This manuscript was published by Taylor & Francis on December 29th , 2023 in the

Journal of the American Society of Brewing Chemists.

The published version can be accessed at
https://doi.org/10.1080/03610470.2023.2267947.

Physical Equations Relating Extract and Relative

Density
∗
Josh Buhl

20 7
January 5, 2024

5, 1.
24
y v Abstract

Equations for converting between relative density and percent by mass

extract based upon physically meaningful values (rather than fitted param-
ar int
eters) are derived from physical principles. An exact equation and two
approximate formulas using one and two physically meaningful parameters
respectively are given based on the partial specific volumes of sucrose and
water in solution, which are calculated for sucrose concentrations varying
between 0 and 20 percent by mass fraction. The predictions of these equa-
nu pr

tions are compared to the ASBC tables and the accuracy of the predicted
values is analysed. The root mean squared error over the entire range of
the ASBC tables of the approximate formula with two parameters is less
-

than 0.016°P and that of the formula using only one parameter is less than
0.023°P . The approximate formula with one parameter provides a simple
Ja re

and easily remembered but nevertheless highly exact means for any brewer
to calculate extract without specialized analytical devices.
Keywords: Relative Density, Specific Gravity, Extract, Partial Molar
p

Volume, Partial Specific Volume, Baumé Modulus, Sucrose, Wort

1 Introduction

Due to the ease of measurement of relative density (specific gravity, SG) and the
necessity of knowing the extract value (E, see below.) for pertinent calculations,
∗
email: [email protected]

1
 methods relating these two quantities are frequently used in both commercial and
home-brewing settings.

Originally, tables of empirical data, e.g. Plato et al.’s [10] data from 1900, or
conversions thereof [12], later their ASBC counterpart [2] and for the last few
decades best-fit equations [8][21][23] based upon these data have been used for
relating SG to E. While useful for calculating the desired values, especially for
automated computations, best-fit equations are usually physically meaningless and
difficult to remember. Modeling the system based upon physical principles is more
satisfying and can shed light on the meaning of some best-fit parameters.

20 7
In the following, no differentiation is made between real and apparent extract. For

5, 1.
the purposes of this article ‘E’ simply refers to the percent by mass of extract in

24
degrees Plato (°P) that corresponds per the ASBC tables [2] to a relative density
value (SG) as measured with a hydrometer.

2
y vRelating Relative Density to Extract
ar int
2.1 Review of previous work

Aside from the rule of thumb estimation customarily phrased as “gravity points1
nu pr

divided by 4”

1000 · (SG − 1)
-

E= = 250 · SG − 250 (1)

4
Ja re

relative density is usually converted to percent extract via best-fit polynomial

equations.[8][21][23]
p

The first best-fit polynomial equations for the ASBC’s version [2] of Plato’s [10]
data relating E to SG appear to have been published by Robbins [21] in 1975.
Robbins found a linear equation (see below) and this second-order polynomial for
the ASBC tables:

E = −205.347 · SG2 + 668.7183 · SG − 463.371 (2)

1
“Gravity points” refer to the thousandths of SG above 1, e.g. an SG of 1.040 corresponds to
40 gravity points.

2
 which has a root mean squared error (RMSE) of approximately 1.68×10−3 °P when
compared to the ASBC tables [2].

In 1980 Siebert [23] again applied best-fit methods using equations of various
form to the ASBC tables. In addition to a first-degree polynomial and a “hyper-
bolic equation” (see below), Siebert fits second through fifth-order polynomials,
compares them to the ASBC tables, and concludes that the following third-order
polynomial is sufficient, since it’s results already have zero residual difference to
the tables’s values in the range 1°P – 20°P , although it does differ minimally for
very low extract values in the 0°P – 1°P range.

20 7
5, 1.
24
SG =1.000019 + 3.865 613 × 10−3 · E+
1.296 425 × 10−5 · E 2 + (3)
y v 5.701 128 × 10−8 · E 3 .

This result was refined in 2009 using improved methods by Cutaia et al. [8]:
ar int
SG − 1 =1.308 × 10−5
+ 3.868 × 10−3 · E
nu pr

(4)
+ 1.275 × 10−5 · E 2
+ 6.300 × 10−8 · E 3 .
-

who also found an eighth-order polynomial giving E as a function of (SG − 1),

Ja re

which currently seems to be the most accurate best-fit equation giving E as a

function of SG, having an RMSE of less than 3.57 × 10−4 °P when compared to
the ASBC tables [2]:
p

E =2.569 × 102 · (SG − 1) + 6.7126 · (SG − 1)2

− 1.4482 × 104 · (SG − 1)3 + 5.1758 × 105 · (SG − 1)4
(5)
− 1.0746 × 107 · (SG − 1)5 + 1.3011 × 108 · (SG − 1)6
− 8.5079 × 108 · (SG − 1)7 + 2.3231 × 109 · (SG − 1)8 .

Hackbarth [13] reported a tenth order polynomial also giving (SG−1) as a function
of E as in equation 4 above, but fitted to the AOAC International’s table [15]

3
 relating specific gravity to degrees Brix and degrees Baumé (see below), instead
of the ASBC tables [2].

In between Siebert and Cutaia, Lincoln [18] performed a “trial & error” best-fit
procedure in 1987, using a computer program to guess coefficients and print the
residuals on a computer screen. Lincoln arrived at two “best-fit” equations, which
are not inverse functions of each other:

E
SG = E
+1 (6)

20 7
258, 6 − 258,2
· 227.1

5, 1.
and

24
y v E = (258 − (205(SG − 1))) · (SG − 1).

Whereas only a single reference to equation 6 was found on the internet ([11]) and
(7)

none in the literature, equation 7 seems to be more well-known and is sometimes

ar int
even cited as the “Lincoln equation” [25]. However, multiplying out Lincoln’s
second equation

E = −205 · SG2 + 668 · SG − 463 (8)

nu pr

shows that this is in fact equation 2 with truncated coefficients, so should perhaps
be more appropriately called “Robbins’ equation.”
-

Cutaia et al. [8] confirm Lincoln’s result arriving at the essentially identical second-
Ja re

order best-fit equation with non-truncated coefficients as Robbins

p

E = −463.37 + 668.72 · SG − 205.35 · SG2 (9)

hence actually confirming Robbins’ result but with a slightly larger RMSE.

Robbins [21] also found a best-fit linear equation arriving at:

E = 244.014 · SG − 243.775 (10)

as did Siebert for the inverse relation [23]:

4
 SG = 0.0039347 · E + 0.999930 (11)

which, if inverted gives:

E = 254.15 · (SG − 0.999930) = 254.15 · SG − 254.13 (12)

both essentially confirming traditional wisdom (equation 1).

20 7
Lastly and most interesting for us, an equation which possibly traces it’s origins
all the way back to Baumé [24] in 1768 and his empirically calibrated hydrometer

5, 1.
24
is in it’s classical form

y v m
s= (13)
m±d

where s is the specific gravity, d are the measured degrees Baumé or °Bé, and m
ar int
is the ‘modulus’ of the hydrometer [6][24]. Here, ‘+’ is used for liquids less dense
than water and ‘−’ for liquids denser than water.

The history of this equation was traced back to it’s common use in the 19th and
early 20th centuries [24] for the frequently used Baumé hydrometers of the time,
nu pr

but no original source for nor derivation of the equation itself could be found.
It could also not be established whether Baumé himself reported any equation
when he invented this hydrometer in 1768, although his original work could not
be obtained for comparison. However, equation 13 is easily derived with basic
-

assumptions, see appendix A. It is the constant in this equation, the modulus m,

Ja re

that will be the focus of our attention.

The correct modulus m for Baumé hydrometers seems to have been quite a point of
contention during the 19th and early 20th centuries[24]. The original Baumé mod-
p

ulus is approximately 140 [24], but due to lack of a standard calibration method
there have been many moduli in use and at one point the US Bureau of Standards
simply set the modulus to the fixed value of 145 [1][24], the Baumé equation 13
defining the calibration via this modulus. This is interesting as the modulus has a
definite physical meaning as will be shown in section 2.2 and in appendix A, and
setting the modulus to an arbitrary number removes this meaning.

Equation 13 is used in the classic brewing textbook by de Clerck [6], which appears
to be the indirect source (via Manning [20], see below) for it’s contemporary usage.
de Clerck uses equation 13 without citing a source for it, but his expression of

5
 the equation is similar to the notation used in Circular No. 59 of the Bureau of
Standards from 1916 on Baumé Hydrometers, so it seems to have been a well-
known relation at the time.

de Clerck briefly relates (see pp. 67-68 in [6]) how empirically graduated Baumé
hydrometers were problematic for use because the relationship of their scale to
actual density was in general unknown, and hence they could not be calibrated
by comparing to a pycnometer. Interestingly, de Clerck refers2 to the modulus as
a ‘factor’ for the calculation, a position already taken in a Technological Paper
of the U.S. Bureau of Standards in 1918 [1] where it is referred to as a ‘constant

20 7
multiplier’, implying that it was possibly considered to be a simple constant of the
respective hydrometer.

5, 1.
24
Indeed, throughout both US Bureau of Standards Circular No. 59 [24] and Tech-
nologic Paper No. 115 [1], in which the modulus is discussed at length, no mention
y v
is ever made that the modulus is related to a physical quantity which is not con-
stant, a fact which might have helped explain the point of contention over the
correct modulus presented in Circular No. 59.
ar int
de Clerck states that for “Plato’s saccharometer” (i.e. a hydrometer reading in
°P, which were calibrated at the time for 17.5 ◦ C), the modulus is 260 (see [5] pp.
31-33), which would give
nu pr

260
SG = (14)
260 − E

or in inverted form
-
Ja re

260
E = 260 − (15)
SG
p

but since no source for the modulus 260 was cited, it’s origin could not be traced
back any further.

This equation was later used by Manning citing de Clerck [5] in an article for
home-brewers [20], but with a different modulus which he states to have found to
be more accurate at the 60°F at which he measured SG
2
In both the French original (2nd edition) [6] and the German translation thereof [7] de Clerck
refers to it as such, but interestingly not in the English translation [5], where it was either lost in
translation or not yet in the 1st edition, the 2nd edition apparently never having been translated
into English.

6
 259
E = 259 − . (16)
SG

Manning’s equation is sometimes cited on internet pages for home-brewers(e.g. [16]),

and appears to be the source of this equation in the home-brewing canon, but
without the specification that it’s intended validity is for 60°F [20], whereas most
measurements are currently taken at 20 ◦ C and tables [2], equations [8] and equip-
ment (e.g modern hydrometers) usually assume this temperature, so this equation
is probably often being used incorrectly. A more accurate modulus for use at 20 ◦ C

20 7
is given in section 2.3.

5, 1.
Equation 14 was also found, possibly unwittingly, by Siebert [23] with fitting

24
parameters A and B in the form

y v SG =
1
(A + B · E)
(17)
ar int
which Siebert calls a “hyperbolic equation”.

