Surge Analysis of Fire Water Piping 4

[shaff]

Theory part:
What are the conditions under which a surge analysis of a piping becomes necessary? What is the basis of surge analysis? What are we checking and trying to achieve by doing surge analysis of a piping section or a piping system?

Coming to the practical case:
There is an existing fire water piping network in the plant. We are installing new equipments as part of the plant expansion and the new equipment also needs fire protection piping with spray nozzles all around. For this we are taking a tapping out of existing fire water header pipe and installing a deluge valve skid and with some butterfly valves, installing the spray nozzles. Valves will be in open position and the line will always be in pressurized condition. Does this line require a surge analysis and why?
Piping Material: GRP
Test Pressure: 27 Barg

We are not installing any pumps or surge vessels and any other pulsating equipment to foresee any surge problems. Does this typically falls under a piping engineer’s scope or a process engineer need to look at this?

Thanks for bearing with my long questions.

 

[LittleInch]

Taking this in parts

Surge analysis (aka water hammer) should be undertaken where you have an effectively incompressible fluid (water in your case) which is subject to rapid start and stop of flow. What constitutes rapid depends on your system and speed of valve closure etc. If the speed of valve closure is faster than the time taken for the pressure wave to go from the valve back to the pressure source and back again a the speed of sound in our fluid, then you have an instantaneous closure, even if closing the valve take a few seconds. The faster the fluid that comes to a sudden halt, the greater the potential surge pressure due to the kinetic energy in the fluid (a v2 term) having to transfer to potential energy in the form of pressure.

what you are checking for is a transient rise in pressure over and above the normal operating pressure and also the deign pressure plus the allowance in the code or material for transient events. Pipelines it is 10%, but piping can be higher. Especially for relatively low rated pressure systems, surge pressure can be orders of magnitude higher than your design pressure. Surge pressure should never exceed the test peruse and usually needs to be lower.

In your case you need to identify how the water flow starts and stops. If there is any chance that someone can isolate the operating section by simply turning a butterfly valve or small ball valve then yes, you need to do a surge analysis in IMHO. Similarly if the pump start or turning on the inlet flow can happen suddenly, especially against a "flat" system, this could also result in transient effects which can literally blow your pipeline apart. If your system is such that you have multiple outlets and can only turn off one at a time then you probably don't need to do an analysis as the total fluid flow gradually reduces over time and there are other places for the pressure wave to go. Only you know how your system is configured and could be operated. It is generally poor practice to rely on operating procedures to reduce surge pressures as you can't rely on people doing what they are supposed to do at the speed they do it, especially in an emergency situation or a relatively rare event.

Usually this is a process led issue with flow assurance / hydraulic analysis determining whether it is an issue which the piping engineer often needs to solve.

GRP / GRE doesn't like transient shocks and repeated transients can severely weaken the pipe without any external visual impact until it suddenly fails.

hope this helps

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[EnergyMix]

Practical case (at least from where I sit): surge analysis is done only after the water event happens and management is wanting you to assure them that they don't have to replace pipes. Much worse in the nuclear industry, where the NRC will second guess your calculations.

Note: all opinions are my own and do not reflect those of my company or my management.

Want to know the do's and don'ts of Eng-Tips? Read FAQ731-376.
English not your native language? Looking for some help in getting your question across to others or understanding their answers? Go to forum1529.

 

[MortenA]

I would say yes if your water supply is a fire water pump that e.g. lifts water from sea level to e.g. a ships deck (FPSO). Usually the pressure will be supplied with a jockey pump. Once the deluge valve opens the presuure will rapidly drop to atmospheric - and below at high point before the main pump can start and lift water from the sea level. Then once the pump catches up the pressure will rapidly increase and you may experience water hammer Especially if you have high point such as fire water monitors at high elevation (above 10 meters with reference to the ring main/deluge skid spray nozzles). Here you may draw a vacuum. If this happens once the pressure rises the vacuum will/can collads similar to collumn separation/collaps. This may cause a very high surge. Also if the monitor is opens, the water will rush towards the nozzle because the monitor valve will have a much higher capacity for air than water. Once the water reaches the nozzle it will slow down very fast due to the much lower capacity for fluid in the valve - this may also cause a water hammer.

 

[RoboCop16]

Shaff,
In my view GRE firewater systems should undergo surge analysis. One consideration is that this is a safety critical system. Another is that, as LittleInch mentions, GRE is not ideal for water hammer and other dynamic loads. On some newly installed platforms operated by a company where I used to work, we regularly had failures in GRE systems due to water hammer or surge. These were pretty nasty failures as well, with the pipe bursting apart at joints without warning, often in association with pressure spikes caused by pump start-up or downstream valve closures. A strongly contributing factor was poor workmanship while making the bonded joints and unfortunately there is no reliable NDE method for identifying poor joints. The joints had passed pressure test but subsequent pressure spikes or vibration can weaken the joints without this being detectable.

You ask whether the surge analysis falls under the Process or the Piping Discipline. It will normally be a cooperation between the two. Process will perform a dynamic pressure calculation using Pipenet or similar software. Maybe the existing system has already undergone such an analysis in which case most of this work is already done, if you can get hold of the calculation. Process will inform the calculated peak pressures and the unbalanced pressures to the Piping Discipline

The piping discipline will determine the whether peak pressures are within the the allowable limits given by the design code (e.g. ISO 14692) and by the manufacturer. Piping should also do stress analysis to ensure that bending stresses from the unbalanced forces created by the pressure wave travelling though the system are within allowable limits. This can be done in a static analysis (similar to what you would do for a slug analysis) with the static forces multiplied by a dynamic load factor.

Whether you chose to do a surge analysis or not, you want to make the GRE system as resilient as you can against dynamic loads by making it well restrained. This means a lot of supports and also as many stops and guides as you can get away while still keeping expansion stresses within allowable limits.

 

[BigInch]

Since firewater systems are constantly under high static pressure, even slight transient pressures may put them over the top. Looking at current practice, including a transient analysis is certainly state of the art, so no engineer should overlook doing a transient analysis as a routine matter of course.

I hate Windowz 8!!!!

 

[shaff]

I have filled the checklist for the surge analysis and have told the GRE pipe manufacture to have a go at Surge Analysis and provide supporting as applicable.
Difficult part still remains - Convincing the Project Manager technically whos always interested in numbers

Thanks to all the insight you have provided.

 

[Ripz]

What's also done, is a Transient Analysis to determine the possibly of formation of vacuum conditions during abnormal operating conditions, eg pump trip. GRP has minimal strength from external pressure, and needs to be protected as such.