SEALING TEST ON SAFETY VALVES

In response to requests from some colleagues here on Linkedin, i am posting today the text in english of the text i published in portuguese.

Starting today, I will start publishing small articles related to valves here on Linkedin every fortnight. In today's text the topic is specifically about the "Sealing Test on Boiler Safety Valves". Testing the Boiler Safety Valve Seal After the set pressure is set and approved, the valve sealing must be checked. Boiler safety valves normally have an open-type bonnet and also have two rings to control the relief differential. The open type bonnet allows for greater thermal dissipation of the steam temperature acting on the spring, which could reduce the set pressure value or prevent the repeatability of that value if the bonnet was the closed type. There are some methods that can be followed to check the sealing of a safety and/or relief valve, however, for boiler safety valves there are basically two methods, the so-called "test with water trapped at the valve outlet" or the "steam test" on the installation itself. Although the ASME Code Section I (boilers) does not cite API Std 527 as a reference for tightness testing on safety valves. ASME Code Section VIII references this API. This API is used as a reference by manufacturers of safety valves for use in boilers. Note: The API Std. 527 is the ASME Code Section VIII reference in Appendix M10 (b) and Section XIII (2021 edition) in paragraph 3.6.4 for safety and relief valves constructed in accordance with ASME Section VIII that will be applied to pressure vessels. With the valve mounted vertically on the test bench, the leak rate, in bubbles per minute, must be determined with pressure at the inlet of the safety valve, which must be maintained at 90% of the set pressure. For valves set at 50 psig or below, the test pressure must always be maintained 5 psig below the set point. The allowable leak tolerance for the seal test through the bubblemeter, according to API Std. 527, is as follows. However, for safety valves with an open bonnet and/or hood, which can only be tested on a test bench in a workshop, with water trapped at the valve outlet, the leakage tolerance is equivalent to 50% of the values mentioned below: holes bubbles allowed D, E and F ("F" hole with a diameter of up to 18 mm) 40 bubbles per minute G to T 20 bubbles per minute. In valves that have an open bonnet (whether for boilers or pressure vessels), the best and most accurate method is with water trapped in the outlet flange, with the water at least ?” above the level of the nozzle sealing surface. In safety valves that have the upper ring, this ring cannot be below the bottom face of the disc support so as not to influence the test result, in case there is any leakage through the valve sealing surfaces. What can happen, if this ring is below that face, is that any bubble that may come off the surface of the water, due to leaks, may come out through the thread between the upper ring and the guide, not being seen at the time of the test. The photo below shows a safety valve with the two rings (upper and lower) before the sealing test, with the upper ring above the lower face of the disc holder: No alt text provided for this image Top ring above bottom face of disc holder and before seal test. Therefore, the allowable leak rate is 50% of that allowed for the bubblemeter test. For the safety of the operator at the time of this test, it is recommended that the leak be viewed through a mirror to avoid direct viewing, which could cause a serious accident if the valve opens. In this test, the mouthpiece ring should also be in the lowest position to avoid accidental “pop” due to valve opening. The maximum time for observing leakage using this method is 1 minute, regardless of valve size. In this test method, there is no need to keep the valve pressurized for a few minutes, according to the inlet gauge and before starting the bubble count, as required if the test method were through the bubblemeter test. In this test method, the visualization of any leak is immediate, while in the bubblemeter test the leak must pressurize the entire body and bonnet, if the valve is a conventional style, or the entire region external to the bellows in the valve body, if it is a valve balanced, before the bubbles can be released on the surface of the water. Bubble spacing depends on the height of the water surface relative to the nozzle sealing surface, water density, width and depth of scratches, etc. on the bench. If during the test a fluid with density smaller than that of water, such as alcohol or kerosene, the spacing between the bubbles (period of time between one bubble and another) will be smaller, consequently, the amount of bubbles released from the surface of the water will be greater.

Note: Reducing the percentage between the bench test pressure and the valve set pressure tends to increase the leak rate because it reduces the force differential that keeps the valve closed. Following this same line of reasoning, increasing that percentage, that is, reducing the value of the test pressure (on the bench), tends to reduce the leak rate, as it now increases the force differential that keeps the valve closed, as it favors the closing force being exerted by the spring.

Note: Inspection with penetrating liquid in the nozzle (still in the inspection phase to check for pores or cracks) is an important item not only for the proper functioning of the valve, but also for the safety of the test operator, as the test fluid because it is compressible, it has a large stored energy.

Leak visualization through black background (Steam Test) – With the valve installed in the process, and operating with water steam, a black paper can be placed behind the discharge pipe and parallel with the outlet point. If there is a leak, it will be easily seen by the black background. According to the ASME Section I code, the safety valve must be subjected to the maximum operating pressure of the boiler, however, below its closing pressure. It will be considered tight and approved if no leakage is seen through the black bottom at that maximum operating pressure. This test procedure is also mentioned in API Std. 527.

In the next article, we will discuss the methods of testing the sealing of safety and relief valves applied to the protection of Pressure Vessels and Pipes. Until then.