Dividing numerator and denominator in the de-Clerck-equation 14 by 260 gives

nu pr

1
SG = E
(1 − 260 )
(18)
1
=
-

(1 − 0.0038462 · E)
Ja re

whereas Siebert found the best-fit of the ASBC tables with equation 17 to be [23]
p

1
SG = . (19)
(0.9999679 − 0.0038555 · E)

As will be shown, equations 14 and 19 are related to the physically correct conver-
sion equation which will be derived from physical principles in the next section,
but it could not yet be determined whether equation 14 was originally empirical,
best-fit or also derived from physical principles and approximations.

7
 2.2 Relative Density, Extract and the Partial Specific Vol-
ume

In this section, equations relating SG to E are derived from basic physical defini-
tions, starting with the partial molar volume.

The partial molar volume of solute i gives the rate of volume increase of the solution
per mole of solute i, ni

20 7
 
∂V
Vi = (20)
∂ni

5, 1.
P,T,nj̸=i

24
and the partial specific volume gives the rate of volume increase per unit mass of
solute i, mi
y v  
∂v
vi = (21)
ar int
∂mi P,T,mj̸=i

hence
nu pr

Vi
vi = (22)
mwi

where mwi is the molecular weight of solute i.

-

Naively, one might expect the partial specific volume to simply be the inverse of
Ja re

the density of the solute. Indeed, for the pure substance and in ideal solutions
this is the case, but the partial molar volume of both solute and solvent in general
depends heavily on the intermolecular forces between solute(s) and solvent(s) and
p

upon their respective concentrations. Typically, even at infinite dilution the partial
specific volume is smaller than the inverse density because an individual molecule
of solute will partially fill the naturally occuring space between the molecules of
the solvent and can attract the solvent molecules in closer together than they
otherwise would be. For example, the partial molar volume of water in itself is
approximately 18 mL mol−1 whereas in ethanol at infinite dilution is well-known to
be approximately 14 mL mol−1 . The partial molar volume can even be negative [9],
i.e. the total volume decreases when adding more solute, e.g. some salts at certain
concentrations. Due to the complexity of the intermolecular interactions, the par-
tial specific resp. molar volume is (aside from using complex computer simulations)

8
 essentially incalculable and for all intents and purposes are quantities which have
to be determined empirically for each and every concentration.

Indeed, Plato et al.’s [10] work of measuring the specific gravity of aqueous so-
lutions of sucrose of known concentration can be interpreted as being equivalent
to determining the partial molar volume of sucrose at the corresponding concen-
tration, since the volume Vtot of a binary solution is given by the partial molar
volumes of it’s components [9]

20 7
Vtot = nA · V A + nB · V B . (23)

5, 1.
24
The partial molar volumes can be calculated empirically from a table of (relative)
density versus concentration by standard methods [9]. Example calculations for
the ASBC version [2] of Plato et al.’s tables [10] are performed in Section 2.4.
y v
Now, if ρ is the density of the solution, mA,B the mass of solute A resp. mass of
solvent B then
ar int
mA + mB
ρ= . (24)
Vtot

and hence
nu pr

1 mA VA mB VB
= · + · (25)
ρ (m + m ) mw (m + m ) mw
| A {z B } | {z A} | A {z B } | {zB}
-

=wA =v A wB =(1−wA ) =v b
Ja re

where w denotes mass fraction. Solving for the mass fraction wA gives
p

1
vB − ρ
wA = . (26)
vB − vA

The density ρ of the solution at 20 ◦ C is

◦C
ρ20
 
◦C ◦C air 20 ◦ C
ρ= ρ20
H2 O · SG20
20 ◦ C − 20 ◦ C · (SG20 ◦ C − 1) (27)
ρH2 O

9
 ◦
where SG20 C
20 ◦ C is the apparent specific gravity measured in air as given in the ASBC
tables.[2]

Substituting for ρ and setting A = s for sucrose and B = w for water gives

1
vw − 20 ◦ C
!
◦C ◦ C ρair 20 ◦ C −1)
ρ20
H2 O · SG20 −
20 ◦ C ρ20 ◦ C
·(SG20 ◦C
H2 O
ws = (28)
vw − vs

20 7
as the conversion equation for calculating the extract as mass fraction of solution
from the specific gravity measured in air. Neglecting the effects due to air (i.e.

5, 1.
24
setting ρair = 0) gives

1
y v vw − ρ20
◦C
·SG20
◦C
H 2O 20 ◦ C
ws = (29)
vw − vs
ar int
and reducing the fraction by v w gives

1
1− v w ·ρ20
◦C
·SG20
◦C
H2O 20 ◦ C
ws = vs . (30)
1− vw
nu pr

As will be shown in section 2.4, the average value of v w in the range 0 ◦ P to

20 ◦ P is v w ≈ 1.0016 mL g−1 whereas the reciprocal density of water at 20 ◦ C is
(see appendix B) 1/ρw = 1/0.998 207 35 g mL−1 ≈ 1.0018 mL g−1 , so the partial specific
-

volume of water in dilute sucrose solutions is indeed close to it’s reciprocal density,
Ja re

◦C
and hence v w · ρ20
H2 O ≈ 1. Noting also that E = 100 · ws reduces this equation to

1
1 − SG
p

E = 100 · (31)
1 − vvws

as a simplified version of equation 28. Inverting this equation and solving for SG
gives

1
SG = (1−vs/vw )
(32)
1− 100
·E

10
 As will be calculated in the next section, the average value3 for the ratio of the
partial specific volume of sucrose to that of water in the range from 0 ◦ P to 20 ◦ P
at 20 ◦ C is vs/vw ≈ 0.617, which gives

1
SG = (33)
1 − 0.00383 · E

and should be compared with equations 18 and 19 above.

Hence, the modulus in the standard Baumé conversion equation 13 is essentially

20 7
100 divided by 1 minus the ratio of the partial specific volume of the solute to that
of the solvent at the temperature for which the relative density is measured, in

5, 1.
24
particular the known Baumé modulus for “Plato’s saccharometer” [5] at 17.5°C is

y v 260 ≈
100
1 − vs/vw
(34)
ar int
and Siebert’s fitting parameter B in equation 17 is approximately the reciprocal
of this value, i.e.

1 − vs/vw
B = 0.0038555 ≈ . (35)
100
nu pr

This explains why the fairly simple equation 13 resp. 15 with just one constant gives
surprisingly accurate results. It’s actually the correct equation with three very
good approximations. After a careful search of the literature4 no other derivation
-

of equation 26, equation 28 nor equation 31 could be found.

Ja re

2.3 Approximate formulas

p

Noting that 100

1−0.617
≈ 261 means that an approximate formula at 20 ◦ C should be

1 261
E = 261 · (1 − ) = 261 − (36)
SG SG
3
Fortunately, the partial specific volume of sucrose is remarkably constant in the useful range
from 0 ◦ P to 20 ◦ P varying in this range by just 0.007 mL g−1 (see table ??.)
4
However, as previously stated, Baumé’s original work published in 1768 could not yet be
obtained for comparison.

11
 and this equation does reproduce the ASBC tables with a root mean squared error
(RMSE) of only 0.0331°P (see Table 1).

However, aside from approximating the non-constant partial specific volumes with
their averages in equation 33, the second main approximation used to derive equa-
◦C 20 ◦ C −1
tion 31 was assuming v w · ρ20 H2 O ≈ 1. Using ρH2 O = 0.998 207 35 g ml (see ap-
pendix B) and the median5 value v w 1 = 1.0016 mL g−1 as calculated in section 2.4
2
◦C
H2 O ≈ /0,999804 ≈ 1.0002. Using this value and the median values for
gives 1/vw 1 ·ρ20 1
2
v s/v w as calculated in section 2.4 v s 1 /v 1 = 0.61747 ⇒ 100/1−0.61747 ≈ 261.42 in
2 w2
equation 30 gives

20 7
5, 1.
1.0002 261.47

24
E = 261.42 · (1 − ) = 261.42 − (37)
SG SG
y v
and indeed this equation reproduces the ASBC tables with an RMSE less than
0.016°P and a mean absolute error (MAE) less than 0.0125°P , but does have a
larger error of -0.05°P to -0.02°P in the low 1.000 – 1.011 SG range as shown in
Table 1.
ar int
Solving equation 37 for SG gives

1
nu pr

SG = 1 1
( 1.0002 − 261.47
· E)
(38)
1
=
(0.9998 − 0.0038245 · E)
-
Ja re

and should again be compared to equation 19. Hence Siebert’s fitting parameter
◦C
A in equation 17 is A ≈ v w · ρ20
H2 O and its deviation from unity is a measure of the
deviation of the partial specific volume of water in dilute sucrose solutions from its
p

reciprocal density. Further, Siebert’s fitting parameter B in equation 17 is more

specifically

◦ ◦
ρ20 C · v w vs ρ20 C
B ≈ H2 O · (1 − ) = H2 O · (v w − v s ). (39)
100 vw 100

If only one constant is desired, then using vs/vw ≈ 0.616 ⇒ m = 100/1−0.616 ≈ 260.4 6
5
Using the median values v s 12 and v w 12 instead of the mean values v s∅ and v w∅ reproduces
the ASBC tables with lower MAE and RMSE.
6
Actually all moduli in the range 260.4 . . . 260.6 produce very good results with only slight

12
 1 260.4
E = 260.4 · (1 − ) = 260.4 − (40)
SG SG

compensates both approximations well, reproducing the ASBC tables with an

RMSE less than 0.0224°P , an MAE less than 0.0198°P and also has minimal error
for low SG and hence should even be prefered to equation 37 in the 1.000–1.011
SG range (see Table 1).

The best whole number modulus at 20 ◦ C is still 260, which reproduces the tables

20 7
with an RMSE of 0.030°P and an MAE of 0.023°P but a maximal error of 0.08°P
at the high end of the scale (see Table 1). The more frequently cited modulus 259

5, 1.
if used to estimate measurements at 20 ◦ C has an RMSE of 0.070°P , an MAE of

24
0.050°P and maximal error of 0.16°P .

Figure 1 shows an overview graph and Figure 2 a close-up of the region of interest
y v
of the rule of thumb linear equation 1, the best-fit second-order equation 2, and
the approximate formula 40.
ar int
Also, since no third-order best-fit function for the ASBC data giving E as a function
of SG could be found in the literature for comparison, a third-order polynomial re-
gression was performed on the ASBC table data for SG = 1.000, 1.001, 1.002 . . . 1.083
using Polysolve [19] which gave the following equation
nu pr

E = − 613.9427 + 1102.9079 · SG+

(41)
− 622.5576 · SG2 + 133.5892 · SG3 .
-
Ja re

and which has an RMSE less than 5.3 × 10−4 °P and a coefficient of determination
R2 > 0.999. A graph of this equation is also included in both figures.
p

2.4 The Partial Specific Volume in Aqueous Solutions of

Sucrose

In this section, the partial molar and partial specific volumes in aqueous sucrose
solutions are calculated from the ASBC’s version [2] of Plato’s data [10] and are
used to examine equation 28.
differences in their MAE, RMSE and maximal errors, with no single modulus minimizing all
errors simultaneously.

