Why is Orifice Beta in between 0.3 to 0.7?

From Where do we get this Numbers and What significance do they have that we must keep the Orifice Beta ratio in orifice calculation in between 0.3 and 0.7 Values.

Orifice Beta Ratio

Let us start with What is BETA ? then Lets get rebellious and take “Extreme cases”!!

BETA =  I.D of Orifice (BORE of Orifice) / O.D of Orifice (I.D of Pipe)

Orifice Beta

For Understanding purpose .Lets get rebellious and take “Extreme cases”

What if we take a BETA of 1

This is Possible if I.D of orifice = O.D of Orifice

But that means There is no restriction to flow so no Pressure Drop .

And No pressure drop hence Flow measurement is not possible

What if we take a BETA of 0.9

This is Possible if I.D of orifice approximately near O.D of Orifice

But that means There is very little restriction to flow so very little Pressure Drop .

And as low pressure drop will be difficult to measure so we don’t go beyond 0.7

Also “UNCERTAINTY INCREASES“, this concept is explained below.

Very High Orifice Beta Ratio

What if we take a BETA of 0.1

This is Possible if I.D of orifice is very small as compared to O.D of Orifice

But that means There is very High restriction to flow so very High Pressure Drop.

Very Low Orifice Beta Ratio

This is Good Right ? High Pressure Drop means it will be easy to measure ?

This is not the only consideration we have !!

It affects a variety of things ( Here’s Few reasons listed below)

1) Process Hydraulics

This Affects Process Hydraulic calculation and also leads to Inefficiency in process as a whole .

As the PUMP will have to PUMP MORE !! As a result Process people in their Datasheet specify maximum allowable Pressure loss in Orifice .

What is the use of we purposefully creating more restriction and then trying to compensate it with more pumping power!

2) More Probability of Cavitation and Flashing

The more we drop the pressure the more is the chances of liquid to reach “Vapor Pressure”

And thus more is the chances for “Cavitation and Flashing”

NEXT important consideration

3) Coefficient of Discharge (Uncertianity)

The estimated uncertainty of the empirical coefficient of discharge for concentric, square-edged, flange-tapped orifice meters that are in compliance with this standard (API 14.3.1) is a function of the Reynolds number and the diameter ratio (β)

i.e:- Cd (Uncertainty) is a function of Reynolds Number & Diameter ratio

( Reference API Manual of Petroleum Measurement Standards Chapter 14.3.1 PAGE NO 31 )

At very high Reynolds numbers (approx. greater than 1,00000) the uncertainty is only a function of the diameter ratio (BETA) (β)

So Here’s Where Beta Ratio again comes into picture

Below is graph from API Standards Chapter 14.3.1 that has Percent Uncertainty Vs Beta Ratio (See Comments in Red)

Coefficient of Discharge

So from this graph we can deduce it is better to come to a nearby moderate value of BETA approx.=0.5 for better accuracy !!

Lastly one unanswered question ?

Why only between 0.3 to 0.7

There is no hard and fast rule it is just due to above reasons we try to stay in the moderate range, this limit is established by “Empirical” Method (That is by experimental means, we have come to this value) .

Some standards say we should be between 0.2 to 0.6

None the less ,stay Safe and select a moderate value

Thanks for Reading !!

PS: This is as per best of my current understanding !!

Author : Asad Shaikh

Profile : Linkedin

Articles You May Like :

When to use Diaphragm Seal ?

Different Types of Orifice Taps

Selection of Flow Instruments

Valve Relation between Cv and Kv

Safety Instrumented System Failure

Don't Miss Our Updates
Be the first to get exclusive content straight to your email.
We promise not to spam you. You can unsubscribe at any time.
Invalid email address

1 thought on “Why is Orifice Beta in between 0.3 to 0.7?”

Leave a Comment