PRESSURE SURGE 5

[PUMPDESIGNER]

I have been analyzing PVC pipe lines for about 12 years with the formula below. I have found it to be useful and with 12 years of experience designing new lines and correcting lines with pressure surge problems I have good confidence that the results are useful and insightful.
However, the results I obtain with this formula seem to give high results, as if the formula gives worst case results scenario.
One factor that seems to be missing is the flexibility of the PVC pipe.
Question 1 - Is there a simple "add on" that I could use in this formula to take into account pipe material flex?
Question 2 - Does anyone know the history behind this formula?
Question 3 - Any other comments on the formula?

PR = [(.070)(V x L)] / T
PR = PRESSURE RISE
.070 = CONSTANT
V = VELOCITY
L = PIPE LENGTH FROM SOURCE TO STOPPAGE
T = VALVE CLOSURE TIME
Richard Neff
Irrigation Craft

 

[PVRV]

Mr. Neff,

Reference to your post dated 29/01/2002; I read an article in "PVC News" that outlines a study conducted by Utah State University to re-evaluate the ability of PVC pipes to withstand cyclic pressure surges.

The study concluded (Professor A.P. Moser) that the correct design methodology must take into account both the average stress and the stress amplification to accurately predict cycles to failure. Since the estimated time to failure for the tested samples was 322,000 cycles where as the pipe samples continued preform until the test was stopped at 3.5 million cycles without failure.

I hope the above helps; and kindly note that I am not advocating the use of PVC, HD/MD/LD-HDPE, RTRP, etc etc

Regards,

 

[stanier]

Tests on specimens that are "as manufactured" may well exhibit good long term results in their pristine condition. Lets see some results where the pipe has been dragged to the trench , rocked around on the back of a truck and generally mishandled.

CSIRO in Australia did some test back in the seventies and the results from that were not although complimentary for uPVC. Search for papaers by L S Burns.

 

[PVRV]

stanier,

The question at hand is related to surge stresses and accuratly depicting failure values. Material mis-handling in all cases may lead to diverse effects. Wether the pipe is made of PVC or Stainless Steel.

Typically the influence of an adquate quality managment system is not undermined and Contract Speciciations or Company General Procedure regarding material handling, stacking hights, transportation/stringing/lowering/jointing procedures are all very well dveloped to obviate premature degradation towards any pipelines preformance.

Regards,

 

[stanier]

PVRV

This is all very well in downtown Washington with all the supervision from the water authority superintendent. Try getting out to the back blocks in SE Asia and see how its done. We dont live in a perfect world and since the economic rationalists have taken control quality has gone out the window.

Pipelines go in having been mishandled. Hence testing should include susceptibility to damge. Give me PE, ABS or PP over FRP,uPVC, MPVC,OPVC or nay other PVC any day.

 

[PUMPDESIGNER]

I have been enjoying this discussion very much. You guys bring out a lot of useful insights.
Understanding pressure surge is tough because surge is dependent on enough variable system factors to make every job unique. Therefore I conclude that perhaps it is wise to use materials less prone to damage from surge.

Now concerning the latest comments, based on my experience using PVC, HDPE, Copper, Steel, CPVC, etc., HDPE is less prone to installation error than the others. Richard Neff
Irrigation Craft

 

[stanier]

Hi POUMPDESIGNER

Dont jump the gun here my son.

PE can be installed successfully by competent contractors.

Check out the website mentioned above

http://www.performancepipe.com.

It has some great technical information there. Get yourself a good supervisor experienced in welding PE and make sure that the contractor has certified welders.

I once had to rectify a pipeline that apssed a hydrostatic test but failed some time later. rember that a hydro test only subjects girth weld defects to half the design stress. There is no NDE I know of that can be repeatedly used in the field to establish if a buttweld is good.

So it comes down to repeatability by using good labour.

Best of luck in your endeavours.

 

[PUMPDESIGNER]

I agree completely with your post Stanier.
I had written more to qualify myself and then erased it, big mistake when sharp guys are afoot.
It is extremely important to have good, calm, competent men with good procedures.
I know that a fool can screw anything up like the old proverb says:
"A man would rather meet a bear robbed of here babies than a fool in his folly". (I had written that but erased it).

My post was unbalanced and left the possibility open that unqualified people could get away with sloppiness.

Richard Neff
Irrigation Craft

 

[PVRV]

stanier,

If quality control cannot be maintained during transportation and installation.

QUOTE
"since the economic rationalists have taken control quality has gone out the window." stanier

How do you propose to :

a) Justify cost nvolved in testing susceptibility to damage.

b) Cost involved in utilizing a superior pipe material.

Prudency would be compelling in taking extra care during handling than choosing options a) or b). Unless HDPE is cheaper than PVC Wonder what the cost ratio between PVC to PP is?

At any rate, your insights are appreciated and I am not located in North America.

 

[Ag93]

Stanier and Pumpdesigner,

There is a fusion equipment manufacuterer that makes a product called the DataLogger. This allows the user to record the critical parameters during each Polyethylene pipe fusion. Have heard from many that this is a great tool. Here's the link :

http://www.mcelroymfg.com/fusion/machines/datalog.htm

 

[chip1998]

Pumpdesigner, I think your formula is the same as the other formula (Jakwsky equation from the Perry Handbook)
It has been Rearranged. Iam not sure whether its in SI units so I will substutute K for the constant 0.07 in your formula.

