Research Project in Support of ASHRAE Standard 195: Method of Test for Rating Air Terminal Unit Controllers

The two main goals of this research project were (1) to validate ASHRAE Standard 195 “Method of Test for Rating Air Terminal Unit Controls” as a reasonable and fair method for evaluating the controllable minimum of VAV boxes and VAV box controllers and (2) to use Standard 195 to determine the controllable minimum for some commonly used VAV boxes and VAV box controllers.  The controllable minimum of a VAV box (195 calls this the “system-under-test”, or SUT) depends on the stability and accuracy of (1) the flow probe provided by the VAV box manufacturer and (2) the stability and accuracy of the pressure transducer and algorithms provided by the VAV box controller manufacturer.  Previous research projects concluded that the flow probe is extremely stable and accurate, even at very low flows, and that the controller is primarily what determines the controllable minimum of the overall SUT.  Standard 195 does not have a way to distinguish between flow probe error and controller error when rating the SUT. 

As part of this research, we developed a way to distinguish between flow probe error and controller error. We found that the flow probes of the two boxes we tested (Price and Nailor) are indeed stable and accurate even at very low flows, but so are most controllers and that the magnitude of the flow probe error is similar to the magnitude of the controller error for most controllers. Therefore, we developed a new methodology for rating the controller separately from the SUT.  We found other shortcomings in Standard 195 that we addressed with our new methodology including:

• We reduced the overall testing burden by reducing the number of samples recorded during steady state and combining the separate accuracy and zero-drift tests into a single test.
• We introduced a way to gather more useful data on the performance of the controllers by running the accuracy tests across a broad range of flow probe signals, instead of individually selected setpoints. 
• We identified flaws in the 195 accuracy and stability test procedures.  In the 195 accuracy test the controller modulates the damper to achieve the target airflow setpoint, with the reference airflow sensor determining how closely the controller holds the setpoint.  But if the controller has a wide deadband within which it will not move the damper, then the tester can game the results by tweaking the fan speed to get the reference airflow closer to setpoint without triggering the controller to move the damper.  Therefore, in our proposed accuracy method the damper is fixed at full open and accuracy is determined by comparing the controller’s airflow reading to the reference airflow sensors.  A similar flaw was identified in the 195 stability test which allows the tester to tweak the fan speed to game the result.  To eliminate this flaw, our proposed stability method requires an automated fan speed controller that must follow an exact sequence of speed changes.

We found that the results of the accuracy tests we ran under the new methodology were repeatable, logical, and fairly represented the accuracy performance of each controller we tested. We also found that our method for separating the controller error and probe error out of the SUT error was validated, as the SUT error was equal to the sum of the controller error and probe error for nearly all flow points tested. Unfortunately we did not identify the stability test flaw in time to test our new proposed stability test method. As such, the stability test results we present in this report are in part a reflection of a test method that relies on the tester to manually select the right flow within the flow deadband, as opposed to a reflection of the controller’s stability performance alone. 

A useful way to describe the controllable minimum for an SUT or for a controller is in terms of the lowest flow probe signal (Pvm) that results in the SUT or controller passing all the subtests in a particular Method of Test (MOT).  With this lowest passing Pvm a designer can easily calculate the lowest controllable velocity (FPM) and CFM from the box manufacturer’s published K-factor.  Table 1 shows the lowest passing Pvm for the 6 controllers we tested using the 195 method (Method 1), our proposed new accuracy method (Method 3) and an interim working version (Method 2). For each of the lowest Pvm-SUTs Table 1 also shows the corresponding minimum velocity for a typical K-factor and the minimum % flow rate based on a typical engineer’s VAV box max flow rate.

Table 1: Lowest Passing Flow Probe Signals (Pvm)

Aside from improving upon the Standard 195 test method and determining the controllable minimums for the controllers we tested, our study also found that the VAV box flow probe’s nominal, calibrated, and instantaneously measured K-factors tend to vary widely. While the controller is operating, the instantaneously measured K-factor can be over 10% greater than the nominal K-factor and 5% greater than the calibrated K-factor. These gaps in K-factor may slightly increase with higher ambient temperatures and lower flows in some cases, more investigation is needed to determine more definitive factors that contribute to the discrepancies we observed.   

Our proposed revised test method evolved over the course of the project.  Method 1 is the current Standard 195 method.  Method 2 and 3 include our proposed accuracy test method but use the current 195 stability test method.  Method 4 includes the same proposed accuracy test method as Method 3 and also includes our proposed stability test method.  Methods 1, 2, and 3 are described in more detail in this report. Method 4 is included in an attachment to this report.  It is formatted as a proposed modification to the current 195 method using the ASHRAE strikethrough and underline formatting. 

While we did not have the opportunity to fully vet the changes to the stability test portion of Method 4, we are confident that Method 4 in its entirety is a reasonable and fair method of test for rating the accuracy and stability of an SUT and for isolating the performance of just the VAV box controller.

We plan to work with ASHRAE to update Standard 195 and to work with Title 24 and ASHRAE Standard 90.1 to require Standard 195 testing of VAV box controllers and potentially require a maximum passing controller-only Pvm.