We know burst testing. After over 40 years, our Micropac pumps have been used for lots of burst testing set-ups. Here are a few pointers including the issue of protection.
How dangerous is burst testing? What extra safety considerations are there?
This is a very common question. It can be very dangerous and a risk to life. In the most extreme cases, you are talking ballistic protection. Precautions can be taken so that these risks are reduced down to negligible.
Test the right type of part on the best test medium with planning, foolproof kit and protection and you will be fine. It just becomes a process that gets carried out every day. You still need to Risk Assess what you are doing and take adequate measures to protect yourself and other people plus the workplace.
What is burst testing?
You are pressurising a component using gas or a hydraulic fluid until it fails. The term “burst” can be misleading. Some parts will simply burst. Others will have a particular component that fails, such as a plug blowing out or a fastener failing and causing a massive leak. There are lots of failure modes in the extreme. We are referring to these as generic “burst testing”, but don’t expect a dramatic burst every time.
Why do you burst test? Why do you need to burst test if you have designed the parts properly?
For 40 years, we have manufactured our Micropac pumps for different forms of hydrostatic pressure testing including functional, proof and burst testing. Learn more about different types of test from our explanation on our web site.
A hydrostatic burst test isn’t something you will do on every production unit. It will be done on a prototype component or system then on a sample from a production batch. For example, for a new moulded part, you would burst test and compare to what you expected from your calculations in your design. Checking actual performance to predicted figures is a big thing for Engineers. There is always a risk of a failure mode that you didn’t foresee or indeed a mistake. In production, on a batch of parts, you will be checking for a consistent process. A burst pressure figure will provide a yardstick for comparison with future production.
Some users like the military will insist on testing to failure on a lot of parts. That would include burst testing amongst other tests. For example, if a designer has optimised a cockpit window to cope with a bird strike at 300 knots, a customer would probably want to test to see if that is so in real life and at what point the part actually fails and by what mode. There is also the risk of “over-designing” a part and it being far too substantial and maybe heavy. That isn’t the best use of resources or much help to the environment if over-weight parts have to be moved around for years. Good Engineering is about optimising design for the function.
Burst testing may also be a requirement of any prevailing pressure equipment regulations and design standards, in order to test to destruction and validate the design. That is vital to know.
How do I calculate how much energy I need to burst my component?
That’s a difficult question. Pressure is the key and how an Engineer looks at likely failure modes and at what pressure the component will burst.
Pressure is what bursts your component. You build up potential energy in the test fluid, in our case by pumping with a Micropac hand pump. You will prefill the component, whether it is 5cc or 500 cc volume and bleed out the air. The hand pump is inputting energy to build up the pressure in the test fluid from atmospheric pressure up to a higher pressure at which the component fails.
An Engineering designer will have calculated what pressure the component will work in everyday use, what pressure a production unit should be tested to (proof pressure) and finally what pressure they think the unit will fail at. Watch our explainer. That’s what Engineers do. They would look at all the ways in which the unit might fail and do calculations to check out strength and other parameters even down to whether deflection might open up a clearance large enough for a seal to extrude. They would also consider longer term issues like corrosion that could weaken a structure or creep of plastics that will cause issues. Note a pressure vessel will need to be reviewed in the context of the Pressure Equipment Directive in the EU. Get to a certain volume or pressure and it needs certifying. “Pressure Vessels” as an engineering area are a very specialist area. We’d subcontract it out to somebody like Compressor and Power Engineering. Our world is testing smaller assemblies and components. Testing pressure vessels is not our expertise.
How much pressure to burst my component or test piece?
You will probably have to get a feel for what pressure something will burst at before setting it up. If all the parts are “bought in” with quoted maximum working pressures, you might simply test to twice or three times that pressure and see what results you get. You can be surprised. If you are using some designed and manufactured parts, it can really be quite surprising how high a pressure they can withstand. Sometimes the failure is in a connector or valve that strictly speaking is not part of the test piece. There again, you can get a dramatic failure on a designed and manufactured test piece, which of course is the whole point of the process. You have a calculated burst or failure pressure and you are looking what happens in real life. Sorry, we can’t come up with a magic answer for what pressure a part will burst at.