13
 Denote the mole fractions of A and B as XA and XB respectively. Dividing
equation 23 by the total number of moles ntot = nA + nB defines the total molar
volume to be

Vtot
Mtot = = X A · V A + XB · V B (42)
ntot

which can be shown to also be equal to

20 7
M WA M WB
Mtot = XA · + XB · (43)
ρsoln ρsoln

5, 1.
24
where MW denotes molecular weight and ρ the density of the solution as given in
equation 27. And since XA + XB = 1
y v M WA M WB
Mtot (XA ) = XA · + (1 − XA ) · (44)
ρsoln ρsoln
ar int
giving Mtot as a function of mole fraction XA of solute A.

Hence, in our case, Mtot as a function of mole fraction of sucrose can be computed
directly from equation 44 by converting SG to ρ with equation 27 and using the
nu pr

ASBC data.

Standard methods employing the Gibbs-Duhem equation show that

-
Ja re

dMtot
V A (XA ) = Mtot (XA ) + (1 − XA ) · . (45)
dXA

and since dMtot/dX = −dMtot/dXA

p

dMtot
V B (XA ) = Mtot (XA ) − XA · . (46)
dXA

dMtot
Performing a standard polynomial fit for Mtot (XA ) allows dXA
and hence V A (XA )
and V B (XA ) to be calculated.

As dataset, the ASBC conversion equation as given in the ASBC tables [2], which
◦C 20 ◦ C
converts SG204◦ Cvac to SG20 ◦ Cair , was applied directly to Plato et al.’s original

14
 data [10] instead of using the ASBC table values themselves. This was done
because the ASBC tables are sensibly tabulated in regular 0.00005 intervals of SG,
whereas Plato et al.’s original data is tabulated by mass fraction, as that is how it
was measured (i.e. Plato et al. measured the relative density of solutions prepared
to a given mass fraction.) Hence, the ASBC extract values are interpolated from
the original data, which introduces small rounding errors. Applying the same con-
version equation directly to the original data produces an dataset equivalent to
the ASBC tables, but eliminates these rounding errors.

Using this data to compute Mtot and performing a fourth-order polynomial regres-

20 7
sion again using Polysolve [19] to find Mtot (Xs ) gives

5, 1.
24
Mtot (Xs ) =18.04745 + 192.314 · Xs
y v + 169.813 · Xs2
(47)
− 8110.67 · Xs3
+ 235694.13 · Xs4
ar int
and has an RMSE less than 3.204 × 10−5 and a coefficient of determination R2 >
0.999.

The calculated data for V S (XS ), V W (XS ), v s , v w and E(v s , v w , SG) in sucrose
nu pr

solutions using the dataset described above and equation 28 are shown in Table 2.
The mean values of the partial specific volumes and their ratio are vs∅/vw∅ =
0.618 274 mL g−1/1.001 55 mL g−1 ≈ 0.61732 and the median values v s 1 /v 1 =
2 w2
-

0.618 454 mL g−1/1.0016 mL g−1 ≈ 0.61747.

Ja re

Using the calculated values for the partial specific volumes in equation 28, E can
be calculated for each SG in the dataset. As can be seen in the example data
in Table 2, the deviation from the known extract value is zero7 across the entire
p

range of the tables.

3 Discussion

Of course, using E to calculate Xs and v s (Xs (E)) to show that E(v s (Xs (E)), SG) =
E is somewhat circular8 , but it demonstrates that extract can be exactly calculated
7
The MAE of the calculated values is less than 6.8 × 10−13 °P .
8
Nevertheless, equation 28 is not tautological, i.e. it doesn’t simply reduce to E.

15
 from the relative density and the partial specific volumes of solute and solvent, if
these are known.

As far as actually calculating E from SG, equation 28 is mostly of theoretical

interest, as the exact partial specific volumes at the given relative density are
also needed in order to perform the computation. However, equation 28 does
provide the basis for understanding and interpreting previously known equations
(e.g. equations 13 14 16 and 19) as well as for deriving simple yet quite exact
approximate formulas whose parameters have physical meaning. It is a remarkably
fortunate circumstance that the partial specific volumes in dilute sucrose solutions

20 7
are astoundingly constant, varying across the entire useful range of concentration
by only about 1%, and hence can be well approximated in formulas by their average

5, 1.
24
values. It is even more fortunate that wort in general models the behaviour of
simple sucrose solutions so closely.

4
y v
Conclusion
ar int
After reviewing many of the approximate equations historically used to calculate
extract from SG and showing their interrelations, an exact equation showing how
E and SG are physically related was derived and it was shown how several of the
classical approximation equations are related to this physical equation. In this way,
nu pr

some of the traditionally used “constants” in these equations were interpreted and
their relationship to the partial specific volume of sucrose in aqueous solutions was
demonstrated. The partial specific volumes of both water and sucrose in solution
-

were calculated at varying concentrations and using these values it was demon-
strated that the derived equation indeed reproduces the ASBC tables exactly. A
Ja re

new third-order best-fit of the ASBC data giving E as a function of SG was pro-
vided, since this had been previously unavailable, as well as two more accurate,
yet simple approximate formulas than had previously been available, which con-
p

tain only physically meaningful parameters. It is hoped that these new equations
will help brewers who need to quickly and easily, yet accurately, compute percent
extract from simple hydrometer readings without expensive equipment.

5 Acknowledgements

The author is indebted to Prof. emeritus Alex Speers formerly of the Canadian
Institute of Fermentation Technology for helping to gain access to several key

16
 references cited in this article, which otherwise would have been difficult to ob-
tain, as well as to the reviewers whose suggestions greatly improved this article.
Additionally, many thanks are due to JASBC Editor-in-Chief Scott Britton and
his editorial staff for their continued patience during the final preparation of this
article.

6 Disclosure

20 7
The author reports there are no competing interests to declare.

5, 1.
24
References
y v
[1] Frederick J. Bates and H.W. Bearce. New Baumé Scale for Sugar Solutions.
Technologic Paper of the Bureau of Standards 115. Department of Com-
merce, Oct. 1918. url: http://dx.doi.org/10.6028/nbst.3715.
ar int
[2] American Society of Brewing Chemists. Tables Related to Determinations
on Wort, Beer, and Brewing Sugars and Syrups. St. Paul, MN: The Society,
2004.
[3] George K. Burgess, ed. Standard Density and Volumetric Tables. Sixth.
nu pr

Circular of the Bureau of Standards 19. Department of Commerce, Oct.

1924. url: https : / / www . govinfo . gov / app / details / GOVPUB - C13 -
d16de779f5660cced79946e8ac0d16a8.
-

[4] P. Chappuis. “Dilatation de l’eau”. In: Traveaux et mémoires du Bureau

Ja re

international des poids et mesures. Vol. XIII. Gauthier-Villars (Paris): Bu-

reau international des poids et mesures, 1907, pp. D.1–D.40. url: https:
//gallica.bnf.fr/ark:/12148/bpt6k6441711s.
p

[5] Jean de Clerck. A Textbook of Brewing [Cours de Brasserie]. Trans. by Kath-

leen Barton-Wright. 1st ed. Vol. II. London: Chapman & Hall, 1958.
[6] Jean de Clerck. Cours de Brasserie. 2nd ed. Vol. II. Heverlee: Univ. de Lou-
vain, 1963.
[7] Jean de Clerck. Lehrbuch der Brauerei [Cours de Brasserie]. Trans. by Paul
Kolbach. 2nd ed. Vol. II. Berlin: Versuchs- und Lehranstalt für Brauerei,
1965.

17
 [8] Anthony J. Cutaia, Anna-Jean Reid, and R. Alex Speers. “Examination
of the Relationships Between Original, Real and Apparent Extracts, and
Alcohol in Pilot Plant and Commercially Produced Beers”. In: Journal of
the Institute of Brewing 67.3 (2009), pp. 318–327.
[9] Howard DeVoe. Thermodynamics and Chemistry. LibreTexts. 2019. Chap. 9.2.
url: https : / / chem . libretexts . org / Bookshelves / Physical _ and _
Theoretical_Chemistry_Textbook_Maps/DeVoes_Thermodynamics_and_
Chemistry/09%3A_Mixtures/9.02%3A_Partial_Molar_Quantities.
[10] Johannes T. Domke, H. Harting, and Fritz Plato. Die Dichte, Ausdehnung

20 7
und Kapilarität von Lösungen Reinen Rohrzuckers in Wasser. Wissenschaftliche
Abhandlungen der Kaiserlichen Normal-Aichungs-Kommision. 1900. url:

5, 1.
24
https://daten.digitale-sammlungen.de/0006/bsb00067725/images/
index.html?id=00067725.
[11] Brewer’s Friend. Plato To SG Conversion Chart. url: https : / / web .
y v
archive.org/web/20220523163739/https://www.brewersfriend.com/
plato-to-sg-conversion-chart/. (accessed: 04.30.2023).
[12] F. Goldiner and Klemann H. Tafel zur Umrechnung der spezifischen Gewichte
ar int
20/4 von Zuckerlösungen auf Zuckerprozente und Tafel zur Umrechnung
der spezifischen Gewichte 20/20 von Akoholwassermischungen auf Geicht-
sprozente Alkohol. Berlin: Institut für Gärungsgewerbe, 1951.
[13] J.J. Hackbarth. “The Effect of Ethanol-Sucrose Interactions on Specific Grav-
nu pr

ity”. In: Journal of the American Society of Brewing Chemists 115 (2009),
pp. 146–151.
[14] Georgia Harris and Val Miller. “GLP 10: Good Laboratory Practice for
-

the Purity of Water”. In: NISTIR 7383, Selected Procedures for Volumetric
Calibrations (2019 Ed). Gaithersburg, MD: NIST Interagency/Internal Re-
Ja re

port (NISTIR), National Institute of Standards and Technology, 2019. doi:

https://doi.org/10.6028/NIST.IR.7383-2019. (accessed: 09.14.2023).
p

[15] AOAC International. “Table 942.33: Degrees Brix, specific gravity, and de-
grees Baumé of sugar solutions”. In: Official Methods of Analysis. Gaithers-
burg, MD: The Association, 2005.
[16] Justin. Plato to Specific Gravity Formulas. url: https://web.archive.
org/web/20230204205435/https://winning-homebrew.com/Plato-to-
specific-gravity.html. (accessed: 04.30.2023).
[17] George S. Kell. “Density, Thermal Expansivity, and Compressibility of Liq-
uid Water from 0° to 150°C: Correlations and Tables for Atmospheric Pres-
sure and Saturation Reviewed and Expressed on 1968 Temperature Scale”.
In: Journal of Chemical and Engineering Data 20.1 (1975), pp. 97–105.

