Multiple-Measurement Accuracy

1. Apr 14, 2010

Staff: Mentor

I ran into an issue on a project where I'm needing to be sure of the accuracy of airflow measurements being taken. There are two basic ways to measure the airflow of an air conditioning system:

1. First is to traverse the main supply duct with a pito-static tube array and a differential pressure gage and measure the whole system airflow at once. These measurements have an accuracy of 2% with a 95% confidence factor. This is a typical rated accuracy for such sensors/gauges.

2. Second is to measure the airflow at each airflow control device, which has a very similar piece of equipment (a pito-static array with a pressure transducer) permanently installed - and add-up the numbers. The accuracy of each device is then about the same as in #1. For this example, I have somewhere around 100 of these devices. What is the accuracy of this airflow measurement?

It seems to me that method #2 should yield an accuracy far better than of 2%, but I'm not sure and I don't know how to calculate the accuracy. Can anyone help?

2. Apr 16, 2010

winterfors

If your measurement errors at each pressure transducer is independent of the error at the other transducers, then the total relative variance will be 100 times smaller and the accuracy (confidence interval) thus sqrt(100) = 10 times smaller (= 0.2% accuracy with 95% confidence)

3. Apr 16, 2010

Staff: Mentor

It is.
Ok, simple enough: so the accuracy is just the square root of the number of measurements times the accuracy of each individual measurement (assuming they are all of the same accuracy).

Thanks very much.