PSV allowable back pressure 1

[processeng01]

Dear All,

When sizing the safety valve vents in power plants, I use the equations given in ASME B31.1 Appendix II. According to the calculations, it seems that for 99% of steam safety valves common in the market, the valve exit pressure is too much (since the SV nozzle as well as the vent tip reach choking conditions)and the typical 10% permissible backpressure rule of vendors(built up or superimposed) is never satisfied. Check out the example given in the Appendix II. The set pressure=900 psig, SV exit flange pressure= 194 psia, the ratio for this case is 194/(900+14.7)= 0.21 (21% of set pressure)

Vendors state that if their permissible back pressure is exceeded, bellow type should be considered. However I do not see many bellow type PSVs offered/sold in the boiler market. Can anyone clarify this? Thanks in advance. Sorry for my bad English.

Processeng01

 

[pleckner]

Which vendor(s) have you talked to?

This is a problem not only in boilers but also in process steam relief to atmosphere. In these cases, most vendors have PSVs specifically designed for steam relief and take the back pressure issue into account.

The whole concept of the 10% variable back pressure (conventional style PSVs) and 40% to 55% for balanced bellows PSVs is so that these valves will stay open or not loose significant capacity. Nothing says you can't have even greater back pressures if the PSV is designed for it and the rquired relieving rate is achieved. Code only says you can't do anything to impede the the ability of the PSV to relieve at the proper rate to protect the equipment. These back pressure percentages are only guidelines and are based on conventional designs.

 

[Ashereng]

There is a difference between constant and variable back pressure also. pleckner alluded to this in his third paragraph.

A constant backpressure is easy, just account for it in the setting of the spring.

Variable backpressure is different. This is where the bellows comes in. If the bellows don't work, then onto a pilot PSV and big bucks.

I see lots of bellows PSVs. I haven't bought a pilot PSV in years.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[iainuts]

Hi Process. You are correct, and this observation surprised me too. I discovered the same thing quite a few years ago.

The problem you're seeing is the apparent high back pressure inside the body of a conventional, unbalanced relief valve during flow, right? In the case I was looking at, there was a female pipe thread in the outlet of the valve, and even without the pipe screwed in, it seemed the outlet port would be choked and back pressure on the poppet would be more than the 10% recommended by ASME. With a pipe screwed in, I seem to remember the back pressure was on the order of 30% or 40% of set pressure inside the valve. I wondered how in the hell it could possibly work, since we would assume that for every psi of back pressure on the valve, the poppet would also see that pressure and have a tendancy to close. Thus, it seemed to me the valve would close or chatter unless the inlet pressure rose to something like 150% of set pressure! That's probably a bad thing if you want to prevent tanks from exploding.

I brought this to the attention of the manufacturer, and discovered a bit about relief valves in the process. They did a test on the valve, measuring back pressure inside the valve body with and without a pipe, and the pressure matched the pressure I had calculated which was well above the 10% of set pressure. Nevertheless, the valve worked flawlessly! It took a bit of patience to understand why.

~

Before a relief valve lifts, the area exposed to pressure is the area of the seat. That area times the pressure results in a force which must be balanced by the spring. So far, that much is easy.

Once the valve lifts however, the area exposed to the inlet pressure is no longer the area of the seat. Your relief valve has a "huddling chamber" which in essence, increases the area exposed to the upstream pressure. That's why those valves have a snap action, the area suddenly increses and the force opening the valve suddenly increases along with it. The spring no longer is just balancing the inlet pressure, it is suddenly overwhelmed because of the pressure acting on a much larger area.

The huddling chamber creates a second restriction. The main restriction can still be the seat diameter, which is choked, but there is a second restriction now between the huddling chamber and inside valve body pressure. It's this pressure in the huddling chamber which allows the valve to function normally even with that high back pressure you are calculating inside the valve body. The pressure inside the body of the RV is high, but that downward force on the poppet is balanced by the even higher pressure inside the huddling chamber.

This imbalance arises because of the increased area. Even though the nozzle of the RV is choked, and even though the pressure inside the valve body is very high, if you do a force balance calculation on the valve, you need to look at the new area which includes the huddling chamber pressure which provides additional upward force on the poppet. It's not really that difficult to calculate, but it is a bit of a surprise if you've never seen it before.

