Orifice Plate Pressure Recovery 1

[mjpetrag]

I have a pipe with an orifice, with the upstream orifice pressure at 60 psig and the downstream pressure is a tank open to atmosphere. There is a 4' long run of pipe downstream of the orifice before it hits atmospheric pressure (about 3.5 psi loss due to length). So is the delta P for orifice plate sizing just the 60 psig - 3.5 psig? Or do you have to factor in the pressure recovery, which is about 16% recovery of the dP? In that case the dP would be 60 - [(60-3.5)*.16+3.5] = 47.5 psig.

I'm a little confused about pressure recovery

-Mike

 

[pmover]

without knowing the design or actual installation, the orifice dP is measured at taps located within a certain distance from the orifice plate (i.e. for flange taps, the taps are 1-inch upstream & downstream of orifice plate). so the actual measured dP will likely be determined by the orifice plate hole diameter relative to the pipe size.

pressure recovery likely not applicable in this installation/situation as the downstream portion of the orifice meter is open to atmosphere. the pressure drop will be greatest across the orifice plate.

what is the purpose of the orifice plate? from 60-psig to nearly atmospheric pressure across the orifice plate, i fail to understand how an accurate or reasonable flow measurement can be achieved.

a little more information about the design, installation, and purpose of the orifice plate will be helpful . . .

meanwhile, good luck!
-pmover

 

[BigInch]

Seems like you want to keep from filling the tank up too fast and trying to do that without using a control valve at the pipe outlet.

It takes a long enough pipe length downstream for the fluid to decelerate, after having just speed up to go through the smaller hole in the orifice plate, to recover any pressure. 4 ft might not be enough. The pressure recovery is theoretically limited to difference in velocity heads at pipeline velocity and at orifice velocity. The trick is to hold the pressure constant enough downstream of the orifice plate so that further fluid acceleration is not allowed and the pressure CAN be recovered. That would normally be accomplished in say around 15 pipe diameters for typical D_orifice/D_pipe ratios, but if that ratio is lower, progressively longer lengths downstream would be required for the fluid to slow to pipeline velocity and recover any pressure lost at the plate. You don't mention the diameters involved, so we don't know what length of pipe might be needed to get full recovery.

You could ingore the pressure recovery, if you were interested in getting the maximum flowrate into the tank.

Don't forget that if the tank were ever to be sealed off from atmospheric pressure, the pressure drop across the orifice would go to zero as flow eventually stopped and you would see full pipe system pressure in the tank, or hopefully only the tank's relief valve pressure.




"I am sure it can be done. I've seen it on the internet." BigInch's favorite client.

"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.live.com

 

[mjpetrag]

The pipe ID is .824" and the orifice hole is 0.33". The thing is I don't know how much length it takes to recover. I can't find anything on that.

This is a minimum flow bypass and the 60 psig is deadhead conditions.

-Mike

 

[mjpetrag]

Also, I made a mistake the downstream length of pipe from the orifice is 70" with 3 90 bends.

-Mike

 

[BigInch]

Unit mass of fluid pushed by the pressure of 13 psig that you will supposedly have downstream of the orifice plate; slowed by virtually no backpressure at the pipe outlet and only a short length of pipe friction, will pretty much say no further deceleration is possible, since those don't appear to add up to 13 psi, so some further acceleration might be expected. Don't expect any pressue recovery unless you can somehow hold a backpressure near what you will have at the outlet from the orifice.

"I am sure it can be done. I've seen it on the internet." BigInch's favorite client.

"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.live.com

 

[mjpetrag]

Thanks BigInch, it makes sense now.

-Mike

 

[BigInch]

Great!

Reduce the problem down until it makes sense. Thinking of a cube of fluid as a dynamic particle exposed to F = M * A moving from V1 to V2 with a little bit of Bernoulli thrown ain usually works pretty well.

"I am sure it can be done. I've seen it on the internet." BigInch's favorite client.

"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.live.com