water hammer valve opening

[rppearso]

Does anybody have any resources/equations/publications/etc for calculating valve opening times to prevent surge of an immediate downstream elbow. In my scenario I have a closed SDV with high differential pressure and an elbow downstream that hammers and I need to calculate how slow to open the valve to prevent hammer of the downstream elbow.

 

[MortenA]

For a simple Jukovski calculation look e.g. at Perry.

For more complicated calculations then i will recommend use of simulation softwae. There are numerous packages avaiable e.g. Flowmaster or pipeline studio, pipenet etc. Also some freeware.

Best regards

Morten

 

[MortenA]

Sorry didnt ready your post fully.

What i mentioned will propably not do what you are looking for since its not water hammer as such. A bend is a fairly low resistance and wont slow the water much - causing a waterhammer/surge.

Best regards

Morten

 

[rppearso]

I have found there is very very little information on this topic out there in regards to actual quantitative calculations that give you a valve opening time. I was hoping for a pointed useful text reference or someother such resource. I cant imagine this is an unstudied or grossly understudied topic, but I guess its not impossible. I have the Wylie/Streeter text (which is extremely kriptic) and does not include valve opening as far as I can tell.

 

[JohnGP]

rppearso,

There seems to be a few references out there - how about

http://www.ajdesigner.com/phpwaterhammer/pressure_increase_equation_valve_closing_time.php

.

Regards,
John

 

[gerhardl]

Suggestions for 'roundabout alternatives' in addition to finding the straightforward answer:

If possible:

a) Longer straight stretch before the bend.
b) Mechanical reconstruction else, larger radius, larger pipeline dimension.
c) Compare with closing times? Necessary closing time calculated for closing the valve, actual construction upstream and downstream, or imagined construction. Opening time = closing time?
d) How long time for opening can be accepted? Use this?
e) Slow opening x% rest faster?
f) Practical experimentation?
g) Down-scaled model?

If you have a valve giving this problem you should also check the consequences on the valve side. Cavitation when opening, and/or backlash flow conditions for the valve?

Pressure peaks when closing, upstream and downstream?

Are the valve conditions generally within allowed limits?

 

[rppearso]

a) Longer straight stretch before the bend. This is a possible recommendation but how much longer?
b) Mechanical reconstruction else, larger radius, larger pipeline dimension. This recommendation would have to be rigoursly document with no other means to mitigate the problem since this would be an expensive option.
c) Compare with closing times? Necessary closing time calculated for closing the valve, actual construction upstream and downstream, or imagined construction. Opening time = closing time? I beleive the pricipals between opening and closing are different, they could be the same but I cant prove it.
d) How long time for opening can be accepted? Use this? This is what the instrument TA has stated and if worse comes to worse I will put in a control to ensure the upstream control valve is fully closed before the SDV is opened and then allow the control valve to go back into normal control mode and HOPE there are no issues lol, if there are we may have to revisit this issue.
e) Slow opening x% rest faster? This is something I have seen done in a stoner surge model for valve closing, supposidly the stoner model can do opening as well but they are very kriptic about there back ground calc (for good reason) and very expensive to engage for a single PHA action item, if this were a full blown medium to large project I would have no problems engaging them.
f) Practical experimentation? This is a thought but would be immpractical during operation of the plant without shutting in the vessel
g) Down-scaled model? No way BP will pay for this for a PHA action item but it would be fun.

If you have a valve giving this problem you should also check the consequences on the valve side. Cavitation when opening, and/or backlash flow conditions for the valve?

Pressure peaks when closing, upstream and downstream? I am not sure, I would assume there is downstream pressure peaks for a very brief time, there is a down stream PI but no operation data when the event occured, shut down of this vessel is uncommon and when it hammers people are leaving the area and no one is looking at a PI

Are the valve conditions generally within allowed limits? I would assume so but this scenario is a special case of bringing the system back online after a shut down, the control valve controls on level so pursumably it closes on its on in automatic when the vessel level is very low and reopens but the concern is when the level is very high it will quickly come wide open to get the level down until the level is within the control range at which point a hammer has already occured.

Part of this is accademic on my part as well it would be very interesting to try to develope an equation where velocity, mass flow and area is changing and the fluid is basicly being accelerated throught the valve modeling the valve as a variable orifice using F=m*dv/dt and F=rho*A*v^2 but the problem here is area is changing and what is the mass?

 

[BigInch]

In the general case, the pressure downstream of a closing valve is initially normal operating pressure minus the wave pressure amplitude, until the wave hits a feature downstream that reflects it and then it will arrive at the originating point at a lesser amplitude, because of friction losses in transit.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[sterl]

If you are making up liquid to a saturated vessel with a cycling valve you have a 2-phase topic...and likely condensation induced shock.

Is the upstream pipe colder than its surroundings and higher than the source liquid level? Is the immediate downstream above or below the operating liquid level?

A slow open/slow close valve will deal with either condition but the question will still remain, how slow is slow enough?