Perfomance Testing on a 19000 PRV

[Instnapp05]

Hi All,

We are currently performing testing for set pressure indication on an assembled 19126M pressure relief valve and failed to set at the desired set pressure having inconsistent readings each lift. Apart from inconsistent readings, the valve itself did not passed the acceptable test for set pressure indication according to consolidated procedure as stated below but instead had a gush at each test.The said valve was removed and another valve identical to the one in question, same size, model, orifice and set pressure was tested in the same manner and passed set pressure indication with a steady stream at ninety degrees from the valve outlet, no gush and noise.

Consolidated procedure
First continuous flow is the point that the discharge first comes down straight out of the nipple on a 19000 valve or out of the flange on a 1900 valve.
In any case the liquid valve set pressure should occur when the discharge first stops curling(COHESION) under the nipple or flange and drops straight (90º) from the Nipple or flange.

VALVE INFO:
Model 19126MCF-2-CC-MS-31-16-10-LA
Size 1” 1500# / 1” 300#
Set Press 1450 PSI
CDTP 1144 PSI
CAP 128 GPM
BP 306 PSI

Inconsistent readings for three test performed;
1.1100
2.1127
3.1118

Note: The above VALVE INFO is identical to the other valve that was tested okay on our bench.

An inspection was done on both valves with reference to 19000 catalogue and revealed that the overall heights were different for A and C dimensions as attached. Can the illustrated dimensions have an effect on the performance testing of the said PRV in comparison to the other PRV that was tested okay?

Thanks and appreciate the feedback.

 

[JAlton]

The difference in overall height and the inlet piping should not affect opening pressure. The longer inlet piping may affect performance after the lift, but not set pressure. In discussions with PRV Manufacturer's and National Board PRV Lab Engineers, both have indicated that when a PRV goes straight to GUSH without a Steady Stream, this is accepted by the NB Lab as the Set Pressure. Some PRVs do go straight to gush rather than exhibit a steady stream (vertical stream off the outlet). The machining tolerances in the disc and nozzle may have some affect on this, so could the spring depending on the range of the spring and the set pressure. I would not "fail" a valve that went directly to GUSH. This is a valve that you know will flow its rated capacity before 110% of Set Pressure. I hope this is helpful.

JAC

 

[bevjones2009]

I agree with JAlton, you must check your internal components against the tolerance data specified on pages 19-28 in the 19000 maintenance book.

Looks like the PSV is a metal seated liquid service valve.
Also make sure your spring ID is correct, looking at your model, size and set pressure I think it should be; W4250NC (1385-1519 psi)

 

[JAlton]

Looking at your Test Results, 1100 psi, 1127 psi & 1118 psi, you have a PRV that is in Tolerance for the CDTP with no trending up or down. The set pressure is established and within 0.7% of Stabilized. I would not be too concerned. I recommend that you do one more test. If it is higher than 1103, but less than 1139, you will have a stable PRV. The initial 1100 psi may have been due to Air Trapped in the inlet of the PRV. Unless you have a specially designed Test Bench, there will always be a small pocket of air in the PRV Inlet. As the pressure increases, the incompressible Liquid compresses the small pocket of air under the seat until the PRV opens and relieves the air pocket (typically followed by a small amount of liquid). This could have given you a lower initial reading. Per ASME PTC 25, the Set Pressure is considered Established and Stabilized when you have at least 3 tests with no significant trend up or down and all tests are within 1% of the average. Your three tests were within +1.07% and -1.035% of the average of 1115 psi. Try four more tests, but discount the initial test and average tests 2, 3 & 4. See attached calculation sheet. Also, the Back Pressure affects Spring Selection. Your spring should be for 1144 psi not 1450 psi. In addition, there are different springs for Liquid PRVs and Air/Gas PRVs. The Liquid springs are typically weaker which results in the "gush" effect when opening. In this instance, the correct spring, W4220NC has a range of 1047 to 1148 psi. This spring in a Air/Gas Service PRV would be 739 to 801 psi. The fact that your set pressure is at the extreme top of the spring range may also be a factor to consider. I hope this is helpful.

JAC

 

[Instnapp]

JAlton and bevjones2009 thanks for the support thus far on this post.

Prior to performing the additional tests that you've recommended above,I've attached a picture of my test bench being used to test the PRV.Can you identify if the setup is suitable for testing water PRVs in comparison to other test benches viewed or worked on during your career.
Appreciate your feedback.

 

[JAlton]

Based on the end view of the Test Bench, I would say that you have an Energy Equipment Bench built buy my old friend, the late Mr. Larry Lane, or a very good copy of his design. Larry was a top hand on PRVs for many years. When I built PRV Test Equipment, he was my toughest competitor. It's a good design for Liquid Testing, a classic J-Tube Design. Very nice set up. Good Luck!

JAC

 

[Instnapp]

JAlton,

Based on your experience with reference to consolidated PT for air, gas and steam service valves, set pressure must be at a slow and uniform rate(starting from 90% of the anticipated set pressure, not to exceed 2 PSI per second)but rapid enough to cause the valve to open.

Does this also applies for testing of liquid PRVs?

Thanks and appreciate the feedback.