PSV Sizing 1

[Hutamaki]

I want to size a PSV as per API RP 520 PT1. Two suppliers come up with different sizes for the same specified flow. I have checked this and come to realise that it is the discharge co-efficient that produces the different results.

According to API RP 520, one can use a discharge co-efficient of Kd = 0.975 when doing prelimenary sizing. Then there are comments under 3.2.4. and 3.2.5. which i am trying to make sense of. It says under 3.2.4 that the actual discharge area for most valve designs will be greater than those specified in API Std 526 which allows the rated capacity to be greater than the estimated capacity for prelim sizing. It then says in the next par that the effective area or effective coefficient of discharge cannot be used with the actual area or rated coeffiecient of discharge.

Now i have a copy of the Crosby sizing program since i order this type of PSV quite often. When i size a PSV using this program i get a certain size (say a 6Q8) and i notice that Kd of 0.975 is used.

When i use another manufacturers program, it says that i need say a 8T10 for the same flow! I noticed on this program the the manufacturer uses a de-rating factor which results in an effective discharge coefficient of 0.9

My understanding is that both these Kd's are effective discharge coefficients. But does that mean that one valve has a higher capacity than the other? Or do you think one supplier is just being more conservative than the other?

In any case, it makes life difficult for me when i do costing and feasibiltiy. When does one get the actual discharge area and rated discharge coefficient - after the valve is manufactured? If someone uses Crosby here i would appreaciate your comments as well as to how you do the prelim and final sizing.

 

[TD2K]

I have an article on this in the office (where I'm not presently) but I'll try to remember the key points (but not necessarily the historical explanation how we got to this point).

If you look in a PSV catalog, there are two ways to calculate the capacity of a PSV. Using the API area and API discharge coefficients or the ASME area and ASME disharge coefficient (you must never mix one area with the other coefficent). You will not get the same capacity using the two sets of data. In my experience, the difference is greatest in the smaller sizes.

The actual capacity of a PSV is determined by testing. Per ASME, the actual capacity is derated by 10% which is the capacity you will get using the ASME area and ASME discharge coefficient.

So, what do you do? If you are looking at a PSV for a specific client or facility, they typically have their preference and I would use the formulas from the appropriate vendor for your sizing. If the PSV vendor isn't set, you don't really have any guaranteed way of ensuring the size you come up with during preliminary sizing is going to be the size the successful PSV vendor confirms.

 

[TD2K]

Here is the paper from Anderson Greenwood.

http://s21.yousendit.com/d.aspx?id=25MLH2R5SS0GK0G1U1V9ZUHE6Y

 

[Hutamaki]

TD2K, thanks for the response. I cannot download the file, is it perharps on the Anderson Greenwood website? If yes, what is the doc name?

I am not very farmiliar with ASME BPV. If i choose an ASME rated PSV then i should use the AMSE derating factor of 0.9 correct. Where can i find the formula for sizing using ASME BPV code (section/par?)

If i use the API RP 520 for sizing and i specify that the PSV is to be according to this code, will the manufcaturer then supply me the same size as per my prelim calc? i.e. will the area and co-efficient change after prelim sizing or can they just make a PSV to be able to flow the same amount of gas resulting in the same effective area and same rated discharge coefficient of 0.975?

 

[TD2K]

The title of the paper is "A new parameter for selecting a pressure relief valve size" by Donald N. Papa, PE. Anderson Greenwood, October 1, 1990. I'm not sure if the paper is on their website or if you have to request a copy.

You don't actually use the ASME 0.9 derating factor, that is built into the individual relief vendor's applicable ASME area and discharge coefficient. I don't believe the equations are in ASME, I've always gotten them from the relief valve vendor's catalogues.

 

[TBP]

I was able to download, save, and print a copy of the Anderson Greenwood paper, with no problem. Thanks

 

[EGT01]

Hutamaki,

The document that TD2K offers can be found on the Anderson Greenwood web site. Go to the following link...

http://www.andersongreenwood.com/literature.asp

Look for the following documents...

Technical Seminar Manual (Tech Sem Manual)
can be found under the section Product Brochures.

Pressure Relief Valve Engineering Handbook (CROMC-0296-US)
can be found under the section Engineering Documents.

The Technical Seminar manual has essentially the same information about difference between API and ASME areas and coefficients but in a slightly revised format.

The second document I've referenced above is the Crosby relief valve handbook. In it you will find excerpts from ASME Section VIII, Div 1 which is one of the Boiler and Pressure Vessel (BPV) codes. API RP-520 is a companion document for ASME Section VIII.

The basic relief valve equations that are used to certify relief valve capacity are found in ASME (see UG-131 and Appendix 11 in the Crosby handbook).

For a relief valve to have a UV stamp, it must be certified according to ASME. You can obtain the ASME certified coefficients and areas from the valve manufacturer. You can also find the values published in the National Board Pressure Relief Device Certifications NB-18 otherwise known as the "Redbook".

http://www.nationalboard.org/NationalBoard/Publications/NB18/RedBook.aspx

For the API coefficients and areas, you will need to obtain those from the valve manufacturer. If you know which manufacturer you want to use, then use either the API or ASME values the manufacturer provides and not the generic values you find in API RP-520.