2 ms·

> If you specify a closed-bonnet PSV (like the vast majority of PSVs) for steam service then the whole valve body can get up to the steam temperature, including

> If you specify a closed-bonnet PSV (like the vast majority of PSVs) for steam service then the whole valve body can get up to the steam temperature, including the spring, and it doesn't lift until a much higher pressure than your bench test indicates. That was surprising. With a quick search about temperature and springs I found 3 items: 1 Spring constant descreases with temperature, not much at small increases but significant at much higher temperatures (that would make it open at lower temperatures than expected) 2 Spring gets longer at increased temperatures (that would make it open at higher temperatures) 3 Durability issues Is effect 2 stronger than 1 at the relevant temperatures? If there's a fixed relation between steam temperature and pressure it should be possible to design one to open at the correct pressure at the temperature that this pressur occurs at?

iamthemonster 3y ago
Yes definitely - the vendors of the PSVs will calculate for you a special compensation factor that allows you to test the PSV at normal room temperature, but still have it open at its set pressure when it is at a higher operating temperature. They call it the "cold differential set pressure". You can calculate your own cold differential set pressure just as a cross-check against the vendor's calculation to make sure there are no misunderstandings but ultimately you should just use the vendor's value. In the case of steam, because it is obviously non-toxic and non-flammable you can just let it go to atmosphere in an emergency, so the best thing is just to let the spring of the PSV be exposed to atmosphere rather than some variable operating temperature.