18
 [18] Robert H. Lincoln. “Computer Compatible Parametric Equations for Ba-
sic Brewing Computations”. In: Technical Quarterly of the Master Brewers
Association of the Americas 24.4 (1987), pp. 129–132.
[19] Paul Lutus. Polynomial Regression Data Fit. url: https : / / arachnoid .
com/polysolve/. (accessed: 04.27.2023).
[20] Martin P. Manning. “Understanding Specific Gravity and Extract”. In: Brew-
ing Techniques 1.3 (1993), pp. 23–26.
[21] George S. Robbins. “Elimination of Tables for Malt and Beer Analysis Com-
putations”. In: The Brewers Digest 50.9 (Sept. 1975), pp. 66–67, 82–83.

20 7
[22] Karl Scheel. “Die Ausdehnung des Wassers mit der Temperatur”. In: Annalen

5, 1.
der Physik und Chemie 47.11 (1892), pp. 440–465. url: https://archive.

24
org/details/sim_annalen- der- physik_1892_47_11/page/440/mode/
2up.
y v
[23] K.J. Siebert. “Routine Use of a Programmable Calculator for Computing
Alcohol, Real Extract, Original Gravity, and Calories in Beer”. In: Journal
of the American Society of Brewing Chemists 38.1 (1980), pp. 27–33.
ar int
[24] S.W. Stratton, ed. United States Standard Baumé Hydrometer Scales. Circu-
lar of the Bureau of Standards 59. Department of Commerce, Apr. 1916. url:
https://www.govinfo.gov/app/details/GOVPUB-C13-ffb59b625df81205bb22839d4476d3d
[25] wikipedia.org. Brix. url: https : / / en . wikipedia . org / w / index . php ?
nu pr

title=Brix&oldid=1149018732. (accessed: 04.30.2023).

A Derivation of the Baumé Equation

-
Ja re

In this section equation 13 is derived without using the theory of partial molar
quantities and it’s relation to equation 26 is shown. In addition, the historical
p

Baumé moduli 140 and 145 are briefly examined.

The total mass of the solution composed of solute(s) and solvent(w) is

mt = ρ · vt = ms + ρw · vw (48)

hence,

ρ · vt − ms
vw = . (49)
ρw

19
 Assuming that the partial specific volume of the solvent is simply its reciprocal
density, i.e. that it is hardly affected by the addition of a small amount of solute,
and that of the solute is different from but proportional to the reciprocal density
of the solute9 gives

1 k
vt = · mw + · ms (50)
ρw ρs
| {z }
=vw

20 7
where k is the assumed proportionality constant.

5, 1.
Combining these equations and solving for ms gives

24
y v ρ − ρw
ms = vt · . (51)
1 − ρw · ρks

further rearranging and using mw = ρw · vw gives

ar int
k
ρw · (1 − ρ · ρs
)
mw = vt · k
. (52)
1 − ρw · ρs
nu pr

Using these results to calculate the mass fraction ws and simplifying gives
-

ms
ws =
Ja re

ms + mw
ρ − ρw
= k
ρ · (1 − ρw · ρs
)
(53)
p

1 1
ρw
− ρ
=
1 k
−
ρw ρs
|{z} |{z}
≈v w ≈v s

9
The partial specific volume of the solute has to be determined empirically, which, perhaps
interestingly, was the original starting point of this entire investigation. Dissolving 50g of sucrose
in 100 ml of water at 20 ◦ C resulted in 132 ml of solution, giving an apparent partial specific
volume of approximately 32/50 = 0.64 mL g−1 . Applying the calculations of section 2.4 to the
Plato Table value for E = 33.4°P gives 0.6394 mL g−1 .

20
 which is essentially equation 26. Expanding the fraction with ρw and multiplying
by 100 to get E gives
ρw
1− ρ
E = 100 ·
1 − ρw · ρks
100 1 (54)
= ρw ·(1 − )
1 − k · ρs SG
| {z }
≡m

20 7
and hence
1
d = m · (1 − ) (55)
s

5, 1.
24
where m is the Baumé modulus, d are the degrees Baumé and s the relative
density (using the classic variables). This is the inverse of equation 13 and can be
y v
compared to equation 31.

If the Baumé modulus for dilute aqueous solutions of sodium chloride at 12.5◦ C is
ar int
140 [1], then it’s partial specific volume at this temperature for low concentrations
should be approximately v N aCl ≈ 1 − 100/140 ≈ 0.286 mL g−1 which is in exact
agreement with an example calculation using partial molar quantity methods in
the textbook by DeVoe [9] who arrives at V N aCl ≈ 16.71 mL mol−1 ⇒ v N aCl ≈
0.286 mL g−1 . This result can be compared to it’s reciprocal density 1/ρs = 1/2.16 ≈
nu pr

0.463 mL g−1 . The fact that the partial specific volume of sodium chloride in water
is considerably lower than it’s reciprocal density is not surprising since it’s an
ionic compound and water is a polar molecule. Had the manufacturers of Baumé
hydrometers in the 19th century used the original modulus of 140 [24] instead of
-

145 (which the Bureau of Standards later fixed as calibration value because that’s
Ja re

what the manufactureres of the time were using. [1][24]), then °Bé would have been
equivalent to mass fraction of aqueous saline solutions analogous to °Bx and °P
which correspond to mass fraction of aqueous sucrose solutions.
p

21
 B Water Density

The calculations in this paper are dependent upon a precise and accurate value
for the density of water at 20 ◦ C in order to calculate the solution density ρsoln for
use in equation 44 from the solution’s relative density in air using equation 27.
◦ −1
The density for water at 20 ◦ C used here, ρ20 C
H2 O = 0.998 207 35 g mL , was calcu-
◦
lated from the value given in the ASBC Tables [2] for the (relative) density D420◦ CC
◦
of water used in the ASBC calculations, assuming ρ4H2CO = 0.999 973 g mL−1 , which

20 7
was a commonly accepted value before 1950 (see Kell [17]) when the tables were
created:

5, 1.
24
◦ ◦ ◦
ρ20 C 20 C 4 C
H 2 O = D4 ◦ C · ρ H 2 O
y v = 0.9982343 · 0.999 973 g mL−1
= 0.998 207 35 g mL −1
(56)
ar int
This value also agrees with modern values (to six figures) if corrections are made
for the various temperature scale definitions assuming 20 ◦ CtEN = 20 ◦ Ct27 =
20 ◦ Ct48 = 19.987 773 3 ◦ Ct90 , and it is assumed that the ASBC tables are in accord
with the IPTS-48 temperature standard. see Kell [17] and Harris et al. [14].
nu pr

◦ −1
However, a case could also be made for using ρ20 C
H2 O = 0.998 234 3 g mL . Although
no source is cited in the ASBC tables, the numerical value 0.9982343 is identical to
that given in the Bureau of Standards’ Standard Density and Volumetric Tables [3]
(which themselves reference Chappuis [4]), which would have been a valid standard
-

reference at the time, and which in the introduction states that the density values
Ja re

◦C 4 ◦C
in g mL−1 are numerically identical to SG20 4 ◦ C values, i.e. they assume ρH2 O is
identically one, which would also explain why the ASBC Tables refer to this relative
density value as just density. Using this value results in slightly different values
p

for v s and v w than reported in section 2.4, but interestingly not in their ratio to
the precision reported in this paper.

P. Chappuis [4] indeed reports a Densité de l’eau of 0.9982343 at 20 ◦ CtEN and

1.000 000 at 4 ◦ CtEN , but no units are given in the table and this author’s command
of the French language was insufficient to deduce whether or not these are actually
◦
relative densities or whether perhaps ρ4H2CO was possibly defined at the time by the
échelle normale to be exactly 1 g mL−1 as implied by the Bureau of Standards
Density Tables [3].
◦
F. Plato et al. also report SG20 C
4 ◦ C = 0.998234 for water (the value in the table

22
 ◦
for 0°P ) but state clearly that this value is relative to ρ4H2CO . Interestingly, F.
Plato et al. cite (see [10] p. 36) the averaged values given by Scheel [22] for their
◦C
calculations, which however was not 0.9982343 but rather SG20 4 ◦ C = 0.9982353.

20 7
5, 1.
24
y v
ar int
nu pr
-
Ja re
p

23
 C Calculation with Approximation Formulas

This table shows calculations using the approximate formulas (36), (37) and (40) for each SG equal to 1.000, 1.001, 1.002, etc. up to 1.083 in the ASBC
dataset [2].

pr ua
Table 1: Data Calculation Using Approximation Equations 36, 37 and 40

ASBC Table values[2] E = 261.42 − 261.47/SG |∆| E = 261 · (1 − 1/SG) |∆| E = 260 · (1 − 1/SG) |∆| E = 260.4 · (1 − 1/SG) |∆|
SG2020 E
Ja
e- ry
1 0 -0.050 0.0500 0.000 0.0000 0.000 0.0000 0.000 0.0000
1.001 0.257 0.211 0.0458 0.261 0.0037 0.260 0.0027 0.260 0.0031
1.002 0.514 0.472 0.0421 0.521 0.0070 0.519 0.0050 0.520 0.0058

n
pr 5
1.003 0.770 0.732 0.0379 0.781 0.0107 0.778 0.0077 0.779 0.0089
1.004 1.026 0.992 0.0343 1.040 0.0138 1.036 0.0099 1.037 0.0115
1.005 1.283 1.251 0.0322 1.299 0.0155 1.294 0.0105 1.296 0.0125
1.006 1.539 1.509 0.0295 1.557 0.0177 1.551 0.0117 1.553 0.0141

in , 2
1.007 1.795 1.768 0.0274 1.814 0.0193 1.807 0.0123 1.810 0.0151
1.008 2.051 2.025 0.0258 2.071 0.0204 2.063 0.0125 2.067 0.0157
24

1.009 2.305 2.282 0.0228 2.328 0.0230 2.319 0.0141 2.323 0.0177

tv 0
1.010 2.560 2.539 0.0212 2.584 0.0242 2.574 0.0143 2.578 0.0182
1.011 2.814 2.795 0.0191 2.840 0.0258 2.829 0.0149 2.833 0.0192
1.012 3.067 3.050 0.0166 3.095 0.0279 3.083 0.0160 3.088 0.0207
1.013 3.321 3.305 0.0155 3.349 0.0285 3.337 0.0156 3.342 0.0208
1.014 3.573 3.560 0.0130 3.604 0.0306 3.590 0.0167 3.595 0.0223

1. 4
1.015 3.826 3.814 0.0119 3.857 0.0311 3.842 0.0164 3.848 0.0223
1.016 4.077 4.068 0.0094 4.110 0.0332 4.094 0.0175 4.101 0.0238
1.017 4.329 4.321 0.0083 4.363 0.0338 4.346 0.0171 4.353 0.0238

