I have found variations in published documents of the Cooling Tower Curves used in DOE2.

Specifically I am looking for the Cooling Tower formulas used in the current version of DOE2 to calculate the two variables below:
FRA = TWR GPM FRA
FWB = TWR GPM FWB

The DOE2 Engineering Reference Manual 2.1A and subsequent Supplement 2.1E are where I am finding conflicting information about the variables f(x,y) to use for each.

They are both Bi-Quadratics and the (a-f) coefficients are clearly and consistently stated in DOE2 documents.

z = a + bx + cx2 + dy + ey2 + fxy

One section of the Supplement 2.1E (copied below) explains the function as:
FRA = f(Trange, Tapproach)

TWR-GPM-FRA
accepts the u-name of a CURVE-FIT instruction that defines a
bi-quadratic equation. That equation will be used to express an
intermediate variable which is a function of the range and
approach. The intermediate variable is in turn used in the
curve TWR-GPM-FWB.

And another section of the Supplement 2.1E (copied below) explains it as
FRA = f(Trange, FRA)

COOLING TOWER
TWR-FAN-FPLR A1RCEL 0.33162901 *0.88567609 0.60556507 0.94848233 0.00000000 0.00000000 TWRFAN CUBIC
TWR-GPM-FRA RNG,FRA *2.22888899 0.16679543 *0.01410247 0.03222333 0.18560214 0.24251871 GPMRA B1*QUAD
TWR-GPM-FRB FRA,OWB 0.60531402 *0.03554536 0.00804083 0.02860259 0.00024972 0.00490857 GPMRB B1.QUAD
TC-CHLR-CAP-FT Teond,Tcw *0.351443 0.056583 *0.600054 *0.045625 *0.000043 *0.000012 CCAPT5 B1.QUAD

Which format is used in the actual simulation?

Many thanks,
Noah

Noah Czech, AIA