Pressure Testing of PVC Pipe

5

You see, the way this works is you provide a bit of information and the other members make an effort to keep you on the straight and narrow. "Give me a procedure" doesn't usually work. You need to tell us what your MAWP is (and what code was used to calculate it), what your test pressure is. What is the length in the branches. How are you going to isolate at the various terminal points (it matters in setting your acceptable leak rate). You have to do SOME up front work to qualify for free engineering.

David Simpson, PE
MuleShoe Engineering

Law is the common force organized to act as an obstacle of injustice Frédéric Bastiat

You might want to consider air testing this beast. With that many branches of unknown length and profile, getting even 90% of the air out looks really tough.

I can only assume each branch has a valve or a end flange or something to test against?

As zdas04 says - you need to show some prior thought and ask some questions, not wait for the answer.

My motto: Learn something new every day

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

Ever seen PVC come apart? I have and it fails in brittle failure at all temperatures. The time I saw it fail was with natural gas (not during a test) and one of the shards went all the way through a meter building. No one got hurt, but it was way more luck than planning. I wouldn't test PVC with air on a bet. Without knowing test pressure it is hard to tell how much bother degassing it will be.

David Simpson, PE
MuleShoe Engineering

Law is the common force organized to act as an obstacle of injustice Frédéric Bastiat

fyi http://www.oshrc.gov/decisions/html_2010/09-1158.h... (probably best to not even think about it -- however, several folks around this one just a few years ago may have gotten off quite "light", particularly in that air testing of pvc lines was apparently performed a couple times previously on this same job, except of course one "Jacky S Brown" to whom the sole idea was apparently convincingly attributed). In any case, and even with hydrotesting this or any piping it would be a good idea to be careful in the process, as it seems like it might be hard to get all air out of maybe particularly extensive branched systems like this.

Here is a procedure:

http://www.kdheks.gov/pws/mds/Appendix_C.pdf

You should install some intermediate valves so that you can test segments instead of the entire pipeline.

• http://files.engineering.com/getfile.aspx?folder=68ba1167-4f93-4a73-9994-dd

A good source for your pressure testing requirements is the pipe supplier, they will have all the information you need on the PVC pipe for hydrotesting.
Just something to consider is that there are subtle difference from one manufacture to another when it comes to pressure testing guidelines.

3km of 6" sch. 80 PVC pipe with 340 2" branches and no valves installed along the length? The first thing you need to figure out is how you will regulate the fluid pressure along the entire 3km length of pipe to the level prescribed by the test procedure. You also need to figure out how you will detect leaks at the hundreds of joints in the system for the period of time the hydro test requires.

I have actually seen it come apart in front of my eyes when using it as a student for a rocket. Repeated pressurisation had caused fatigue at a joint and it did indeed disintegrate violently. However PVC is used for lots of fluids and if there is a lot of air in this thing, the effect would be the same. Failure caused by any means is clearly not good for PVC, but if most of this is buried then it needs the ends or A/G sections to be protected against failure.

I knew I didn't like PVC pipe for some reason and now I remember...

My motto: Learn something new every day

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

The OP logged on today, but seems to have less interest in this thread than we do.

David Simpson, PE
MuleShoe Engineering

Law is the common force organized to act as an obstacle of injustice Frédéric Bastiat

"I knew I didn't like PVC pipe for some reason and now I remember... "

Then thank Ullas2711.

Irrigation pipe? Maybe his maximum operating pressure is only 15 foot of water?

Maybe. Unfortunately, this not a student question.

Hi All,
Thanks very much for the response. We have decided to install isolation valves at an interval of 500 mtr and then test the pipe. The working pressure is 10 bar and will be tested at 15 bar.

Can anyone help me in calculating friction drop in looped pipes?

thanks in advance........

Ullas…

For water, EPANET is free and easy to use. http://www.epa.gov/nrmrl/wswrd/dw/epanet.html. More sophisticated programs are available, such as WaterCad, H2ONET/H2OMAP, KYPIPES, etc., but you have to pay for these. The increased sophistication is in the added features, not the solving technique.

Although I do not recommend it, small systems can actually be solved by hand if you know the techniques. In college, we had to solve a three-loop system, by hand, using the Hardy Cross Method. Never again. Hardy Cross is primitive, but it is simple enough to implement on a programmable calculator if you're up for a challenge. Years ago, I wrote a Hardy Cross program for my HP-41CX and HP-42S calculators that can handle small looped systems (99 pipes max and 6 pipes per loop max), but they aren't in publishable form at this time.

Here is some info I found about the Hardy Cross Method in case you are curious about it:
http://en.wikipedia.org/wiki/Hardy_Cross_method
http://www2.dem.uc.pt/antonio.gameiro/ficheiros/ar...
http://www.cee.mtu.edu/~vgriffis/CE%203620%20mater...

If you do a Google search for [hardy cross method spreadsheet] you will find several that are available for download. I haven't tried them since I use EPANET, a commercial network modeling program, or my calculator programs.

Fred

==========
"Is it the only lesson of history that mankind is unteachable?"
--Winston S. Churchill

Mfgr recommends that you must never use air for testing PVC/CPVC plastic pipe
http://www.harvel.com/sites/www.harvel.com/files/d...

you must get smarter than the software you're using.