Dec-2022

Cooling tower acceptance testing

In a recent blog, August 8, 2022, the subject of an independent check-rate or rating of a cooling tower’s thermal design was discussed.

Allen Feltzin
Becht

Article Summary

This enables a determination of whether or not the stated or quoted fan horsepower for given design conditions is based on acceptable data and whether or not the resultant horsepower prediction is reasonable. It is a calculation “on paper” and may precede the construction of a new tower or a retrofit/upgrade to an existing tower.

Once the tower is erected however, a new dimension is added. The tower “as built” will perform in part, according to the above criteria, and in part, due to the care and attention exercised during construction. For example, if the fill or packing is not fitted properly at the end walls or around structural members, air bypassing will affect performance. If the fan tip clearance to the shroud is excessive, the fan(s) will not perform at their design efficiency. It therefore becomes necessary to conduct a performance test or acceptance test to verify that the tower will deliver the cold water temperature at the rated horsepower.

One exception to this is with small package cooling towers. In some cases the manufacturers have certified certain designs with the Cooling Technology Institute (CTI) and tests have already been conducted to document performance. The end-user or customer should verify that the particular package tower selected has, in fact been certified; then an acceptance test is generally not necessary.

This is not the case for field erected towers however. Therefore, it is important, perhaps imperative, that an acceptance test be conducted. Owner-operators may wish to include this in the terms and conditions of a purchase order, or even in an RFP. The procedure is given in CTI ATC-105. There are two basic techniques described in the CTI document: the Characteristic Curve method and the Performance Curve method. Although they are both related to the intrinsic design, one or the other is selected as the basis for the test. In this discussion, the Characteristic Curve method will be described, along with some other requirements pertinent to the test.

To begin with, the conditions during the test have to be within certain limits of design conditions.

Wet-bulb temperature     Â± 8.5°C (15°F)
Dry-bulb temperature*     Â± 14.0°C (25°F)
Range    Â± 20%
Circulating water flow    Â± 10%
Barometric pressure     ± 3.5 kPa (1”Hg)
Fan driver output power     ± 15%

As an example, a tower cannot be acceptance tested at a 50ºF wet bulb if the design wet bulb was 75ºF. The results would be invalid.

Suitable instrumentation is also described in the test procedure.

Characteristic Curve Method
To illustrate, the same design conditions will be utilized that were given in the August 08 discussion. The demand curves with a fill characteristic curve are reproduced in Figure 1.

The design L/G was 1.047 and the corresponding KaV/L was 2.603 for a 7ºF approach to the 65ºF wet bulb.

The test L/G is determined from the water flow rate and fan horsepower at the time of the test. Let us assume that the result gave an L/G of 0.90. The test KaV/L is obtained by numerical integration (the method will not be described here but is illustrated in ATC-105). The result is a KaV/L of 2.500. These two coordinates are plotted as shown in Figure 2.

A line segment is then drawn through the test point parallel to the original fill characteristic curve as shown in Figure 2. An L/G is determined at the point where the test line segment intersects the design (7ºF) approach curve. It will be seen that this occurs at an L/G of 0.95. This represents the value of L/G necessary to actually produce design cold water when the tower is operating at design conditions. Note that compared to the original design L/G of 1.047, L/G actual is smaller than L/G design, meaning that for constant L, G has to be greater. More air would need to be drawn through the tower to produce the design cold water temperature. Therefore, the test indicates that the tower is operating at less than 100% capability.
The capability is given by the ratio of the two values for L/G

(0.95/1.047) x 100 = 90.7%.

As explained in the August 08 discussion, there is a cubic relationship between airflow and power: it takes roughly 30% more horsepower to pull 10% more air through the tower (in other words, G increases by 10%), other things remaining constant.

In this particular case, it would take approximately [1/ (0.907)] ^3 or 34% more horsepower (if in fact that could be obtained). There could be limitations on motor amps or fan pitch angle.

Who should perform a cooling tower acceptance test? There are several CTI certified agencies that could conduct an “official” test. It may be agreed with the tower manufacturer that he could perform an unofficial test and only resort to a certified test if the results are unacceptable. If the latter course is chosen, it is still important to ensure that the test is conducted with proper instrumentation and within the test limits mentioned previously.

Towers with plastic film fill need to be “conditioned” before testing. The recommended period is 1,000 hours of operation.

In conclusion, a cooling tower acceptance test provides valuable information about whether the cooling tower as built will deliver the design cold water temperature at the design (stated) fan horsepower when operating at design conditions.