Volume of water required for hydrotest 5

[AKPipeFighter]

Is there a chart somewhere that shows me the volume of water required (after a pipeline is full) to raise the pressure to 200#s?? I'm working a COE job and I'm required to reach test pressure by adding not more than 10 gallons of water once the line is full. I have a 12" welded steel main, a 6" welded steel main, a 10" ductile iron main with flexible vic joints and an 8" ductile iron main with flexible vic joints. It seems to me that each line may require a different volume of water to reach test pressure depending upon the pipe material, the type of joints, the lining of the pipe (cement lined DI) and etc. This COE requirement seems very arbitrary. Any ideas??

 

[zdas04]

The key is your starting point. If you fill the line and let it breathe for a while, let the temperature equilibrate, then pressure it up to a few psi while venting it. Call it "full" at that point and the volume that needs to be added to reach test pressure is very small. If the water is saturated with air and at warmer temperature than the graound then the volume can be really large (compared to your 10 gallons).

On a full system, every degree F will change pressure about 100 psi. If you pump in cooler water than the ground temperature then you'll have to vent water to maintain pressure. There really isn't a rule of thumb, but there are ways to make the system work as long as there isn't a major leak.

The COE (Controllable Operating Expense?) limit is pretty arbitrary, but most limits like that are quite arbitrary.

David

 

[pipedream]

just a suggestion but fill the ductile pipe full of water and let it sit at least over night. the cement lining will absorb some of the water and has some air trapped in it.

just my opion.

just wondering ow are you filling he pipe lines........in other words will there be a static pressure in the pipe before you start your test?
G3

 

[Yorkman]

I agree with what zdas04 said, the key to making a hydrostatic test go well is making sure you have vented all the air from the system. That can be a challenge if the installing contractor was a little too frugal with the vents. As far as the 10 gallon rule I'm not famliar with that one, my only thought is the inspector wants to know that all the air is out of the system. Water being virtually non-compressable if all the air is out it will take very little water or effort to pressurize the system. Like zdas04 said you may have to vent water to control the rise in pressure due to higher ambient temperatures. If you're testing against isolation gate valves, you may want to install a pancake if they will allow it to maintain the pressure. Not knowing the system, if you have any compression or expansion tanks you may want to isolate those also.

I'm not a real engineer, but I play one on T.V.
A.J. Gest, York Int.

 

[CRG]

“Pipe Line Rules of Thumb Handbook”, E.W. McAllister, Editor, Fourth Edition, Chapter 5, “Hydrostatic Testing” has the some equations and examples that will show you how to answer your own question.

Is your job at Ft Wainwright? Does the project specification/contract outline this 10 gallon requirement?

 

[CRG]

In some lines that require hydro testing there may be air pockets that do not have high point vents. Two examples are externally pressured bellow expansion joints and open gate valves in the system. Both the gate valve bonnets and the expansion joints have air pockets that might add up to substantial quantities of trapped air depending on how many are in the system. So, the quantity of water to reach final pressure depends on the length and diameter of the pipe, the number of air pockets that cannot be vented, and the final test pressure. So, the 10 gallon amount is completely arbitrary.

Last summer, in Alaska, I heard from a second hand source about a Corp of Engineer rep at Ft. Wainwright stating that 10 gallon limit. All you can do is try to limited or eliminate air pockets, and ensure that the system is vented at very low pressure. As for the number of high point vents, if additional vents are required, request a change order to install them.

 

[AKPipeFighter]

Thanks for all the quick answers, I appreciate all the advise. The COE (Corp or Engineers) and yes the project is on Ft. Wainwright. I believe that we have sufficient air vents on the current system and will endevour to install enough on each system. However, I still think the 10 gallon limit is just an arbitrary number pulled from nowhere.

In addition, the limit is not found anywhere in my specs. There are no expansion tanks in the systems but I do have trap assemblies with sparging tanks.

It just seems to me that if I fill a 1/4" line 2" long perhaps it only takes a minute amount of water to bring the line to 200#s, while a 20" line 10 miles long most definately takes more water to raise the pressure to 200#s. Therefore, I'm assuming that some guidance exists somewhere as to the amount needed to bring a line to a specific pressure once the line is full and purged of air.

 

[codeeng]

For each atmosphere increase in pressure water would decrease in volume by 46.4 parts per million.

 

[AKPipeFighter]

Codeeng, do you know the reference book that the 46.4 parts per million come from? I've built a handy spreadsheet for the lines that I'll be testing but would like to have a reference for the factor.