7
1.018 4.580 4.573 0.0068 4.615 0.0349 4.597 0.0172 4.604 0.0243
1.019 4.830 4.825 0.0047 4.867 0.0365 4.848 0.0179 4.855 0.0253
1.020 5.080 5.077 0.0031 5.118 0.0376 5.098 0.0180 5.106 0.0259
1.021 5.330 5.328 0.0021 5.368 0.0383 5.348 0.0177 5.356 0.0259
1.022
1.023
1.024
1.025
1.026
5.580
5.828
6.077
6.325
6.572
2 5.579
5.829
6.078
6.327
6.576
0.0015
0.0006
0.0012
0.0023
0.0039
5.618
5.868
6.117
6.366
6.614
0.0384
0.0400
0.0402
0.0409
0.0420
5.597
5.846
6.094
6.341
6.589
0.0169
0.0176
0.0167
0.0165
0.0167
5.605
5.855
6.103
6.351
6.599
0.0255
0.0265
0.0261
0.0262
0.0268
 Table 1: Data Calculation Using Approximation Equations 36, 37 and 40

ASBC Table values[2] E = 261.42 − 261.47/SG |∆| E = 261 · (1 − 1/SG) |∆| E = 260 · (1 − 1/SG) |∆| E = 260.4 · (1 − 1/SG) |∆|
SG20
20 E
1.027 6.819 6.824 0.0051 6.862 0.0427 6.835 0.0164 6.846 0.0270
1.028 7.066 7.072 0.0058 7.109 0.0429 7.082 0.0157 7.093 0.0266

pr ua
1.029 7.312 7.319 0.0069 7.356 0.0437 7.328 0.0155 7.339 0.0268
1.030 7.558 7.566 0.0076 7.602 0.0439 7.573 0.0148 7.584 0.0265
1.031
1.032
Ja
7.803
8.048
7.812
8.058
0.0089
0.0096
7.848
8.093
0.0447
0.0450
7.818
8.062
0.0147
0.0140
7.830
8.074
0.0267
0.0264

e- ry
1.033 8.293 8.303 0.0099 8.338 0.0449 8.306 0.0129 8.319 0.0257
1.034 8.537 8.548 0.0107 8.582 0.0452 8.549 0.0123 8.562 0.0255
1.035 8.781
n 8.792 0.0110 8.826 0.0451 8.792 0.0113 8.806 0.0248

pr 5
1.036 9.024 9.036 0.0118 9.069 0.0455 9.035 0.0107 9.049 0.0246
1.037 9.267 9.279 0.0122 9.312 0.0454 9.277 0.0098 9.291 0.0240
1.038 9.509 9.522 0.0131 9.555 0.0459 9.518 0.0093 9.533 0.0239

in , 2
1.039 9.751 9.765 0.0136 9.797 0.0459 9.759 0.0084 9.774 0.0234
1.040 9.993 10.007 0.0135 10.038 0.0455 10.000 0.0070 10.015 0.0224
25

1.041 10.234 10.248 0.0140 10.280 0.0455 10.240 0.0062 10.256 0.0219

tv 0
1.042 10.475 10.489 0.0141 10.520 0.0452 10.480 0.0048 10.496 0.0210
1.043 10.716 10.730 0.0137 10.760 0.0443 10.719 0.0031 10.736 0.0196
1.044 10.956 10.970 0.0138 11.000 0.0440 10.958 0.0019 10.975 0.0187
1.045 11.195 11.209 0.0145 11.239 0.0442 11.196 0.0012 11.213 0.0184
1.046 11.435 11.449 0.0137 11.478 0.0430 11.434 0.0010 11.452 0.0166

1. 4
1.047 11.673 11.687 0.0144 11.716 0.0433 11.671 0.0016 11.689 0.0164
1.048 11.912 11.926 0.0137 11.954 0.0422 11.908 0.0036 11.927 0.0147
1.049 12.150 12.164 0.0136 12.192 0.0416 12.145 0.0051 12.164 0.0136

7
1.050 12.387 12.401 0.0140 12.429 0.0416 12.381 0.0060 12.400 0.0130
1.051 12.624 12.638 0.0139 12.665 0.0411 12.617 0.0074 12.636 0.0120
1.052 12.861 12.874 0.0134 12.901 0.0401 12.852 0.0093 12.871 0.0105
1.053 13.098 13.110 0.0124 13.137 0.0388 13.086 0.0116 13.107 0.0086
1.054
1.055
1.056
1.057
1.058
13.333
13.569
13.804
14.039
14.273
2 13.346
13.581
13.816
14.050
14.284
0.0130
0.0121
0.0118
0.0111
0.0109
13.372
13.607
13.841
14.075
14.308
0.0389
0.0376
0.0369
0.0357
0.0351
13.321
13.555
13.788
14.021
14.253
0.0123
0.0145
0.0161
0.0182
0.0197
13.341
13.575
13.809
14.042
14.275
0.0082
0.0064
0.0051
0.0034
0.0022
1.059 14.507 14.517 0.0103 14.541 0.0341 14.485 0.0216 14.508 0.0006
 Table 1: Data Calculation Using Approximation Equations 36, 37 and 40

ASBC Table values[2] E = 261.42 − 261.47/SG |∆| E = 261 · (1 − 1/SG) |∆| E = 260 · (1 − 1/SG) |∆| E = 260.4 · (1 − 1/SG) |∆|
SG20
20 E
1.060 14.741 14.750 0.0092 14.774 0.0326 14.717 0.0240 14.740 0.0014
1.061 14.974 14.983 0.0087 15.006 0.0317 14.948 0.0258 14.971 0.0028

pr ua
1.062 15.207 15.215 0.0077 15.237 0.0303 15.179 0.0281 15.202 0.0047
1.063 15.439 15.446 0.0073 15.468 0.0295 15.409 0.0298 15.433 0.0061
1.064
1.065
Ja
15.671
15.903
15.678
15.908
0.0065
0.0053
15.699
15.930
0.0282
0.0266
15.639
15.869
0.0319
0.0345
15.663
15.893
0.0078
0.0100

e- ry
1.066 16.134 16.139 0.0046 16.159 0.0255 16.098 0.0364 16.122 0.0117
1.067 16.365 16.368 0.0035 16.389 0.0239 16.326 0.0389 16.351 0.0137
1.068 16.595
n 16.598 0.0029 16.618 0.0230 16.554 0.0407 16.580 0.0152

pr 5
1.069 16.825 16.827 0.0019 16.847 0.0216 16.782 0.0430 16.808 0.0171
1.070 17.055 17.056 0.0005 17.075 0.0198 17.009 0.0457 17.036 0.0195
1.071 17.284 17.284 0.0003 17.303 0.0185 17.236 0.0478 17.263 0.0213

in , 2
1.072 17.513 17.511 0.0016 17.530 0.0169 17.463 0.0503 17.490 0.0234
1.073 17.741 17.739 0.0023 17.757 0.0158 17.689 0.0523 17.716 0.0251
26

1.074 17.970 17.966 0.0044 17.983 0.0132 17.914 0.0557 17.942 0.0281

tv 0
1.075 18.197 18.192 0.0049 18.209 0.0123 18.140 0.0575 18.167 0.0296
1.076 18.425 18.418 0.0069 18.435 0.0099 18.364 0.0607 18.393 0.0324
1.077 18.652 18.644 0.0082 18.660 0.0082 18.589 0.0633 18.617 0.0347
1.078 18.878 18.869 0.0090 18.885 0.0070 18.813 0.0654 18.842 0.0364
1.079 19.105 19.094 0.0112 19.109 0.0044 19.036 0.0689 19.065 0.0396

1. 4
1.080 19.331 19.318 0.0129 19.333 0.0023 19.259 0.0717 19.289 0.0421
1.081 19.556 19.542 0.0139 19.557 0.0009 19.482 0.0740 19.512 0.0441
1.082 19.782 19.766 0.0163 19.780 0.0020 19.704 0.0777 19.735 0.0474

7
1.083 20.007 19.989 0.0182 20.003 0.0042 19.926 0.0809 19.957 0.0502

MAE 0.0124 0.0300 0.0228 0.0197

MAX 0.0500 0.0459 0.0809 0.0502

2 RMSE 0.0159 0.0331 0.0304 0.0224

 D Data Calculation

This table summarizes the data calculations for the full dataset using Equations 28, 46 and 47 as described in section 2.4.

Table 2: Data Calculation Using Equations 28, 46 and 47

pr ua
′
Plato’s Table[10] SG20
20 ρ Xs Mtot Mtot (Xs ) Mtot Vs Vw vs vw E(v s , v w , SG20
20 ) ∆
E SG20
4
0 0.9982
Ja
1.0000 0.99821 0.00000 18.0476 18.0475 192.314 210.362 18.0476 0.61456 1.00180 0.000 0.000

e- ry
0.1 0.9986 1.0004 0.99860 0.00005 18.0577 18.0576 192.332 210.379 18.0476 0.61461 1.00179 0.100 0.000
0.2 0.9990 1.0008 0.99898 0.00011 18.0679 18.0677 192.350 210.397 18.0476 0.61466 1.00179 0.200 0.000
0.3 0.9994 1.0012 0.99937 0.00016 18.0780 18.0779 192.367 210.415 18.0475 0.61471 1.00179 0.300 0.000

n
pr 5
0.4 0.9998 1.0016 0.99976 0.00021 18.0882 18.0881 192.385 210.432 18.0475 0.61476 1.00179 0.400 0.000
0.5 1.0002 1.0019 1.00015 0.00026 18.0984 18.0983 192.402 210.450 18.0475 0.61481 1.00179 0.500 0.000
0.6 1.0006 1.0023 1.00054 0.00032 18.1086 18.1085 192.419 210.467 18.0475 0.61486 1.00179 0.600 0.000
0.7 1.0010 1.0027 1.00092 0.00037 18.1188 18.1188 192.437 210.484 18.0474 0.61492 1.00178 0.700 0.000

in , 2
0.8 1.0013 1.0031 1.00131 0.00042 18.1290 18.1291 192.454 210.501 18.0474 0.61496 1.00178 0.800 0.000
0.9 1.0017 1.0035 1.00170 0.00048 18.1393 18.1394 192.471 210.518 18.0473 0.61501 1.00178 0.900 0.000
27

1 1.0021 1.0039 1.00209 0.00053 18.1496 18.1497 192.488 210.535 18.0473 0.61506 1.00178 1.000 0.000

tv 0
1.1 1.0025 1.0043 1.00248 0.00059 18.1599 18.1600 192.505 210.552 18.0473 0.61511 1.00178 1.100 0.000
1.2 1.0029 1.0047 1.00287 0.00064 18.1703 18.1704 192.521 210.569 18.0473 0.61516 1.00178 1.200 0.000
1.3 1.0033 1.0051 1.00326 0.00069 18.1807 18.1807 192.538 210.585 18.0473 0.61521 1.00178 1.300 0.000
1.4 1.0037 1.0055 1.00365 0.00075 18.1911 18.1911 192.554 210.602 18.0473 0.61526 1.00178 1.400 0.000
1.5 1.0041 1.0058 1.00404 0.00080 18.2015 18.2016 192.571 210.618 18.0473 0.61531 1.00178 1.500 0.000

