A bursting disk manufacturer has provided an equation for the calculation of relief forces as witnessed upon rupture. which equates to:

F=2PA

P=Burst disc rating
A=Discharge area


My questions are:

1. Are bursting(rupture) disk event comparible to
relief valve events.
2. should a DLF apply? (Dlf=2 assuming rupture is instantaneous?)
3.are there any rules of thumb regarding imblances of forces as the event travels through a multi elbow system? (ie at which point does the distance between elbows on the discarge become sgnificant.)

This is complex question I can give you some approximate answers....

1)Yes and no the relief valve may simmer or chatter before lifting. When or if this occurs think of a pogo stick. Lift, close, lift, close.... until a final lift occurs. So a relief valve in many ways can be nastier than a rupture disc.


2)Yes

3)Hmmm think of the shock wave as it travels down the system. When it first enters an elbow pair it creates an unbalanced force in that pair. When it gets to the far elbow of that pair the unbalnced force nearly disappears. So the force unbalance travels down the pairs one after another until the end.

This means that A)The longer the distance between the elbow pairs the more time the unbalanced force is applied and therefore the greater the piping system will participate in it.

B)The outlet of the system if it is open to atmosphere will behave in a fashion as discussed in B31.1

Finally do a search on this forum you will a plethora of advice.

On your Q1 (aside from the issues pointed out by Mr. Luff), in terms of the application of CAESAR II, a rupture disk burst causes impulse type loads that are very similar to those caused by relief valves. In reality, the key difference is the speed of application of the unbalanced loads (rupture disks being faster to open than relief valves), but I've yet to find a PRV manufaturer who could tell you how fast their valves open. For that reason, a conservatively-quick opening rate is often presumed; along the lines of a rupture disk. The faster the load is applied, the higher the DLF, to a maximum of 2.

On your Q2, the equation offered by the manufacturer must be considering the suddenly-applied unbalanced pressure force (P*A), multiplied by the conservative DLF value of 2. In many cases, applying this 2PA force in a static load will provide good 'ballpark' results. There are cases though, where presuming DLF of 2 is overly conservative.
If you are not sure, I would suggest developing your own force-time profiles based on your particular circumstances and employing time history analysis.

On your Q3, there are several variables that come into play. If you have a large pressure and/or area, your P*A load can be significant, and in these cases, lesser load durations can still develop a significant system response.

The DLF generator in CAESAR's dynamic input processor will give you an idea of how various natural frequencies will respond to your particular force-time profile. That, in conjunction with the magnitude of the unbalanced pressure load may help you decide when it is necessary to consider some loads and ignore others.