
                                  MEMORANDUM


Tetra Tech, Inc.
10306 Eaton Place, Suite 340
Fairfax, VA 22030
phone	703-385-6000
fax	703-385-6007

TO:	Paul Shriner and Jan Matuszko, EPA
FROM:	John Sunda (SAIC) and Kelly Meadows
DATE: 	June 15, 2009

SUBJECT:	Review of Existing Cooling Tower Optimization

EPA requested a review of existing cooling tower installations to examine the degree of optimization (i.e., minimized makeup withdrawals) exhibited by these systems.

Estimates of theoretical make-up volumes in gpm/MW expected for different types of power plants and different cycles of concentration (COC) were developed.  The COC is the cooling tower variable that has the greatest influence on make-up volume requirements.  A COC of 1.5 is a design maximum for saltwater.  For freshwater systems, a COC of 1.5 is considered the baseline below which the system should be considered to be operating poorly with respect to flow reduction.  A COC of 4.0 or above should be considered as optimized.  Estimated flow values for a COC of 2.0 are included as a midpoint between the two.  Attachment A contains an analysis that shows the calculated gpm/MW for each closed-cycle facility that completed a detailed questionnaire for both DIF and AIF.  These values are shown in comparison to the calculated make-up volumes for selected COC values of 1.5, 2.0, and 4.0.  Note that other process flow such as requirements for flue gas desulphurization could add up to 2.0 gpm/MW to the expected values but this analysis does not attempt to adjust for that.  Also included are two graphs that show the make-up volumes graphically for different plant types and how the make-up volume varies with COC for an example tower.

Excluding the partially closed-cycle systems and cooling ponds, there are 20 intakes. Seven operate in an optimized manner, seven operate in between a COC of 4.0 and 1.5, and six operate with a COC less than 1.5.  Only 2 facilities could be considered outliers (greater than twice the COC = 1.5 flow) and one appears to be misclassified in the database.

