Hydrotest pressurization rates...

[jrjones]

A co-worker and I were discussing a specification for hydrotests of piping and pressure vessels. While there is a lot of information on temperature limitations during a hydro (to avoid brittle fracture, etc), I can't find anything that provides guidance on pressurization rates. For example, if you have a piping system with a hydro pressure of 1800 psig, should you ramp up at 100 psig/minute?..50?..200?

My thoughts are unless you could shock the system by going to full pressure in microseconds such that you have some sort of kinematic effect...there is no need for a limit.

On the practical side, I know that hydro contractors bring in these big pd pumps and sometimes can overshoot the desired pressure because it may only take one pump stroke after the line is inventoried with water. So you may want them to slow down...but is that even practical.

API 570 give guidance on going to the max hydro pressure for x time and then dropping to the design pressure for inspection...but nothing on rate.

What about the depressure rate?

Your thoughts on the topic would be appreciated...

jrjones

 

[vesselfab]

When we hydro piping, we generally add in an accumulator to add volume to control the overshoot of pressure

On vessels that we build, there is enough volume that this is not a problem.

 

[dcasto]

You don't use a big pump to pump up the pressure. It only takes a hand hydraulic pump. If doing a test under B31.3 you should stopp and perform leak checks, thats what its all about.

 

[jrjones]

I understand about the use of small pumps...but...during a major shutdown of a operating unit, the practice we use is to hire a single contractor for all hydros. This contractor can fill the lines with water (or glycol) and pump up with one unit.

Aside from the pump issues, what would people suggest for pressure up and pressure down rates.

jrjones

 

[zdas04]

A safety manual chapter at my previous company said

Pressure changes faster than 10 psi/minute can introduce localized stresses in excess of the average stresses calculated and should be avoided. It is prudent to allow a hold time of several minutes at MOP, MAWP, and every 100 psi above MAWP to allow stresses to equalize

Before anyone asks, the manual doesn't give a reference and as I recall from chairing the committee that wrote it, that particular language came from a similar document in use in a refinery for over 50 years.


It may be just crap, but it feels safe and I've never had a line fail while following it (no way to know if any of them would have failed if I'd gone up quicker).

DCASTO's comment is important. When I have to do hydro's, I'll fill a line with a centrifugal pump, never a PD pump, until I get water out all the vents then 25-30 psi higher. Then let it sit overnight and turn the pumps back on while venting the last of the air out. Finally, use a very small pd pump to get pressure. Filling with a PD pump feels pretty reckless to me.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The harder I work, the luckier I seem

 

[BigInch]

It depends on if you are testing a vessel or a piping system. In addition to the rules mentioned above, it is far more important when testing piping systems to fill at a rate that forces air along ahead of the water, but not so fast that any air becomes entrained. A stabilization period is also helpful after filling, but before pressurization to allow any entrained air to be let out the vents and to allow the pipe/vessel temperatures equalize with the water temperature. You'll get a much nicer test plot in a lot less time, if you do these things.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[jrjones]

I agree with the fill up considerations to ensure that the air is out of the system. I'm not sure about time for stress equalization. As long as I do not shock the system (piping or a vessel) the stresses are, for the most part, primary (load controlled).
While the 10 psig/minute that zdas04 seems too slow, what is being proposed in the specification I am reviewing is upwards of 150-200 psig/minute. This "seems" fast, but I really don't have a basis. I have seen hydros done at >600 psig/min. I agree this seems crazy, and it is a reason for the specification, but I still have no sound basis to pick a rate (and justify it to the spec writer). I do agree with the tight control once the pressure>MAWP.
Just for clarification, inventorying and venting the line is not an issue...just the pressurization and depressurization.
Thanks for the continued discussion...jrjones

 

[zdas04]

200 psig/min takes you from zero to 900 psig in 4.5 minutes. I've never had a pump at a pipeline test that could do that. 10 psig/minute covers the same range in an hour and a half, add in 5 minutes for a soak at MOP, MAWP, 700 and 800 and you're closer to 2 hours--for a pipeline of any length that has typically been about as fast as the pump would run, and bigger pd pumps really scare me as you approach test pressure.

David

 

[jrjones]

I here you about the scary pd pumps. I have been in the field during a shutdown and had the inspector call me over to a hydrotest because of a "problem". I knew that the pipe in question was 150# class...when I arrived the inspector asked me if going over the test pressure (~350 psig) was okay. I said that it depended on how much, but that the piping was new and okay upto ~450psig. Then I looked at the guage (the top one on the system) and say 825 psig!!! What's even funnier...the flanges didn't leak. Anyway, that was a busy day.
I am really interested in this discussion because I have literally witnessed 500+ hydros (on piping, vessels, tanks, pump casings, valves, etc). I am usually called after the line is holding pressure, but sometimes I have been there, knowing that I specified a ramp of 100 psig/minute max and seen the pressure go from 0 to 800 in 10 seconds. I have seen this more that I would care to mention (again...the reason for the specification) but it does make me question whether those rates are okay...I have never seen a pipe fail hydro other than through flanges or valve packing, etc. So maybe rates > 200psig are okay??

 

[dcasto]

jrjones,
See this thread: thread378-193569