9 replies to this topic

[SPOOK]

Dear All,

Kindly refer to the attached excel sheet of a Liquid LPG unloading pipeline at a port. I want to calculate the surge pressure in the described pipeline (58 m) in the emergency scenario when the ROV is shut down.

The required data are also mentioned in the spread sheet. If any other data is required let me know.

There are a couple of methods to calculate the surge pressure that is as under:

METHOD 1:

P = (0.070) V * L / T

where, P = Pressure rise above static pressure. psi
V = Velocity of the fluid flow, ft/sec
L = Length of the pipe, ft
T = closure time of the valve, sec

Hence

P = 0.070 x 9.834 X 190.2887 / 5
= 26.19 psi
= 1.831 kg/cm2

This is for case 1 where the time for valve closure is 5 secs


P = 0.070 X 9.834 X 190.2887 / 10
= 13.095 psi
= 0.915 kg/cm2

This is for case 2 where the time for valve closure is 10 secs

METHOD 2

The results second method was provided by my client who gave a different value of surge. This method used the STONER SOFTWARE. In this method the dependency of the surge pressure on the valve closing time is very negligible. The output of the results is as under

Vavle Closing Time Surge Pressure
(mins) (psi) kg/cm2
0.1 310 21.8
0.2 310 21.8
0.3 310 21.8
0.4 310 21.8
0.5 309 21.72
0.6 308 21.65
0.7 307 21.58
0.8 306 21.51
0.9 305 21.44
1 305 21.44
2 298 20.95
3 293 20.6

I am not trying to compare the software with the hand calculations. At least they should be within agreeable range.

Any one has a some other method or guidelines kindly share with me.

Kindly give me your valuable insights and comments.

Waiting eagerly for reply

Regards
SPOOK

Attached Files

•  surge_pressure_LPg_pipeline.xls   893.5KB   1708 downloads

[ankur2061]

SPOOK,

I am attaching a section from the book 'Pipeline Rules of Thumb Handbook' by E.W. McAllister which describes the method to determine maximum surge pressure in a liquid-filled pipeline when a valve is suddenly closed. Please note that the equation mentioned in the attachment does not take into account the valve closure time. Essentailly it means that the fluid behaviour is considered as steady state and not transient. For transient analysis you are advised to read the section 'Transient Pressure Analysis' of the above mentioned book. I believe that any manual calculations of surge pressure for pipeline transient (time dependent) flow is a herculean task and recognized software such as 'Stoner' and 'Olga' should be used.

To the best of my knowledge 'Stoner' is an industry recognized software and engineering consultants like Bechtel do subscribe to 'Stoner' for steady state and transient surge analysis in pipelines. Stoner however is not recommended for multi-phase fluid transport pipelines. 'Olga' is the preferred software when dealing with multi-phase fluid transport in pipelines.

Hope I have been able provide you some assistance in resolving your query.

Regards,

Attached Files

•  max_surge_pressure_liquid_filled_pipeline_for_valve_closure.doc   20KB   1594 downloads

[SPOOK]

SPOOK,

I am attaching a section from the book 'Pipeline Rules of Thumb Handbook' by E.W. McAllister which describes the method to determine maximum surge pressure in a liquid-filled pipeline when a valve is suddenly closed. Please note that the equation mentioned in the attachment does not take into account the valve closure time. Essentailly it means that the fluid behaviour is considered as steady state and not transient. For transient analysis you are advised to read the section 'Transient Pressure Analysis' of the above mentioned book. I believe that any manual calculations of surge pressure for pipeline transient (time dependent) flow is a herculean task and recognized software such as 'Stoner' and 'Olga' should be used.

To the best of my knowledge 'Stoner' is an industry recognized software and engineering consultants like Bechtel do subscribe to 'Stoner' for steady state and transient surge analysis in pipelines. Stoner however is not recommended for multi-phase fluid transport pipelines. 'Olga' is the preferred software when dealing with multi-phase fluid transport in pipelines.

