cylinder abuse, scarring of cylinder, rusting or other - automatic fail.

The only downside to utilzing water internally is drying the cylinder thoroughly, quickly, and efficiently. If you can dry a cylinder quickly under 5 minutes - I would be very interested. Just note that the cylinder can not reach 150 degree C.

We already utilize this approach with a heated airtorch / industrial hairdryer - but due to the rapid heat, slow cool down - majority of the cylinders loose pressure. When opened for inspection, they pass, refilled, they pass. Have to shorten the cycle.

If we where to utilize N2 in a controlled environment, fill tank slowly, watch for leakage, raise to required hydrostatic pressure cylinder rating - hold for 30 seconds then gradual / slow release.

Any thoughts on improvement?

Look at a compressed air fill station. Substitute nitrogen for the compressed air cylinders in the fill bank. Use electrically switched air valves to fill, and dump the nitrogen. Monitor with accurate digital pressure gauges. If you add an event recorder to the system, you could record date, time, pressure and cylinder numbering for record keeping. You could test at least two of your cylinders in each fragment containment section of the fill station. Most fill stations fil 2 bottles at a time. How many cylinders do you need to test at a time?

Just one.

I would rather have a chamber that can be easiliy operated, small and portable. In testing over 5,000 cylinders to date, I have yet to fail one due to hydrotesting.

In regards to the compressed air fill station, do you know of a website or a link of similar system?

If a rupture was to occur, is water still the best medium for the blast? Reviewing our phsyics / mechanics - if you have a blast, the reaction to the water ruptures a small rupture disk (4") on the side of the chamber. Is there anything else that doesn't consist of a liquid that could be utlized to abosb the blast, muffle the sound?

Reviewing the YouTube videos online, there is at times a considerable blast. As mentioned before. The cylinders are tested during manufacturing, most are quite sound by the time we receive them. If they are not damaged - they usually pass our hydrostatic tests.

In high pressure cylinder testing the use of water is used internally, and the amount of water within the test chamber is also used to idenify the amount of expansion with the use of berrets and / or manometer like equipment.

As we all know, glass isn't all that reliable on the back of a truck - any suggestions on how we can monitor the amount of discharged air from the test chamber? Is there an electronic measurement tool that can identify the amount of expansion, or better yet a wrap around the cylinder that can measure the amount of expansion?

I have found the following compressed air breathing apparatus vendors.

• American Airworks
• Eagle Compressors
• Foster Rescue Products
• Haskel International
• Hypress Equipment
• MAX-Air Compressors
• Onspot of North America
• Scott Health & Safety
• Trace Analytics

If water is a concern to your operation, Perhaps you should eliminate it entirely. The water filled tank filling chambers were better at absorbing the heat from the filled tanks than they were at containing the fragments of a failed cylinder. This next idea is beyond my personnel experience, but here goes anyway.Current construction cranes have load sensors built into them to determine safe lifting practices. If your cylinder was set into a frame that fit tightly around it, with load sensors against your cylinder, and strapped tightly in place, Any expansion beyond a prescribed amount would show on the load sensors, and be recorded. This frame could be set into the fill station. This would allow you to accurately measure ther deflection of the pressurized cylinder, as well as containing any fragments if there was a catastrophic failure.

Check out the vendors above, and remember that these fill stations are designed for rapid filling of larger cylinders than you are testing, and are filled to 2250PSI to 5000PSI repeatedly. These fill stations are very safe, and not large at all is a single cylinder is being serviced. Take a look and let me know.