1. 4
1.6 1.0045 1.0062 1.00443 0.00086 18.2120 18.2120 192.587 210.634 18.0473 0.61535 1.00178 1.600 0.000
1.7 1.0048 1.0066 1.00482 0.00091 18.2224 18.2225 192.603 210.651 18.0472 0.61540 1.00177 1.700 0.000
1.8 1.0052 1.0070 1.00521 0.00096 18.2329 18.2329 192.620 210.667 18.0472 0.61545 1.00177 1.800 0.000

7
1.9 1.0056 1.0074 1.00560 0.00102 18.2434 18.2435 192.636 210.683 18.0472 0.61550 1.00177 1.900 0.000
2 1.0060 1.0078 1.00599 0.00107 18.2539 18.2540 192.652 210.699 18.0472 0.61554 1.00177 2.000 0.000
2.1 1.0064 1.0082 1.00638 0.00113 18.2645 18.2645 192.667 210.715 18.0472 0.61559 1.00177 2.100 0.000
2.2 1.0068 1.0086 1.00677 0.00118 18.2750 18.2751 192.683 210.730 18.0472 0.61563 1.00177 2.200 0.000
2.3
2.4
2.5
2.6
2.7
1.0072
1.0076
1.0080
1.0084
1.0088
1.0090
1.0094
1.0098
1.0102
1.0106
2
1.00716
1.00755
1.00794
1.00834
1.00873
0.00124
0.00129
0.00135
0.00140
0.00146
18.2856
18.2962
18.3069
18.3175
18.3282
18.2857
18.2963
18.3069
18.3176
18.3282
192.699
192.714
192.730
192.745
192.760
210.746
210.762
210.777
210.792
210.808
18.0472
18.0471
18.0471
18.0471
18.0471
0.61568
0.61573
0.61577
0.61582
0.61586
1.00177
1.00177
1.00177
1.00177
1.00177
2.300
2.400
2.500
2.600
2.700
0.000
0.000
0.000
0.000
0.000
 Table 2: Data Calculation Using Equations 28, 46 and 47
′
Plato’s Table[10] SG20
20 ρ Xs Mtot Mtot (Xs ) Mtot Vs Vw vs vw E(v s , v w , SG20
20 ) ∆
E SG20
4
2.8 1.0091 1.0109 1.00912 0.00151 18.3389 18.3389 192.776 210.823 18.0471 0.61590 1.00177 2.800 0.000
2.9 1.0095 1.0113 1.00951 0.00157 18.3496 18.3497 192.791 210.838 18.0471 0.61595 1.00176 2.900 0.000

pr ua
3 1.0099 1.0117 1.00991 0.00163 18.3604 18.3604 192.806 210.853 18.0471 0.61599 1.00176 3.000 0.000
3.1 1.0103 1.0121 1.01030 0.00168 18.3712 18.3712 192.821 210.868 18.0470 0.61604 1.00176 3.100 0.000
3.2
3.3
1.0107
1.0111
Ja
1.0125
1.0129
1.01069
1.01109
0.00174
0.00179
18.3819
18.3927
18.3819
18.3927
192.835
192.850
210.882
210.897
18.0470
18.0470
0.61608
0.61612
1.00176
1.00176
3.200
3.300
0.000
0.000

e- ry
3.4 1.0115 1.0133 1.01148 0.00185 18.4036 18.4036 192.865 210.912 18.0469 0.61616 1.00176 3.400 0.000
3.5 1.0119 1.0137 1.01188 0.00191 18.4144 18.4144 192.879 210.926 18.0469 0.61621 1.00176 3.500 0.000
3.6 1.0123
n
1.0141 1.01227 0.00196 18.4253 18.4253 192.894 210.941 18.0469 0.61625 1.00176 3.600 0.000

pr 5
3.7 1.0127 1.0145 1.01267 0.00202 18.4362 18.4362 192.908 210.955 18.0469 0.61629 1.00175 3.700 0.000
3.8 1.0131 1.0149 1.01306 0.00207 18.4471 18.4471 192.922 210.969 18.0469 0.61633 1.00175 3.800 0.000
3.9 1.0135 1.0153 1.01346 0.00213 18.4580 18.4580 192.936 210.983 18.0468 0.61637 1.00175 3.900 0.000

in , 2
4 1.0139 1.0157 1.01385 0.00219 18.4690 18.4690 192.951 210.997 18.0468 0.61641 1.00175 4.000 0.000
4.1 1.0143 1.0161 1.01425 0.00225 18.4800 18.4800 192.965 211.011 18.0468 0.61646 1.00175 4.100 0.000
28

4.2 1.0147 1.0165 1.01465 0.00230 18.4910 18.4910 192.978 211.025 18.0467 0.61650 1.00175 4.200 0.000

tv 0
4.3 1.0151 1.0169 1.01504 0.00236 18.5020 18.5020 192.992 211.039 18.0467 0.61654 1.00174 4.300 0.000
4.4 1.0155 1.0173 1.01544 0.00242 18.5131 18.5131 193.006 211.053 18.0467 0.61658 1.00174 4.400 0.000
4.5 1.0159 1.0177 1.01584 0.00247 18.5241 18.5241 193.020 211.066 18.0466 0.61662 1.00174 4.500 0.000
4.6 1.0163 1.0181 1.01623 0.00253 18.5352 18.5352 193.033 211.080 18.0466 0.61666 1.00174 4.600 0.000
4.7 1.0167 1.0185 1.01663 0.00259 18.5464 18.5463 193.047 211.093 18.0466 0.61669 1.00174 4.700 0.000

1. 4
4.8 1.0171 1.0189 1.01703 0.00265 18.5575 18.5575 193.060 211.106 18.0465 0.61673 1.00173 4.800 0.000
4.9 1.0175 1.0193 1.01743 0.00270 18.5687 18.5686 193.073 211.120 18.0465 0.61677 1.00173 4.900 0.000
5 1.0179 1.0197 1.01783 0.00276 18.5799 18.5798 193.086 211.133 18.0465 0.61681 1.00173 5.000 0.000

7
5.1 1.0183 1.0201 1.01823 0.00282 18.5911 18.5910 193.099 211.146 18.0464 0.61685 1.00173 5.100 0.000
5.2 1.0187 1.0205 1.01862 0.00288 18.6023 18.6023 193.113 211.159 18.0464 0.61689 1.00173 5.200 0.000
5.3 1.0191 1.0209 1.01902 0.00294 18.6135 18.6135 193.125 211.172 18.0464 0.61692 1.00173 5.300 0.000
5.4 1.0195 1.0213 1.01942 0.00300 18.6248 18.6248 193.138 211.185 18.0463 0.61696 1.00172 5.400 0.000
5.5
5.6
5.7
5.8
5.9
1.0199
1.0203
1.0207
1.0211
1.0215
1.0217
1.0221
1.0225
1.0229
1.0233
2
1.01982
1.02022
1.02062
1.02103
1.02143
0.00305
0.00311
0.00317
0.00323
0.00329
18.6361
18.6475
18.6588
18.6702
18.6816
18.6361
18.6474
18.6588
18.6702
18.6816
193.151
193.164
193.176
193.189
193.201
211.197
211.210
211.223
211.235
211.248
18.0463
18.0463
18.0462
18.0462
18.0461
0.61700
0.61704
0.61707
0.61711
0.61715
1.00172
1.00172
1.00172
1.00171
1.00171
5.500
5.600
5.700
5.800
5.900
0.000
0.000
0.000
0.000
0.000
6 1.0219 1.0237 1.02183 0.00335 18.6930 18.6930 193.214 211.260 18.0461 0.61718 1.00171 6.000 0.000
 Table 2: Data Calculation Using Equations 28, 46 and 47
′
Plato’s Table[10] SG20
20 ρ Xs Mtot Mtot (Xs ) Mtot Vs Vw vs vw E(v s , v w , SG20
20 ) ∆
E SG20
4
6.1 1.0223 1.0241 1.02223 0.00341 18.7044 18.7044 193.226 211.272 18.0461 0.61722 1.00171 6.100 0.000
6.2 1.0227 1.0245 1.02263 0.00347 18.7159 18.7159 193.238 211.284 18.0460 0.61725 1.00171 6.200 0.000

pr ua
6.3 1.0231 1.0249 1.02303 0.00353 18.7274 18.7274 193.250 211.296 18.0460 0.61729 1.00170 6.300 0.000
6.4 1.0235 1.0253 1.02344 0.00359 18.7389 18.7389 193.262 211.308 18.0459 0.61732 1.00170 6.400 0.000
6.5
6.6
1.0239
1.0243
Ja
1.0257
1.0261
1.02384
1.02424
0.00365
0.00371
18.7504
18.7620
18.7504
18.7620
193.274
193.286
211.320
211.332
18.0459
18.0458
0.61736
0.61739
1.00170
1.00170
6.500
6.600
0.000
0.000

e- ry
6.7 1.0247 1.0265 1.02465 0.00377 18.7736 18.7736 193.298 211.344 18.0458 0.61743 1.00169 6.700 0.000
6.8 1.0251 1.0269 1.02505 0.00383 18.7852 18.7852 193.310 211.356 18.0457 0.61746 1.00169 6.800 0.000
6.9 1.0255
n
1.0273 1.02545 0.00389 18.7968 18.7968 193.322 211.367 18.0457 0.61750 1.00169 6.900 0.000

pr 5
7 1.0259 1.0277 1.02586 0.00395 18.8085 18.8085 193.333 211.379 18.0456 0.61753 1.00169 7.000 0.000
7.1 1.0263 1.0281 1.02626 0.00401 18.8202 18.8202 193.345 211.390 18.0456 0.61756 1.00168 7.100 0.000
7.2 1.0267 1.0285 1.02667 0.00407 18.8319 18.8319 193.356 211.402 18.0456 0.61760 1.00168 7.200 0.000

in , 2
7.3 1.0271 1.0290 1.02707 0.00413 18.8436 18.8436 193.368 211.413 18.0455 0.61763 1.00168 7.300 0.000
7.4 1.0275 1.0294 1.02748 0.00419 18.8554 18.8554 193.379 211.424 18.0455 0.61766 1.00168 7.400 0.000
29

7.5 1.0279 1.0298 1.02788 0.00425 18.8672 18.8672 193.390 211.436 18.0454 0.61769 1.00167 7.500 0.000

tv 0
7.6 1.0283 1.0302 1.02829 0.00431 18.8790 18.8790 193.401 211.447 18.0454 0.61773 1.00167 7.600 0.000
7.7 1.0287 1.0306 1.02869 0.00437 18.8908 18.8908 193.412 211.458 18.0453 0.61776 1.00167 7.700 0.000
7.8 1.0291 1.0310 1.02910 0.00443 18.9027 18.9026 193.423 211.469 18.0453 0.61779 1.00166 7.800 0.000
7.9 1.0295 1.0314 1.02951 0.00449 18.9145 18.9145 193.434 211.480 18.0452 0.61782 1.00166 7.900 0.000
8 1.0299 1.0318 1.02991 0.00456 18.9264 18.9264 193.445 211.491 18.0452 0.61786 1.00166 8.000 0.000