Hope I have been able provide you some assistance in resolving your query.

Regards,

Ankur

Thank you for your promt repy. Indeed this throws some light on my topic. However, you did not comment on my method. anyways. If you have a soft copy of the foresaid book would you mind sharing it with me?

Appreciated

REgards
spook

[ankur2061]

Thank you for your promt repy. Indeed this throws some light on my topic. However, you did not comment on my method. anyways. If you have a soft copy of the foresaid book would you mind sharing it with me?

Appreciated

REgards
spook

Spook,

I haven't come across the method you have adopted so I can't make any comments.

As regards sharing the book, this is a copyrighted book and it is unethical to share copyrighted stuff.

However, further details of the book are as follows:

Title: Pipeline Rules of Thumb Handbook
Author: E.W. McAllister
Publisher: Gulf Publishing Company, Houston, Texas

Check out the online bookstore of Amazon for the title and you can order it from there.

Regards,
Ankur.

[narendrasony]

Dear Ankur,

Where can we use these equations for pipe-line surge pressure calculations?

I think just before the valve closure the flow may be steady state flow (i.e. the properties like pressure, velocity and temperature are time independent), But the moment the valve is closed suddenly, the flow will no longer remain in steady state. Pressure and velocity will essentially become a function of time and may be oscillating as well. Please correct me if I'm wrong.

So, can we use the steady state equation for surge pressure calculation ? Do you mean to say that it can be used for lines 100% filled with liquid with no void space. For that first we need to find out for a given flow whether the given line will be completely filled or partially. Then if the line is found to be 100% liquid-filled, use the given equation. Please throw some light on this.

Also, have you heard about the vessels experiencing surge pressure ?

Regards
Narendra

[ankur2061]

Dear Ankur,

Where can we use these equations for pipe-line surge pressure calculations?

I think just before the valve closure the flow may be steady state flow (i.e. the properties like pressure, velocity and temperature are time independent), But the moment the valve is closed suddenly, the flow will no longer remain in steady state. Pressure and velocity will essentially become a function of time and may be oscillating as well. Please correct me if I'm wrong.

So, can we use the steady state equation for surge pressure calculation ? Do you mean to say that it can be used for lines 100% filled with liquid with no void space. For that first we need to find out for a given flow whether the given line will be completely filled or partially. Then if the line is found to be 100% liquid-filled, use the given equation. Please throw some light on this.

Also, have you heard about the vessels experiencing surge pressure ?

Regards
Narendra

Narendra,

The equation provided by me in the attachment is for transient condition but it considers a transient for instantaneous valve closure. If you observe, the valves which close in big pipelines generally take a few seconds to close. It is here when the time delay factor comes into picture that a massive amount of iterative calculations are required due to the changing behaviour of the shock wave or pressure surge which in turn is a result of the infitesmal movement of the valve with time.

The equation in the attachment will give you gross values for pressure surges. But if you are looking for accuracy based on the valve closure time, it is best recommended that a pipeline surge analysis software such as 'Stoner' or 'Olga' be used which will give you accurate results for scenarios such as valve closures with time delays.

Hope I have been able to clear some doubts.

Regards,
Ankur.

[Joyy]

Hi SPOOK,

See attached Joukowski formula for pressure surge estimation given in Perry. However, this equation shall be used only in single lines and not in branched networks where you would require transient analysis simulation softwares like Wanda, Pipenet, TLNet, others.

Your method 1 looks similar to the equation 6-206 given perry, which is a simplified equation for rough estimation and used where the valve closure time exceeds time required for the pressure wave to travel from the valve to the pipe inlet and back.

The magnitude of the pressure surge is determined by the velocity of the fluid in the pipe and the rate at which this velocity is changed. As a general rule a sudden velocity change of 1 m/s will generate a pressure surge of 10 bar. The effect of a sudden velocity change depends on the length of the pipe and of the closing speed of the end valve.
As a general rule pressure surges may be ignored if the pipe is shorter than 100 m and the closing time of the end valve is longer than 5 s.

