I saw a discussion on air flow measurement in a combustion application over at ISA Safety email list. I wanted to share the question and the answer provided by DP Flow specialist, Vernon Binger.

In a boiler installation is a combustion air duct. The air duct is not build correctly with enough strait length and equipped with a venturi dP measurement transmitter, but the measurement is not correct, because there is an unstable air stream.

Now the contractor wants to install an annubar flow measurement. The annubar flow measurement consist of a pipe with different holes in it across which a trust arises which causes a dP that is measured by a transmitter.

Question is now is this air flow measurement principle adequate enough to be used in a SIF?

Has anybody some experience with this type of measurement and can it be used in a SIF?

Here was Vernon's response:

The Rosemount model 485 Annubar, with it's "T" shape, generally creates 40 to 100% more DP than other types of area averaging pitot devices. The Rosemount Annubar is often used in air duct type applications, even in exhaust stack applications.

It's "T" shape design is considered to be resistant to plugging from particulate in the flow stream. However, it is of course not impossible to plug it, in extreme dirty flow streams. In some applications customer's install an intermittent purge system to blow out the Annubar with a source of higher pressure air, or clean N2. The frequency of the intermittent purge cycle would depend on the application, sometimes once per day, sometimes once per hour is required.

But, if the issue also involves a lack of straight duct at the required installation point, that may also not be an ideal application for and Annubar. We would need to see the size of the duct and some sketch depicting the upstream and downstream shape of the duct. In some situations perhaps multiple Annubars spanning the flow profile at different locations can provide a better overall average indication of the velocity in the duct. It will be helpful to know the expected duct pressure and temperature, as well as the required min, norm, and max or full scale flow rates required.

Regarding the venturi primary element which is not providing a reliable measurement, usually venturi primary elements do not require a significant amount of straight duct in order to provide a usable flow indication. The most common error with venturi installations is that they were sized with too large of a throat diameter, or beta ratio. Often the venturi is sized to provide some given DP at the expected normal or full scale flow rate, but no adequate consideration is given to the minimum flow rate that will be required for the end user's application. Then at some min flow condition, there is simply not enough DP created, and the DP instrument is not able to provide a reliable flow indication. The beta ratio should have been smaller to provide at least 0.5" H2O (0.124 kPa) DP even at the lowest flow rate for which a reliable measurement will be required.

Does anyone else have anything to add?