How can I make burst testing a lot safer?
There are some basics that are very sensible before you start talking ballistics.
Don’t stand watching a part burst. Burst testing isn’t a social occasion for the troops to stand and watch. You probably won’t see what happened anyway because it happens so quickly but are taking a risk. Having people around is a real risk.
Avoid using high pressure gas unless you really have to. That would not be our area of expertise. For example, a spacecraft pressure vessel that will be used for storing gas will often be burst tested on a gas in order to see what actually happens. Will a fragment damage the spacecraft hull? The results can often be very hazardous during testing due to the energy in compressible gas and the mass of parts that are blown off. A large heavy fragment propelled by a large volume of gas can have enormous kinetic energy. Some of the results are truly horrific. If you don’t need to use gas, opt for water or even oil. That has very little stored energy under pressure. We don’t get involved at all in gas and that includes testing on shop air, which isn’t always the cheap and easy option that it sounds like.
Any explosion risk due to an inflammable gas as the medium or traces of vapour such as petrol are a massive hazard. One of our contractors was foolish enough to pressure test a petrol tank that had traces of petrol in it. We would suggest that testing on gas needs professional help in assessing risks.
Even on fluids like water, use whatever physical protection that is adequate to catch any projectiles that are propelled from the test piece as it bursts. That means risk assessment and some attempt to estimate the mode of failure and energy in fragments. Even at a basic level, we would always use a polycarbonate chamber like our MTE enclosure to simply contain fluids spraying out upon burst. That also provides some protection from parts propelled out subject to assessing how much energy is involved. Add in whatever ballistic protection is required on top of what the standard MTE can provide. Our MTE-B provides substantially increased protection.
Look at possible failure modes and risk assess what protection you need.
In a perfect world, you build a test chamber that can resist projectile damage, set up some sort of protective chamber in your shop or test in your shop with nobody there and nothing critical that could be damaged. Burst testing does present real issues and risk of serious injury.
What happens when a part bursts?
This is the million dollar question. Once you know this, you can assess what risk there is of injury. What are the various modes of possible failure?
We have done a lot of burst testing over many years. So, what are common modes of failure in our world? Every business is different, so we can’t comment on parts that we have never used or even seen. We can make some general observations.
Surprisingly, we would say that very often a material in a component yields or distorts and there will be a dramatic leak. The pressurised fluid simply sprays out. The MTE enclosure is a low cost solution to catch that spray. You definitely don’t want oil spraying over a floor as it is so slippery. Even water is a trip hazard.
For sheet metal components or indeed pressed or moulded parts with areas or “diaphragms” that distort under pressure and eventually rupture or shatter, risk can be more difficult to weigh up. A rupture can be spectacular, with a blast of water caught in the MTE enclosure. A glass window that shatters could be more risky if fragments are not caught. Sometimes, thin sheet metal fabrications can be quite surprising. There may be quite spectacular distortion of a ductile material then suddenly it is a weld that is peeled back causing a leak. There may be a rupture of material itself, but we find that welds seem more the cause of failure.
What about hydraulic hoses and fittings? Hose material can burst, although safety factors are high. An end fitting blowing off is more spectacular, as you have a hose that whips and maybe an end fitting that could become a projectile unless it is held captive by pipework. Consider anti-whip wires to reduce risk. Compression fittings will fail at some pressure, normally with a bang and a spray of fluid. The fitting will only become a projectile if it isn’t captive on pipework.
Maybe the big unknown is a component that has a potentially very dangerous mode of failure. Burst testing of a casting? That could suddenly shatter and the fragments could have enough energy to cause damage or hurt somebody. Anything with a single point of failure is very worrying. A hydraulic sealing plug that blows out under pressure. A threaded plug that strips its thread and blows out. Or a single fastener (as opposed to multiple fasteners) holding a pressure assembly together. If the fastener fails due to the load, the pressurised assembly could suddenly blow apart and cause damage. Designers generally don’t like single points of failure.