Your fluid is LPG with a typical density of 550 kg/m3 (will vary depending on composition of C3, C4 and its temp.) has a much lower fluid bulk modulus of elasticity (in the range of 300x10^6 Pa) as compared to that of water (2193x10^6 Pa). Thus will result in lower surge pressure.
And the distance to travel in your case is lesser than 100m.
So, I feel if you calculate from Perry formula, you will get a pressure rise (Delta P) of around 0.5 bar. Add operating pressure to it to get the pressure surge.
Correct me if I am wrong, in your case you are loading LPG from the ship tanker into LPG tanks. Right.
Only areas of concern I could see is that the flexible hose and couplings are designed for this pressure or not.
Also the general rule for closure time is 1 inch per second for a valve upto size DN 600 and upto a max 30 sec for higher size valve. So for a 8 inch valve you can have 8 seconds closure time.

Hope this information helps.

Best Regards,
Joyy

Attached Files

•  Hydraulic_Transient__Perry_Chapt_06__6_44_.pdf   430.9KB   1289 downloads

[gvdlans]

Be aware that the "time of flow stoppage" as mentioned in Perry stands for the "effective valve closure time". This is the time in which the flow through the valve is reduced from 90% of the initial value to 0. This time is often much shorter than the total valve closure time (time in which the valve goes from the open to the closed position). There are some rules of thumb that relates the actual to the total valve closure time, depending on the type of valve (gate valve, ball valve etc.).

In relation to Total Valve Closure Time, this is typically about the last 5% for gate valves, about the last 15% for butterfly valves, about the last 25% for ball valves and about the last 30% for plug valves, when these valves are at the end of a long pipeline system.

Note that this is a rule of thumb only, for better values you should use software such as Pipenet or Wanda, where you can use the actual valve characteristics as input to the calculations.

[SPOOK]

Dear Ankur,

just to extend narendras question a bit. "Where to use this equations?"

Talking about pressure surges in liquid pipelines with pump, there are chances that the pressure surge will occur due to many circumstances. Some of which are emergency closure of control valves, triping of pumps, etc....

Now the equations that we are discussing here about (the ones that I have used and one suggested in Perry and other references). Do these equations give the highest surge pressure among all the case of possibility of surge occuring in the pipeline?

OR

We need to find a point in the pipeline where the possibility of maximum surge occurs and apply the equations???


Lets take my case for example, now there could be two cased one emergency closure of the ROV due to (say) leakage at the jetty and second tripping of the ship LPG pump. Now do we analyse both the situations and decide in which case the surge will be higher and apply the equation or just apply the equation irrespective of the cause of the surge?



Regards

SPOOK

[Qalander (Chem)]

Dear Ankur,

just to extend narendras question a bit. "Where to use this equations?"

Talking about pressure surges in liquid pipelines with pump, there are chances that the pressure surge will occur due to many circumstances. Some of which are emergency closure of control valves, triping of pumps, etc....

Now the equations that we are discussing here about (the ones that I have used and one suggested in Perry and other references). Do these equations give the highest surge pressure among all the case of possibility of surge occuring in the pipeline?

OR

We need to find a point in the pipeline where the possibility of maximum surge occurs and apply the equations???


Lets take my case for example, now there could be two cased one emergency closure of the ROV due to (say) leakage at the jetty and second tripping of the ship LPG pump. Now do we analyse both the situations and decide in which case the surge will be higher and apply the equation or just apply the equation irrespective of the cause of the surge?



Regards

SPOOK

Dear Spook Undoubtedly the analysis has to be thorough and complete with maximum possible degree of accuracy.
Especially when the pipeline in question

Has multiple turns(Ell's,Bends,offset spools etc. and) varying grade levels thereby gas/vapors pocketing may get initiated.

In such circumstance with sudden valve closure;certain degree of surge may be envisaged and most probably can be estimated with phased out segmental analysis.

Hope this proves helpful!