1. 4
8.1 1.0303 1.0322 1.03032 0.00462 18.9384 18.9384 193.456 211.501 18.0451 0.61789 1.00166 8.100 0.000
8.2 1.0308 1.0326 1.03073 0.00468 18.9503 18.9503 193.467 211.512 18.0451 0.61792 1.00165 8.200 0.000
8.3 1.0312 1.0330 1.03114 0.00474 18.9623 18.9623 193.478 211.523 18.0450 0.61795 1.00165 8.300 0.000

7
8.4 1.0316 1.0334 1.03155 0.00480 18.9743 18.9743 193.488 211.533 18.0450 0.61798 1.00165 8.400 0.000
8.5 1.0320 1.0338 1.03195 0.00487 18.9863 18.9863 193.499 211.544 18.0449 0.61801 1.00164 8.500 0.000
8.6 1.0324 1.0343 1.03236 0.00493 18.9984 18.9984 193.510 211.554 18.0448 0.61804 1.00164 8.600 0.000
8.7 1.0328 1.0347 1.03277 0.00499 19.0105 19.0105 193.520 211.565 18.0448 0.61807 1.00164 8.700 0.000
8.8
8.9
9
9.1
9.2
1.0332
1.0336
1.0340
1.0344
1.0349
1.0351
1.0355
1.0359
1.0363
1.0367
2
1.03318
1.03359
1.03400
1.03441
1.03482
0.00505
0.00512
0.00518
0.00524
0.00530
19.0226
19.0347
19.0469
19.0591
19.0713
19.0226
19.0347
19.0469
19.0591
19.0713
193.530
193.541
193.551
193.561
193.572
211.575
211.586
211.596
211.606
211.616
18.0447
18.0447
18.0446
18.0446
18.0445
0.61810
0.61813
0.61816
0.61819
0.61822
1.00164
1.00163
1.00163
1.00163
1.00162
8.800
8.900
9.000
9.100
9.200
0.000
0.000
0.000
0.000
0.000
9.3 1.0353 1.0371 1.03523 0.00537 19.0835 19.0835 193.582 211.626 18.0445 0.61825 1.00162 9.300 0.000
 Table 2: Data Calculation Using Equations 28, 46 and 47
′
Plato’s Table[10] SG20
20 ρ Xs Mtot Mtot (Xs ) Mtot Vs Vw vs vw E(v s , v w , SG20
20 ) ∆
E SG20
4
9.4 1.0357 1.0375 1.03564 0.00543 19.0958 19.0958 193.592 211.636 18.0444 0.61828 1.00162 9.400 0.000
9.5 1.0361 1.0380 1.03605 0.00549 19.1081 19.1081 193.602 211.646 18.0444 0.61831 1.00161 9.500 0.000

pr ua
9.6 1.0365 1.0384 1.03647 0.00556 19.1204 19.1204 193.612 211.656 18.0443 0.61834 1.00161 9.600 0.000
9.7 1.0369 1.0388 1.03688 0.00562 19.1327 19.1328 193.622 211.666 18.0442 0.61837 1.00161 9.700 0.000
9.8
9.9
1.0373
1.0377
Ja
1.0392
1.0396
1.03729
1.03770
0.00569
0.00575
19.1451
19.1575
19.1451
19.1575
193.632
193.642
211.676
211.686
18.0442
18.0441
0.61840
0.61843
1.00161
1.00160
9.800
9.900
0.000
0.000

e- ry
10 1.0381 1.0400 1.03811 0.00581 19.1699 19.1699 193.651 211.695 18.0441 0.61845 1.00160 10.000 0.000
10.1 1.0386 1.0404 1.03853 0.00588 19.1824 19.1824 193.661 211.705 18.0440 0.61848 1.00160 10.100 0.000
10.2 1.0390
n
1.0409 1.03894 0.00594 19.1949 19.1949 193.671 211.715 18.0440 0.61851 1.00159 10.200 0.000

pr 5
10.3 1.0394 1.0413 1.03935 0.00601 19.2074 19.2074 193.681 211.724 18.0439 0.61854 1.00159 10.300 0.000
10.4 1.0398 1.0417 1.03977 0.00607 19.2199 19.2199 193.690 211.734 18.0439 0.61857 1.00159 10.400 0.000
10.5 1.0402 1.0421 1.04018 0.00614 19.2324 19.2325 193.700 211.744 18.0438 0.61859 1.00158 10.500 0.000

in , 2
10.6 1.0406 1.0425 1.04060 0.00620 19.2450 19.2450 193.709 211.753 18.0437 0.61862 1.00158 10.600 0.000
10.7 1.0410 1.0429 1.04101 0.00627 19.2577 19.2577 193.719 211.762 18.0437 0.61865 1.00158 10.700 0.000
30

10.8 1.0415 1.0434 1.04143 0.00633 19.2703 19.2703 193.728 211.772 18.0436 0.61868 1.00157 10.800 0.000

tv 0
10.9 1.0419 1.0438 1.04184 0.00640 19.2830 19.2830 193.738 211.781 18.0435 0.61870 1.00157 10.900 0.000
11 1.0423 1.0442 1.04226 0.00646 19.2956 19.2957 193.747 211.791 18.0435 0.61873 1.00157 11.000 0.000
11.1 1.0427 1.0446 1.04268 0.00653 19.3084 19.3084 193.757 211.800 18.0434 0.61876 1.00156 11.100 0.000
11.2 1.0431 1.0450 1.04309 0.00659 19.3211 19.3211 193.766 211.809 18.0433 0.61879 1.00156 11.200 0.000
11.3 1.0435 1.0454 1.04351 0.00666 19.3339 19.3339 193.775 211.818 18.0433 0.61881 1.00156 11.300 0.000

1. 4
11.4 1.0440 1.0459 1.04393 0.00673 19.3467 19.3467 193.784 211.828 18.0432 0.61884 1.00155 11.400 0.000
11.5 1.0444 1.0463 1.04434 0.00679 19.3595 19.3595 193.794 211.837 18.0432 0.61887 1.00155 11.500 0.000
11.6 1.0448 1.0467 1.04476 0.00686 19.3724 19.3724 193.803 211.846 18.0431 0.61889 1.00155 11.600 0.000

7
11.7 1.0452 1.0471 1.04518 0.00693 19.3853 19.3853 193.812 211.855 18.0431 0.61892 1.00154 11.700 0.000
11.8 1.0456 1.0475 1.04560 0.00699 19.3982 19.3982 193.821 211.864 18.0430 0.61895 1.00154 11.800 0.000
11.9 1.0460 1.0480 1.04601 0.00706 19.4111 19.4112 193.831 211.873 18.0429 0.61897 1.00153 11.900 0.000
12 1.0465 1.0484 1.04643 0.00713 19.4241 19.4241 193.840 211.883 18.0429 0.61900 1.00153 12.000 0.000
12.1
12.2
12.3
12.4
12.5
1.0469
1.0473
1.0477
1.0481
1.0486
1.0488
1.0492
1.0496
1.0501
1.0505
2
1.04685
1.04727
1.04769
1.04811
1.04853
0.00719
0.00726
0.00733
0.00739
0.00746
19.4371
19.4501
19.4632
19.4763
19.4894
19.4371
19.4502
19.4632
19.4763
19.4894
193.849
193.858
193.867
193.876
193.885
211.892
211.901
211.910
211.919
211.928
18.0428
18.0427
18.0427
18.0426
18.0425
0.61903
0.61905
0.61908
0.61911
0.61913
1.00153
1.00152
1.00152
1.00152
1.00151
12.100
12.200
12.300
12.400
12.500
0.000
0.000
0.000
0.000
0.000
12.6 1.0490 1.0509 1.04895 0.00753 19.5026 19.5026 193.894 211.937 18.0425 0.61916 1.00151 12.600 0.000
 Table 2: Data Calculation Using Equations 28, 46 and 47
′
Plato’s Table[10] SG20
20 ρ Xs Mtot Mtot (Xs ) Mtot Vs Vw vs vw E(v s , v w , SG20
20 ) ∆
E SG20
4
12.7 1.0494 1.0513 1.04937 0.00760 19.5157 19.5157 193.903 211.946 18.0424 0.61919 1.00151 12.700 0.000
12.8 1.0498 1.0517 1.04979 0.00767 19.5289 19.5289 193.912 211.955 18.0423 0.61921 1.00150 12.800 0.000

pr ua
12.9 1.0502 1.0522 1.05021 0.00773 19.5422 19.5422 193.921 211.964 18.0423 0.61924 1.00150 12.900 0.000
13 1.0507 1.0526 1.05064 0.00780 19.5554 19.5554 193.931 211.973 18.0422 0.61926 1.00149 13.000 0.000
13.1
13.2
1.0511
1.0515
Ja
1.0530
1.0534
1.05106
1.05148
0.00787
0.00794
19.5687
19.5820
19.5687
19.5820
193.940
193.949
211.982
211.991
18.0421
18.0420
0.61929
0.61932
1.00149
1.00149
13.100
13.200
0.000
0.000

e- ry
13.3 1.0519 1.0539 1.05190 0.00801 19.5954 19.5954 193.958 212.000 18.0420 0.61934 1.00148 13.300 0.000
13.4 1.0524 1.0543 1.05233 0.00808 19.6087 19.6088 193.967 212.009 18.0419 0.61937 1.00148 13.400 0.000
13.5 1.0528
n
1.0547 1.05275 0.00815 19.6222 19.6222 193.976 212.018 18.0418 0.61940 1.00147 13.500 0.000

pr 5
13.6 1.0532 1.0551 1.05317 0.00822 19.6356 19.6356 193.985 212.027 18.0418 0.61942 1.00147 13.600 0.000
13.7 1.0536 1.0556 1.05360 0.00829 19.6491 19.6491 193.994 212.036 18.0417 0.61945 1.00147 13.700 0.000
13.8 1.0541 1.0560 1.05402 0.00836 19.6626 19.6626 194.003 212.045 18.0416 0.61947 1.00146 13.800 0.000

in , 2
13.9 1.0545 1.0564 1.05445 0.00843 19.6761 19.6761 194.012 212.054 18.0415 0.61950 1.00146 13.900 0.000
14 1.0549 1.0568 1.05487 0.00849 19.6897 19.6897 194.021 212.063 18.0414 0.61953 1.00145 14.000 0.000
31