Those are a few limited notes on modes of failure. One over riding observation is that volume of the part you are testing is crucial as well as the mode of failure. If a part has 5cc of fluid in it (plus the hose from the pump), the stored energy will be one hundredth of the stored energy in a 500cc component. We talk about stored energy later. All we know is that whether the system is 5cc or 500cc, the sudden failure of a plug blowing out is equally dangerous. We would never not protect a small assembly under test. All burst testing is hazardous.
One thread that runs through these observations is that if you can make a possible point of failure captive in some way that is helpful. Fittings and pipework captive to the rest of a system, hoses restrained by anti whip wires or a simple sheet metal plate covering sealing plugs. That’s a thought.
How much energy is in my burst test? How does velocity, mass and ballistics work in burst testing?
You are into the area of ballistic protection, but diving straight into bullets and bombs misses some of the basic figures and risks for YOUR burst testing challenge.
If you research on the web, there is so much information focussed on bullets. UL752 provides guidance on standards for bullet proof materials. These are relatively small mass and propelled by a very rapidly expanding body of high pressure gas. Our burst during a hydrostatic pressure test is powered by a fixed body of fluid that in reality is not very compressible at all. So, comparison with a bullet is only helpful to a point. Maybe it is quite unhelpful. The bullet masses and muzzle velocities are available. Velocity is a key (and crucial) measure and although you can read that an AK47 has a muzzle velocity of 715 m/s and a bullet mass of 7.9g, we don’t think you need to dive in and design protection for this projectile. For our challenge of safe burst testing, we are told that a velocity figure of 300m/s is a more realistic (but rather terrifying) starting point in any calculations. We need to focus in on maybe larger fragments propelled by the energy stored in our test fluid that we have pumped up to pressure. Along the way, we must remember that small mass items like sealing plugs can blow out, but they are propelled by our largely incompressible test fluid and not a rapidly expanding body of gas.
The web will offer ballistic protection, including chambers and panels which will undoubtedly offer a solution at a price. Possible, an enormous price. You could dive in and order panels to a bulletproof specification. Blast protection is another whole area. Firms offer blast protection screens and windows. But is there an alternative? Stay with us.
Maybe stored energy is a starting point. Engineers know that for pressurised fluid, you can’t get out more than you put in. As you pump up your test system using our Micropac pump, you are building up potential energy in the water (or other fluid) in the hose and the component. You have already prefilled the system and bled out the air, Water is hardly compressible, although a number of strokes will be required to build up the pressure as there is some residue of dissolved air in water and the hose itself will deform a little.
We would look at the amount of energy that the operator is putting into that fluid whilst pumping from zero pressure up to test pressure. Note that you will have already pre-filled the hose, test piece and bled out the air at the highest point. You don’t want to be pressurising a slug of air to high pressures due to safety issues. We don’t like pressurised air. Once we have filled the volume at atmospheric pressure, we can look at the energy required to build the pressure up to our burst pressure. That is quite easy and is calculated as force x distance moved by his or her hand that is moving the handle. As a rough calculation, let’s say 5 double strokes of the handle x 300mm movement x a mean load of 200 newtons if the maximum pressure on the pump needs 400 Newtons. If the potential energy is Force x distance moved, that is 590 Joules. If it actually takes ten strokes and not five, then double the potential energy. That’s interesting. We have made the assumption that there is no significant friction or losses as you build pressure. That’s fair. Note the maximum force is the same on our 49cc/100 bar, 25cc/200 bar, 12cc/400 bar and 7cc/700 bar pumps. Five double strokes of the 49cc pump has shifted 250cc at 100 bar. At the other extreme, the 7cc pump has shifted 35cc at 700 bar. The potential energy is the same.
When the part bursts, the potential energy will be converted to the kinetic energy that a moving part (or parts) has imparted to it. In reality, all the 590 joules won’t be converted, but that gives us a feeling for a maximum figure. The web tells us that a bullet has an energy of between 103 and 20,000 joules. An air gun pellet is down at “only” 7.5 joules at which point the weapon is heavily regulated. On the face of it, we are way higher than an air pistol pellet, which in itself is quite a nasty hazard.