Thanks, I really appreciate the help

 

[CRG]

The same reference I gave above, “Pipe Line Rules of Thumb Handbook,” has the following:

Temp Compressibility Factor
32 degF 3.35X10^-6 gal/(gal*psig)
50 degF 3.14X10^-6 gal/(gal*psig)
68 degF 3.01X10^-6 gal/(gal*psig)
122 degF 2.89 X10^-6 gal/(gal*psig)

The factor, 46.4 from codeeng is for water slightly warmer than 50 degF (3.14^14.7=46.2).

If you require further information, Kirk J. in your office on site has my email address (I assume he is still up there). Otherwise, Brad B. in Anchorage has my contact info.

 

[codeeng]

See this link for reference:

http://hyperphysics.phy-astr.gsu.edu/hbase/tables/compress.html

 

[AKPipeFighter]

Thanks Codeeng,

I had already found that site.

 

[Arcticdeserteng]

AKPipefighter,
I do not have a copy of "Pipe Line Rules of Thumb", sounds like a great book! I believe it is possible to calculate the volume change using the bulk modulus which is a ratio of the pressure stress to the volumetric strain. A quick look in Marks Handbook should help with the calculation. Hope this provides another way to your solution.

Keep your stick on the ice!
Arcticdeserteng

 

[AKPipeFighter]

What's "Marks Handbook" ?

 

[RobertACookPE]

I did my hydro-static testing many years ago on (shorter) nuclear power plant piping, where temperature changes (while the pipes and reactor vessels were filled solid) were mostly from running pumps or heaters, not from environmental temperature changes.

But similar rules will apply, particularly since your pipe is so long.

Note: Find out WHY the ten gallon rule is imposed: I suspect that it comes from some "authorized" leakage value for the actual fluid. The person/regulator who wrote that spec wants to assure himself/herself/her congresscritter that no leaks have occurred where they can't be seen, and that no leaks are being disguised by the hydro pump. But if 10 gallons won't pressure the pipe - even assuming it can be completely 100% vented and the water and pipe and fill are at a steady state temperaure! - then you'll never pass the test.

1) Calculate as close as possible (including elbows, fittings and valves!) the INTERNAL volume of your entire pipe run. You may find that the 10 gallon rule is impposible to meet if the pipe run is long enough.

2) To calc the volume needed to raise pressure in the entire pressurized run, assume you could fill the pipes and all branches 100% full at a static temperature to NOP (normal 100% operating pressure.)

Then, to raise pressure 1% PAST the completely full/static temperature point/completely pressurized point, the added volume of water will have to do two things: compress the water, AND expand the pipe walls (because the added stress inside the pipes). It is these two EXTRA volumes you need to calculate and compare to the arbitrary 10 gallon limit.

In other words, going from 300 psig in a completely full pipe (perfectly non-expanding) pipe using perfect (non-compressible) water to 330 psig in a real pipe filled with real water will require three volumes of water:
Vol1 = amount to raise pressure 30 psig,
Vol2 = amount to fill the expanded volume of the pipe as the pressure expanded the pipe, and
Vol3 = amount to make up for the small compressibility of the water.

Obviously, a 30 psig increase is so small that it is almost neglible, but your real hydro pressures are much higher.

NOTE: 8 US gallons = 1.0696 ft^3.
200 ft of sch 40 10" pipe is 109.46 ft^3
200 ft of sch 40 20" pipe is 385.92 ft^3
so your 10 gallons is only a small fraction of even a short pipe run.
It won't take long for your 1.0696 ft^3 of "allowed water" to become useless in raising pressure.

Try this, once you have calc'ed the actual volume needed under iso-thermal conditions.

Remove all bellows and pumps, in-line eqpt that is pressure-sensitive (we were NEVER allowed to hydro-test bellows!) and insert blank flanges and flanged spool pieces in their stead. Check for check valves and locations of your primary secondary, and backup pressure gages and relief pipes. You DON'T want the increase in temperature during the test to overpressurize the pipe! You DON'T want a branch to be isolated and over-pressurized because a check valve closed tightly and "isolated" the branch as temperature rises.

Fill the pipe completely, vent it, and leave it attached to a standpipe/expansion tank for two days to equalize temperature. Watch the standpipe level through one full day to see the effect of daytime temperature on volume as temperature raises in sunlight. Then start your hydro test.

Try to measure your expected temperature change now. It might really screw up your work despite your best efforts at venting and filling.

As always, check my work. Check my assumptions.