14.1 1.0553 1.0573 1.05530 0.00856 19.7032 19.7032 194.030 212.072 18.0414 0.61955 1.00145 14.100 0.000

tv 0
14.2 1.0558 1.0577 1.05572 0.00864 19.7169 19.7169 194.039 212.081 18.0413 0.61958 1.00144 14.200 0.000
14.3 1.0562 1.0581 1.05615 0.00871 19.7305 19.7305 194.049 212.090 18.0412 0.61961 1.00144 14.300 0.000
14.4 1.0566 1.0585 1.05657 0.00878 19.7442 19.7442 194.058 212.099 18.0411 0.61963 1.00144 14.400 0.000
14.5 1.0570 1.0590 1.05700 0.00885 19.7579 19.7579 194.067 212.108 18.0410 0.61966 1.00143 14.500 0.000
14.6 1.0575 1.0594 1.05743 0.00892 19.7716 19.7716 194.076 212.117 18.0410 0.61969 1.00143 14.600 0.000

1. 4
14.7 1.0579 1.0598 1.05785 0.00899 19.7854 19.7854 194.086 212.126 18.0409 0.61971 1.00142 14.700 0.000
14.8 1.0583 1.0603 1.05828 0.00906 19.7992 19.7992 194.095 212.136 18.0408 0.61974 1.00142 14.800 0.000
14.9 1.0587 1.0607 1.05871 0.00913 19.8131 19.8131 194.104 212.145 18.0407 0.61977 1.00141 14.900 0.000

7
15 1.0592 1.0611 1.05914 0.00920 19.8269 19.8269 194.114 212.154 18.0406 0.61979 1.00141 15.000 0.000
15.1 1.0596 1.0615 1.05956 0.00927 19.8408 19.8408 194.123 212.163 18.0406 0.61982 1.00140 15.100 0.000
15.2 1.0600 1.0620 1.05999 0.00935 19.8547 19.8547 194.132 212.173 18.0405 0.61985 1.00140 15.200 0.000
15.3 1.0605 1.0624 1.06042 0.00942 19.8687 19.8687 194.142 212.182 18.0404 0.61988 1.00139 15.300 0.000
15.4
15.5
15.6
15.7
15.8
1.0609
1.0613
1.0617
1.0622
1.0626
1.0628
1.0633
1.0637
1.0641
1.0646
2
1.06085
1.06128
1.06171
1.06214
1.06257
0.00949
0.00956
0.00963
0.00971
0.00978
19.8827
19.8967
19.9108
19.9249
19.9390
19.8827
19.8967
19.9108
19.9249
19.9390
194.151
194.161
194.171
194.180
194.190
212.192
212.201
212.211
212.220
212.230
18.0403
18.0402
18.0401
18.0400
18.0399
0.61990
0.61993
0.61996
0.61999
0.62002
1.00139
1.00138
1.00138
1.00137
1.00137
15.400
15.500
15.600
15.700
15.800
0.000
0.000
0.000
0.000
0.000
15.9 1.0630 1.0650 1.06300 0.00985 19.9532 19.9531 194.200 212.240 18.0398 0.62004 1.00136 15.900 0.000
 Table 2: Data Calculation Using Equations 28, 46 and 47
′
Plato’s Table[10] SG20
20 ρ Xs Mtot Mtot (Xs ) Mtot Vs Vw vs vw E(v s , v w , SG20
20 ) ∆
E SG20
4
16 1.0635 1.0654 1.06343 0.00993 19.9673 19.9673 194.210 212.249 18.0397 0.62007 1.00136 16.000 0.000
16.1 1.0639 1.0659 1.06386 0.01000 19.9815 19.9815 194.220 212.259 18.0396 0.62010 1.00135 16.100 0.000

pr ua
16.2 1.0643 1.0663 1.06430 0.01007 19.9958 19.9958 194.230 212.269 18.0395 0.62013 1.00135 16.200 0.000
16.3 1.0648 1.0667 1.06473 0.01015 20.0101 20.0101 194.240 212.279 18.0394 0.62016 1.00134 16.300 0.000
16.4
16.5
1.0652
1.0656
Ja
1.0672
1.0676
1.06516
1.06559
0.01022
0.01029
20.0244
20.0387
20.0244
20.0387
194.250
194.260
212.289
212.299
18.0393
18.0392
0.62019
0.62022
1.00134
1.00133
16.400
16.500
0.000
0.000

e- ry
16.6 1.0661 1.0680 1.06602 0.01037 20.0531 20.0531 194.270 212.309 18.0391 0.62025 1.00132 16.600 0.000
16.7 1.0665 1.0685 1.06646 0.01044 20.0675 20.0675 194.281 212.320 18.0390 0.62028 1.00132 16.700 0.000
16.8 1.0669
n
1.0689 1.06689 0.01052 20.0820 20.0820 194.291 212.330 18.0389 0.62031 1.00131 16.800 0.000

pr 5
16.9 1.0674 1.0693 1.06733 0.01059 20.0964 20.0964 194.302 212.340 18.0388 0.62034 1.00130 16.900 0.000
17 1.0678 1.0698 1.06776 0.01066 20.1110 20.1110 194.312 212.351 18.0387 0.62037 1.00130 17.000 0.000
17.1 1.0682 1.0702 1.06819 0.01074 20.1255 20.1255 194.323 212.361 18.0386 0.62040 1.00129 17.100 0.000

in , 2
17.2 1.0687 1.0706 1.06863 0.01081 20.1401 20.1401 194.334 212.372 18.0384 0.62043 1.00129 17.200 0.000
17.3 1.0691 1.0711 1.06906 0.01089 20.1547 20.1547 194.344 212.383 18.0383 0.62046 1.00128 17.300 0.000
32

17.4 1.0695 1.0715 1.06950 0.01097 20.1693 20.1693 194.355 212.394 18.0382 0.62049 1.00127 17.400 0.000

tv 0
17.5 1.0700 1.0719 1.06993 0.01104 20.1840 20.1840 194.367 212.405 18.0381 0.62053 1.00127 17.500 0.000
17.6 1.0704 1.0724 1.07037 0.01112 20.1988 20.1987 194.378 212.416 18.0380 0.62056 1.00126 17.600 0.000
17.7 1.0708 1.0728 1.07081 0.01119 20.2135 20.2135 194.389 212.427 18.0378 0.62059 1.00125 17.700 0.000
17.8 1.0713 1.0733 1.07124 0.01127 20.2283 20.2283 194.400 212.438 18.0377 0.62062 1.00124 17.800 0.000
17.9 1.0717 1.0737 1.07168 0.01134 20.2431 20.2431 194.412 212.449 18.0376 0.62066 1.00124 17.900 0.000

1. 4
18 1.0721 1.0741 1.07212 0.01142 20.2580 20.2580 194.424 212.461 18.0374 0.62069 1.00123 18.000 0.000
18.1 1.0726 1.0746 1.07256 0.01150 20.2729 20.2728 194.435 212.473 18.0373 0.62072 1.00122 18.100 0.000
18.2 1.0730 1.0750 1.07299 0.01157 20.2878 20.2878 194.447 212.484 18.0372 0.62076 1.00121 18.200 0.000

7
18.3 1.0735 1.0755 1.07343 0.01165 20.3027 20.3027 194.459 212.496 18.0370 0.62079 1.00121 18.300 0.000
18.4 1.0739 1.0759 1.07387 0.01173 20.3177 20.3177 194.471 212.508 18.0369 0.62083 1.00120 18.400 0.000
18.5 1.0743 1.0763 1.07431 0.01181 20.3328 20.3328 194.484 212.520 18.0368 0.62086 1.00119 18.500 0.000
18.6 1.0748 1.0768 1.07475 0.01188 20.3478 20.3478 194.496 212.533 18.0366 0.62090 1.00118 18.600 0.000
18.7
18.8
18.9
19
19.1
1.0752
1.0757
1.0761
1.0765
1.0770
1.0772
1.0777
1.0781
1.0785
1.0790
2
1.07519
1.07563
1.07607
1.07651
1.07695
0.01196
0.01204
0.01212
0.01219
0.01227
20.3629
20.3781
20.3932
20.4085
20.4237
20.3629
20.3781
20.3932
20.4084
20.4237
194.508
194.521
194.534
194.547
194.560
212.545
212.557
212.570
212.583
212.596
18.0364
18.0363
18.0361
18.0360
18.0358
0.62094
0.62097
0.62101
0.62105
0.62108
1.00117
1.00117
1.00116
1.00115
1.00114
18.700
18.800
18.900
19.000
19.100
0.000
0.000
0.000
0.000
0.000
19.2 1.0774 1.0794 1.07739 0.01235 20.4390 20.4390 194.573 212.609 18.0357 0.62112 1.00113 19.200 0.000
 Table 2: Data Calculation Using Equations 28, 46 and 47
′
Plato’s Table[10] SG20
20 ρ Xs Mtot Mtot (Xs ) Mtot Vs Vw vs vw E(v s , v w , SG20
20 ) ∆
E SG20
4
19.3 1.0779 1.0799 1.07783 0.01243 20.4543 20.4543 194.587 212.622 18.0355 0.62116 1.00112 19.300 0.000
19.4 1.0783 1.0803 1.07827 0.01251 20.4696 20.4697 194.600 212.636 18.0353 0.62120 1.00111 19.400 0.000

pr ua
19.5 1.0787 1.0807 1.07871 0.01259 20.4850 20.4850 194.614 212.649 18.0351 0.62124 1.00110 19.500 0.000
19.6 1.0792 1.0812 1.07916 0.01267 20.5005 20.5005 194.628 212.663 18.0349 0.62128 1.00109 19.600 0.000
19.7
19.8
1.0796
1.0801
Ja
1.0816
1.0821
1.07960
1.08004
0.01275
0.01283
20.5159
20.5314
20.5159
20.5314
194.642
194.657
212.677
212.691
18.0348
18.0346
0.62132
0.62136
1.00108
1.00107
19.700
19.800
0.000
0.000

e- ry
19.9 1.0805 1.0825 1.08049 0.01291 20.5470 20.5470 194.671 212.706 18.0344 0.62140 1.00106 19.900 0.000
20 1.0810 1.0830 1.08093 0.01299 20.5625 20.5626 194.686 212.720 18.0342 0.62145 1.00105 20.000 0.000

n
pr 5
MEAN 0.61827 1.00155 0.000
MEDIAN 0.61845 1.00160
σ 0.00187 0.00021

in , 2
33

tv 0
1. 4
7
2
 E Figures

pr ua
Ja
e- ry
n
pr 5
in , 2
34

tv 0
1. 4
7
Figure 1: Graph comparing the various approximation equations, overview.

2
 pr ua
Ja
e- ry
n
pr 5
in , 2
35

tv 0
1. 4
7
2
Figure 2: Graph comparing the various approximation equations, region of interest.
 F List of Captions
1. Table. 1 This table shows calculations using the approximate formulas (36),
(37) and (40) for each SG equal to 1.000, 1.001, 1.002, etc. up to 1.083 in
the ASBC dataset [2].

2. Table. 2 This table summarizes the data calculations for the full dataset
using Equations 28, 46 and 47 as described in section 2.4.

3. Fig. 1 Graph comparing the various approximation equations, overview.

20 7
4. Fig. 2 Graph comparing the various approximation equations, region of in-

5, 1.
terest.

24
y v
ar int
nu pr
-
Ja re
p

36