That tells us that the energy built up before bursting your test piece is relatively high and worth taking seriously. In a test set-up with a hose, the fluid suffers loses as it flows out of the hose and test piece in a fraction of a second. At least we have a starting point. Serious stuff!
The pressure reached is another consideration. We are suggesting that needing 250cc up to 100 bar or 35cc to 700 bar as the amount of fluid to build up from a prefilled test component to burst pressure require exactly the same energy. Instinctively, you would think that if the failure mode is a plug blowing out like a bullet or a heavier end cap blasting off, this happening at 700 bar would be much more dangerous and demanding for your method of protection. For example, a hydraulic sealing plug is retained by interference as it is expanded out during installation. They are a great product and we must have used 500K over the years Take a 6mm plug and apply a massive overload of 700 bar and you have a force 0f 438lbf or 1954N forcing it out. If that fails, it will come out “like a bullet.” It will certainly come out with a bang and be dangerous. That’s way above the rated pressure but that’s what burst testing is about. Using potential energy at ½ mv2 and a notional velocity of 300m/s with a mass of 1 gramme, that gives us an energy of 45 Joules. That’s dangerous if it happened. Could the velocity reach 300m/s? That seems to be a commonly quoted worst case for projectiles from fluid burst testing. As noted previously, if your possible failure analysis suggests that a (say) 100g heavier machined component will be fired from the test piece upon “burst” following the failure of a single component retaining it, working the kinetic energy of this object at a notional 300m/s gives a kinetic energy of 9,000 Joules. That is way more that our estimated 590 Joules of stored energy, so it can’t travel at 300m/s. You can’t win more energy than you put into the fluid. Remember there is a square in the potential energy formula so the velocity isn’t much less at 108m/s. That is still a major hazard. As noted, the full stored energy won’t be imparted to the projectile, but even a proportion of the energy would be enough to be a serious risk. We think it also brings home the extra risk in testing on high pressure compressible gas. The ultimate nightmare with projectiles is any risk of a rapidly expanding volume of gas behind them. Munitions take this one step further and have burning gas to expand the volume incredibly quickly.
What pressure should I pressure burst test at? How much is too much?
We have already answered the slightly different question of “How much pressure to burst my chamber or test piece?”
The obvious answer the “what pressure should I burst at” question is “enough to complete the test.” Ask the designer but be aware of “higher pressure testing” being quite specialist. So, what do you say if the designer tells you that 2800 bar is their guess?
Review whether you are functional, proof or burst testing. If you are “burst” testing or pressurising a component up until it fails, the designer will have a good idea of figures. He or she will have designed for a working pressure, a proof pressure and maybe your ultimate test to failure will just verify the the safety factor. There is always the risk of a failure mode that is unforeseen. Once it is designed and manufactured, the pressure is what it is. You don’t really have any choice at that point. You simply source test kit and protection that you hope will cover the range of possible failures. Sometimes a part simply won’t fail at the pressures you can provide. Maybe you have over-designed it and should (or could) modify the part. Or maybe the safety factor built in is so high that you simply accept that you have been unable to burst test without running at an even higher pressure.
For our company and product range, testing up to 700 bar is very common. We manufacturer to 1000 bar but that is moving into something more specialist for us. You are into 1000 bar fittings, valves, hoses and sealing in your design. We don’t get involved in higher pressures.
What can Sarum Hydraulics offer for burst testing?
After 40 years, Sarum Hydraulics are the experts on pressure testing. Our Micropac pressure test pumps and reservoirs power test rigs throughout the world. Our hoses, fittings or even our knowhow help our customers design and set up rugged, reliable and good value pressure testing rigs. We can supply drag pointer or electronic pressure gauges to record at what point your test piece fails. If you simply need to contain a fluid burst and achieve a basic level of ballistic protection, our MTE enclosure is a neat and inexpensive solution. If you need enhanced ballistic protection, our MTE-1-B will provide that extra assurance. We have looked to the technology behind larger inspection panels on machine tools. There you might see higher mass components at slower velocities than small mass bullets at high velocity. We have calculated 590Joules of energy at five strokes of our hand pump, double that at ten strokes and so on. Looking at EN DIN 12415 for the results on a 500mm square inspection window, our material will absorb up to 4960J. We think that this assessment of performance is more useful for an application where we know the stored energy and the propellant is incompressible than looking at ammunition where the propellant is gas and velocities much higher. As always, we welcome comments.