The pressure inside a relief valve body is actually quite high during flow, much higher than the 10% we might expect. The codes are reasonable however, they really only intend to limit the back pressure in the vent header to 10% of set pressure. The back pressure inside the valve is of little concern to us. The manufacturer should have already considered that in the design. Hope that helps.

 

[EGT01]

Processeng01,

Here's a few more thoughts for consideration....

First, the 10% backpressure rule is based on a ratio of gauge pressures. In the B31.1 Appendix II example, the backpressure as percent of set pressure is...
(194 - 14.7)/910 = 19.7%
In my copy of B31.1, set pressure is 910 psig.

As I see it, the purpose of B31.1 Appendix II is more for evaluating the forces and stresses on the piping system than it is for sizing the piping. Of course, it does seem a little odd to give an example that does not follow the general rules for relief valve pipe sizing but I don't think anywhere in the example they state the valve is a conventional type valve. I do see that the ASME web site

http://cstools.asme.org/csconnect/CommitteePages.cfm?Committee=N10020200

is listing a tracking# 02-3307, App. II Sample problem correction, SC Proposal, but I don't know if it is related.

For the sake of discussion, let's assume the valve in the example is a conventional type. In that case, I can't offer any other explanation other than what PLeckner and Iainuts have suggested. I would add that for ASME certified relief valves, the capacities are determined according to ASME PTC 25. The test arrangement shown in PTC 25 for a valve tested with steam and discharging to atmosphere is essentially like that in the B31.1 example, basically an elbow attached to the relief valve outlet. So if your plant installation is the same as the test arrangement used to certify the relief valve, then I think you wouldn't have any additional concerns for backpressure.

The following comment is no different than what Iainuts pointed out but, note that choked pressure (P1) evaluated at the outlet of the elbow in the B31.1 example has a backpressure (alone) greater than the 10% rule allows. What this means is that even with zero amount of discharge piping, the P1 pressure will exist at the relief valve outlet flange and the exit from the body of the valve becomes a choke point. Even if you were to provide an expansion fitting at the relief valve outlet flange, the outlet from the valve will always potentially be a choke point. For additional information about this condition, the AIChE/CCPS publication, "Guidelines for Pressure Relief and Effluent Handling Systems" includes a discussion on the subject.

As additonal reference to the effects of back pressure, have a look at the Anderson Greenwood Technical Manual

http://www.tycoflowcontrol-na.com/getlit.asp?lit=ld/Tech

Sem Manual.pdf
Section IX shows the effects of built-up back pressure as it increases above that equal to the overpressure at 10%.

 

[EGT01]

Just realized that the link to the AG Tech Manual didn't paste correctly. Here's a better one...

http://www.tycoflowcontrol-na.com/ld/Tech Sem Manual.pdf

 

[processeng01]

Dear All,

Thanks for your input. The literature indicates for high backpressure at the SV outlet flange, one should consider bellow or balanced spindle type. I am just wondering if Crosby HCI, HCE models and Consolidated's 1700, 2700 series (which are typical steam SVs) are what the literature calls balanced spindle type. They do not look like bellow type. Most of them have a vent hole close to bonnet which discharges steam during lifting. I think that is for making sure that the pressure above seat is always close to atmospheric. Am I making sense or totally confused?

 

[EGT01]

Processeng01,

The type of questions you are asking are best asked of the valve manufacturers. Contact your local supplier, I'm sure they will be glad to assist you with any technical questions about their products.

Of course it doesn't hurt to ask those questions here, and maybe there is someone who has experience with those valves in particular. I can't say I have experience with either valve model but from what I see in their product brochures, these valves are backpressure assisted closing.

The manufacturer info for the Consolidated valve as found in their SV-1 catalog does describe that the vent is associated with the backpressure assisted closing. That documentation also indicates a back pressure limit of 25% of set pressure but I don't see anywhere that they refer to the valve as a "balanced" type valve. Again, your best choice is to discuss these features with the valve manufacturer to make sure you understand them.

If you are interested, you can download the Consolidated catalog here

http://www.consolidatedvalve.com/internet/pages/documentwarehouse/documentsearchresults.cfm

look for "SV-1: Consolidated Safety Valve Catalog" but I'll warn you it is a large file, about 15 mb.