Safety Valve Size and Piping Connections

[Sharik]

A safety valve with a socket welded inlet connection, indicates an inlet size on the valve nameplate of 0.75". How does this work? If the inlet piping is 0.75" NPS, Sched 80, the pipe ID is 0.742". But if the inlet piping is 0.75", Sched 160, the pipe ID is only 0.619". Wouldn't the Sched 160 piping create an inlet flow restriction?

 

[gerhardl]

Hi Sharic,

You are in principle right. As any restriction in any piping will give a resistance, this will of course also give a resistance and a reduction of flow. Here it would however, at a normal fluid speed in the pipeline, probably not be significant or measurable at all.

A question is also the nominal smallest throughflow opening of the valve, which could be much less than the nominal piping connection opening. Actual capacity of the valve is, as known, calculated on basis of pressure difference before and after the valve, and limited by max. acceptable velocity (speed limitation well below sound of air / noise limits) for the valve, thus giving the nominal necessary seat orifice of the valve.

Please also remember that the pressure on the seat (giving the lifting opening of the valve) is not dependent on the size of the connection. Only if the flow is considerably disturbed by a too restrictive opening will the capacity and operation be disturbed.

The main and most common operational disturbance of both PSVs and check valves is the other way around: a too large orifice, giving a 'gulping' action.

 

[Sharik]

Thank you Gerhardl. I understand calculating the actual flows capable in piping to the PSV. But what about ASME Section VIII, Div. 1, Paragraph UG-135 (b)(1) which requires the cross sectional area of the piping to be at least equal to the cross sectional area of the PSV inlet?

 

[gerhardl]

Sorry, I am not familiar enough with the standard to comment on this one!

I read your question to be how absolutely one should require the actual inner diameter/area of inlet piping to be 'equal' to the inlet diameter of the valve if the two are nominally equal, but real piping somewhat slightly less due to piping schedule.

For larger valves this will be largely a theoretical question, but for smaller valves it could of course in some cases have a measurable difference.