Removing water after hydrotest 1

[donf]

I'm looking for experience, suggetions, advice on dehydrating a tubing system after a hydrostatic pressure test (I did search this forum. Found some quite spirited discussions on testing - but not what I'm looking for).

Here are the particulars:
Code in effect = ASME B31.3
Process Fluid = Proprietary liquid. Adversely reacts with water.
Fluid Category = Normal Fluid Service
Max operating pressure = 3500 psig
Max operating temperature = 140F
System capacity = About 30 lineal feet of 1/4" OD tubing; several valves.
Joints = twin ferrule compression

Testing with the process fluid has been ruled out (for now).

A pneumatic test would eliminate the concern of residual water reacting with the process fluid. I've been involved with a number of pneumatic tests, so I'm aware of the safety concerns. The problem is that the max pressure readily available is 2200 psig nitrogen.

The next option is hydrostatic testing. As stated above, the process fluid reacts with water, so I'm checking on a possible test fluid that's compatible with the process fluid. If that doesn't pan out, I'm left with testing with water......

Which brings me back to the reason for my post: How to dehydrate the system after the hydro? Here are my initial thoughts:

1. Design the system to be as free draining as possible.
2. Determine how dry is dry enough.
3. Determine method(s) to dehydrate. Vacuum? Nitrogen or Argon sweep? Other?
4. How to verify system sufficiently dry?
5. ??

Thanks in advance for your time.

donf

 

[zdas04]

First if the only major issue with pneumatic testing is availability, check the Oil Field service companies (Haliburton, Schulmberger, BJ, Weatherford). They all have liquid nitrogen available that can come out of their trucks at really high pressure (I've seen 10,000 psi).

In terms of drying, does your proprietary fluid react with water vapor or just liquid water. If liquid is the problem, then you can probably get where you need to be with a LP nitrogen purge. If you have to get it to a 32 F dew point at 3500 psig then you have to get it down to around 4 lbm/MMCF. The only way I know to get there is really hot nitrogen purge, then seal it with the hot nitrogen inside. The substance you use is less important than the temperature at start.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.
The plural of anecdote is not "data"

 

[chicopee]

About blowing in dry compressed air?; or about injecting pure alcohol which is used by the propane industry to remove traces of water from their bulk tanks?; or heat up the 30' lineal feet of 1/4" tubing in a large oven?

 

[moltenmetal]

You have a system of 30 lineal feet of 1/4" tubing with a few valves, with a design pressure of perhaps 4000 psig? Note that max operating means nothing- what's the relief pressure?

Pneumatically test it to 1.1x the relief pressure and skip the water!

It would be crazy to hydrotest such a small system in my opinion unless people cannot be removed from the area during the test.

 

[TomDOT]

Yep, that is a tiny volume. No need to get a services company, just get Air Liquide or someone to deliver a bottle of dry compressed nitrogen. Easy.

 

[zdas04]

Funny how your mind can pick and choose what it expects to see. I read the 1/4" as 4" and "30 lineal ft" as "30 miles". No you don't need a truck of liquid nitrogen. You have many options (including putting the assembly in an oven an baking the water out).

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.
The plural of anecdote is not "data"

 

[dicksewerrat]

If the thing is too big to put in the oven in the lunchroom, take it to an auto paint shop.

Richard A. Cornelius, P.E.

http://WWW.amlinereast.com

 

[donf]

Thanks to all for your responses.

General - The section of high pressure tubing is connected to a larger piping network, so drying in an oven is impractical. (dicksewerrat - thanks for the paint shop oven suggestion. I may use that in another application).

zdas04, TomDOT - I found a gas booster that I can hook to a standard bottle of Nitrogen (~2400 psig) and boost the pressure to what I need. Process fluid does react with vapor as well.

chicopee - Looking at using alcohol as a possible test liquid.

moltenmetal - We're in agreement on the RV set pressure being the number to use.

Again, thanks to all for your help. I'll let you know which option we use.

donf

 

[gtaw]

Have you considered testing the system with distilled water and then drying the system by blowing it down with compressed nitrogen or argon? Either gas will be supplied with a dew point down around -76 degrees F with a moisture content around 10 ppm if it is ordered to AWS A5.32 specification for shielding gas. That may be sufficient to remove all but trace amounts of moisture. Will the system be capped and sealed to keep moisture out after the test is completed an the system is "dried"?

Best regards - Al

 

[moltenmetal]

This is silly, guys- it's 1/4" tubing! There is no need to hydrotest it.

You can rent a 6000 psig cylinder of nitrogen and test it right off the bottle. No need for a bottle booster pump.

 

[donf]

Moltenmetal - I'm definitely leaning toward using nitrogen. I was told that 2400 psig was the max I could get in a bottle. Thanks for setting me straight.

donf

 

[moltenmetal]

6000 psig is a readily available cylinder nitrogen pressure. We always keep a couple around for such tests, which we do routinely.

Keep people out of the area. Make sure that pressure can't get anywhere beyond the test and do damage (i.e. by leaking through a valve etc.). Ensure the tubing is properly supported. Put the cylinder at a safe distance from the line to be tested, install an excess flow check at the discharge of the cylinder regulator to limit flow during a failure, and stand by the block valve during the test to shut it off if there's a failure etc. Above all, make sure that nobody manipulates the fittings under test to try to stop leaks- that's what is going to get someone hurt. Bring pressure up in stages with a hold in between, and snoop (external soapy water test) joints for leaks. Leave the system at pressure for a time at each stage before "snooping". Depressurize prior to ANY attempts at repair.