What’s the difference between a simple burst fluid protection chamber and a ballistic chamber?
If a component fails during hydraulic pressure testing, invariably the pressurised water or oil will suddenly spray out, releasing the energy. This spray of fluid will represent a release of energy, although being a fluid flow, our standard MTE chamber will withstand the sudden release. The chamber is manufactured with impact resistant polycarbonate and as we are using a fluid like water or oil rather than high pressure gas, the release will be captured within the chamber.
The moment you have to consider that a test component might fail and produce projectiles travelling at speed, you are into ballistic protection. The standard chamber offers a level of protection that will be adequate for the majority of applications but not all. Move to the MTE-1-B chamber and you have additional resistance in line with the EN DIN 12415 standard for inspection windows.
Why are manual hydraulics a good solution for pressure testing?
We would say that the combination of low cost and operator control make manual hydraulics a winner. If you have a volume production process, the cost of a power operated rig with automated control may be justified.
If you need a powered prefill and manual pressurisation up to burst pressure, don’t overlook our DUO pump units. These are widely used for this very purpose. It is very attractive to prefill and build a system up to a certain pressure using the air driven pump then seamlessly carry on to burst pressure with the manual hand pump that is built in.
Do I need to have a calibrated pressure gauge to pressure test? Do I need glycerine filled pressure gauges?
As with any measuring equipment, something that is not calibrated and traceable back to National Standard has limited value in a quality system. You simply don’t know how accurate it is. You would use a calibrated gauge. Excessive shock is the enemy of any borden tube pressure gauge. Our normal pressure gauges used for simple pressure testing are always glycerine filled, as they provide damping of the mechanism if the pressure suddenly drops when a burst or leak takes place. However, the drag pointer gauge has a spindle running through a hole in the front face, so traditionally aren’t glycerine filled. Talk to a specialist like Ashford Instrumentation with your precise requirement.
Do I need a drag pointer gauge?
We would say yes. In theory, you could keep your eye on the needle of a standard pressure gauge and record its maximum position, but it never seems to work that way. A drag pointer gauge has a second pointer which is pulled around by the original pointer and records the maximum figure. The burst happens in an instant and you have lost the moment.
Is a digital pressure gauge more suited to burst testing?
Yes, a digital pressure gauge that features a “maximum pressure reached” option is perfect, but at a price. Hydrotechnik are our go-to for digital pressure gauges, including digital gauges with a data logging feature.
Should I be cycle testing or burst testing?
We have been discussing one specific type of testing in this paper. Burst testing is a key process for pressurised parts. But don’t overlook the destructive effect of large numbers of cycles, even at much lower pressures. In real life, a failure after a certain number of cycles at working pressure can be far more dangerous than a burst test at a greatly elevated pressure. Some components like fuel tanks have to be cycle tested to meet Standards. A need to cycle test to gain assurance on performance applies increasingly with mouldings from reformulated polymers. Pressure vessels of any type from composites definitely need careful attention to the effect of pressure cycling. There are lots of applications for cycle testing as an addition to burst testing.
Our Micropac PTR-A low cost pressure cycling rig is a great starting point. Talk to us about your application.
Can I use a digital camera to watch what is happening during my burst test?
We have already said that you should not stand at risk to yourself watching an event that you won’t even see. The burst will happen in milliseconds, although the deformation that leads to it will normally take longer. A few years ago, if cameras were mentioned, the response would have been that a high speed camera was very exotic and expensive. An iPhone 15 has a maximum frame rate of 60 frames per second which is amazing but maybe limited if you need to see what actually happens. Cameras running to 1000 frames per second are commonly available at what seem reasonable prices. We don’t have any experience on high speed cameras, but send us any interesting burst test videos. Designers and quality people need to know the mode of failure so they can modify something. Maybe the camera isn’t a bad call.
Micropac hydraulic hand pumps have been the tool of choice for pressure testing for over 40 years. We are always pleased to talk about your application.
Leave A Comment