Cold Differential Test Pressures of Relief Valves

[Ripz]

Hi,

I have a general question regarding Cold Differential Test Pressures of Relief Valves. I understand the concept and it's relationship to Backpressure.

What i'd like clarification on is, how does operating temperature affect the CDTP, why and by how much.
How do designers know how much compensation is needed based on the Operating temperature. Is there a rule of thumb or a specific formula?

 

[Duwe6]

CDTP is also to compensate for the spring getting weaker as its temperature increases. This derate [increase in test pressure] comes from the PRV manufacturer, and usually starts at 200°F. THUS:

If your process runs at 550°F, and the 200 psi PRV is located such that the bonnet & spring are typically 400°F, you are to apprise the PRV Test Facility that you have a 400° Normal Operating Temp [not 550°]*. The compensation will be on the order of 2%, so your CDTP will be 204 psi 'cold' and allow the PRV to relieve at exactly 200 psi when at its Normal Operating Temp.

*The process temperature is pretty immaterial. What matters is the actual bonnet & spring temperature. Generally means that you get to crawl/climb around with a temp 'gun' while your plant is running.

 

[Ripz]

I understand that it's due to the spring strength decreasing. But i came across two manufacturers, approximately the same operating temperature, pressure, design and backpressure. Yet one manufacturer stated a compensation of 1 bar, while another had a compensation of 12 bars. This vast difference is what confuses me regarding the actual calculation used to come up with this factor.

A follow up to this question is, if we were to have a problem with premature lifting of the PRV, and wish to perform and adjust via Electronic Valve Testing, would you set the RV to the Set Pressure or the CDTP?

 

[steamer1966]

Ripz

At our plant we recently have experienced a premature drum PSV lift, This was a new PSV and was bench tested at a qualified shop (air was the test medium and compensated). What we proceeded to do was perform a PrevenTest (Trevi test) the valve while in service and re-sealed after the adjustments were made. They still set it within the allowable tolerance, but I requested we set them to the higher side for that variance. This worked just fine.

 

[Duwe6]

The manufacturer's CDTP offset is not calculation based, it is experience based.
And I agree 1 bar vs. 12 bar is a big difference. Somebody [or both] may be wrong.

I would "Trevitest" while having been at normal operating temperature and pressure for 'a while' [long enough that the entire system is stable, temp-wise]. At normal operating conditions, you test using the 'actual' pressure, not the CDTP. Your PRV is not 'cold', so CDTP cannot apply.

 

[Ripz]

Thanks for your advice. This is what we practice. With the Trevi test.

Wanted to understand the rationale behind the offsets. Since you say it's experience based, that sucks for us, because we've have a designer here, who gives us a different CDTP each time the valve lifts prematurely. Make us feel like guinea pigs.

 

[Duwe6]

Just go 'visit' the PRV, and bring a temperature 'gun'. Take the temperature of the bonnet [actual spring if you can get to it]. Give that temperature to your "VR" PRV shop. They have access to manufacturer's tesing, so the CDTP that they give you will be pretty accurate.

Theoretical CDTP will only get you started. Now you have to go out and "ask the PRV".