Lets take a= wave celerity(celocity as defined by MortenA
Time to close valve should be at least equal to:
T=2L/a (From basic Physics)otherwise closure will be classified as instantaneous.This is used to determine whether water hammer will occur. If Time of closure greater than T, then you are safe for water hammer.

Jakowsky's Formula
Surge Pressure (H)=a*V/g
V=change in velocity which is equal to actual velocity
a=is defined above and can be rearranged to:

a=(2L/T) or (L/T)=a/2
Your Formula is: H=KVL/T=(K/2)Va=k'va
k'=new constant incorporating g.
This will work when you deal with one material and probably a specific diameter(celerity changes with diameter and amterial thichness and strength).Thats why it sometimes gives conservative estimates. You are likely to underestimate as well as lond as your case has a D/t and E above those assumed in your formular( The higher the D/t, the higher the celerity, the higher the hammer).Thans why a steel pipe will generate about twice the surge pressure developed by an equivalent PVC dia PVC pipe. My advise- stick to the equation given by MartenA.

 

[PUMPDESIGNER]

Thank you chip1998. PUMPDESIGNER

 

[stanier]

The Joukowsky equaation has limitations in that it doesnt represent the scenario of column separation and the two columns recobining. Streeter & Wylie caution engineers that the surge pressure can be many times higher in such cases. Also with thermoplastic pipe one has to consider the on set of full vacuum particularly in a buried pipeline scenario. If any ovality exists the eresistance to buckling i increased.

Suggested reading :-

Streeter & Wylie
Thorley
Boldy
Zhou, Hicks and Steffler
Molin
Janssen Sharing knowledge is a way to immortality

 

[MortenA]

Stainer

Its however far from every surge scenario where column separation is a risk. Could you give a better reference than "Streeter & Wylie" (i maen: where does the artickle apper, when, etc)? I have understood that Joukowsky is a conservative estimation.

BestRegards

Morten

 

[stanier]

MortenA

Sorry in just quoting Streeter and Wylie like that, it wsnt a paper although the gentlemen in question have certainly published.

Streeter & Wylie's Book Fluid Transients was published in 1993. It is a classic in this industry so much so that AFT provided me with a copy when I first started using AFT Impulse. They had a special print run done as it was out of print. Any good technical library (university/consulting engineers/etc)should be able to produce a copy for you.

Fluid Transients by Thorley or Pressure Surges by Swaffield and Boldy are other very good practical books on the subject. Sharing knowledge is a way to immortality

 

[mk2000]

PUMPDESIGNER and all
I have done a lot of research in valve design and water hammer in water mains. One solution I have found is a water valve with damping built in to it, (in opening and closing). An other option which is a little more expensive is to use a Feed back loop in conjunction with a proportional valve. They use this in some whishing machines to avoid water hammer and achive better control.

 

[PUMPDESIGNER]

mk2000 - You are correct that the valve closing is the easiest way to get help reducing the surge. Actually, I have described the desired valve as a linear closing valve such as would be possible with a motor operated ball valve. Solenoid / diaphragm valves are one of the chief sources of very fast non-linear closing speeds where 90% of the flow is stopped within less than one second. The damper you speak of would be great on a valve for this.

PUMPDESIGNER

 

[mk2000]

I have done some work on domestic slow acting valves. These valves are mainly used in house hold appliances. Depending the size of your application you might be able to adapt these valves in to your application. But I think I have seen valves in larger application with the same sort of functionality.

Also if you can do a bit of testing you can get a linear closing/proportional solenoid valve and put a big Green Cap on the terminals. This will reduce the response time of the valve. However you have to select a DC operated coil.

 

[PUMPDESIGNER]

mk2000 - Very interesting. Green Cap signifies DC?
How would this proportioning valve work if you have a couple of minutes to explain to me. I have been experimenting with internal shaping to force hydraulic linearity in the face of collapsing downstream pressure which causes diaphragm to lose support and collapse onto the seat.

PUMPDESIGNER

 

[mk2000]

PumpDesigner

What we have developed is a conversion of a digital diaphragm valve in to a proportional valve. I do this by changing the spring behind the armature and that makes it proportional (this is on a DC coil). But we us this to control the temperature of the water. However you can use the same concept to change it to a slow acting valve.

As the valve opening is some function of the coil voltage Flow = (X)Voltage + B (this is the ideal situation, the function is not exactly linear but we can approximate it). By putting a large Green cap on the valve when the voltage across the valve is dropped to zero, then the green cap will start to discharge at some rate depending on the capacitance of the cap. As the cap discharges the valve closes slowly and therefore eliminating water hammer. This also works the other way. If there is a shock wave propagating through the pipes to the valve then because of the sprung system and the way diaphragm valves work it will dampen some of the shock and reduce the water hammer effect.

The approach you are taking only looks at half of the problem. Do not forget that the valve control has two sides you need to consider Dynamic effects and Static effects. Your analysis only looks at the dynamic effect only that is why you get the collapsing effect.