Orifice design for pressure drop

pjo

When designing an orifice plate to reduce the pressure of a constant flowing liquid, is it best to have one large hole in the center or multiple smaller holes?

One example: Pipe diameter = 10 inches, Flow = 2800 gpm water, pressure drop = 150 psi, temperature = 115 F. The hole diameter needs to be about 6 inches.

If many smaller holes are better, how are they sized? Do you have there cross-sectional areas add up to the cross-sectional area of the 6 in hole?

Thank you.

stewartcs

Never really thought about it much to be honest. However, I imagine that the smaller the hole the larger the pressure loss. That is, it wouldn't be the same as adding up the area of the small holes so they equal the same area of the large hole.

An additional consideration would be if the smaller holes became blocked over time due to the fluid flow (i.e. scale or some type of build up).

CS

Studiot

Without details of the orifice purpose it's difficult to comment.

BS1042 gives details of hydraulic orifices for measurement purposes.

Further discussion of the equations may be found in this paper

http://www.publish.csiro.au/paper/EA9690449.htm

I am sorry there must be American equivalent standards but I don't know them.

Jobrag

It has to be a single hole, the downstream pressure is the pressure in the vena contracta if you had multiple holes you would be measuring the correct pressure. The American standard is ASME PTC 19.5

Harry Hazard

I'm not sure about the standards, but a large single hole or several smaller ones should produce about the same result.

Bernoulli's principle tells us that as the flow area decreases (at the orifice), the velocity increases and the pressure decreases, assuming the fluid is incompressible. If the area of the large hole and the "total" area of the smaller holes are the same, then the pressure drop at the orifice should be the same.

Of course, nothing ever works exactly like theory.

minger

As Jobrag said, you want it to be a single hole. With multiple holes, you will get uncertainty in the pressure drop from some holes being in a different flowfield than the rest, in addition to effects of one hole on another.

From a more practical note, as Jobrag mentioned, the ASME orifice sizing specs are written for a single hole. They already did the hard work for you.

Studiot

I am guessing but since the OP has not come back this is not a real design exercise but a coursework or book question. I am not sure why the pressure reduction is required.

However others have chosen to discuss the issue so here are my thoughts. This is not a simple Bernoulli application.

The relationship between pressure drop and flow rate is not calculable with sufficient accuracy for measurement purposes. Flowmeters based on this principle have to be individually calibrated. However we can estimate ball park figure, which could be accurate enough to, for instance, compare with a finite element flow model.

Whatever, The more holes one has the greater the circumference to area ratio or the more circumference one needs to provide a given area. This alters the friction and in turn the Reynolds number and the discharge coefficient.
As has aready been pointed out the flow regime is not constant across a pipe section and so the distribution of the holes would play a part. It is also likely that if they were too close together the local flow regime near one hole would influence its neighbours.
However, offset and non circular holes are standard practice details can be found in the reference below.

All such meters have a 'discharge coefficient'. This is the ratio of actual discharge to the theoretical. For a single hole plate it is around 0.65. Orifice plates have the advantage of flatter characteristics with flow rate.

An engineer wishing to answer some of these questions would need to use the temperature to look up the viscoscity in standard tables. Using this and the pressure drop he could then calculate the Reynolds number and armed with this could enter more standard tables to look up the discharge coefficient. Using this he could derive a flow rate.

Further information and a standard calculator is available at

http://www.flowmeterdirectory.com/fl...ice_plate.html