Growth rate for the two unstirred methods, in new billion cells per gram of
extract. These are flat across the whole 0–3.5 B/g range, and aren't a
fraction of the stirred curve.
still: 0.4 B/g. Kai's own figure, with his own caveat: "The 0.4 Billion
new cells per gram proposed here is a rough guess based on the data points
I have so far."
shaken: 0.62 B/g. This is the weakest number in this module. It isn't
in any Braukaiser text. It's Homebrew Dad's read-off of the unlabelled bar
chart on slide 17 of the NHC 2013 deck (still / shaken / stirred, all under
foil). We use it because it's the only published figure for the method,
and brewers without a stir plate do shake. Treat it as an order of
magnitude, not a measurement.
This gap is the case for a stir plate. At any ordinary inoculation rate a
stirred starter sits on the 1.4 plateau and makes three and a half times
the yeast a still one does from the same DME.
Why these are flat, and why the bands cross
Held to 3.5, a flat 0.62 crosses the stirred fit at
(2.33 − 0.62) / 0.67 = 2.55 B/g, and 0.4 crosses it at 2.88. Above those
rates the model says an unstirred starter grows more yeast than a stirred
one.
That isn't a transcription error, and we don't "fix" it. Kai
both states the flat range explicitly ("a growth rate of 0.4 Billion per gram
should be assumed over the full initial cell density range (0 – 3.5 B/g)")
and gives the mechanism for the shape: "This effect of dropping yeast yield
is not as pronounced for a non agitated culture … because in such a culture
not all yeast cells have the same access to wort nutrients due to
sedimentation." A stirred starter has every cell competing for the same
wort, so crowding has a bigger effect. He explicitly says the modes can't be
related by a scaling factor ("stirred and still starters are expected to
show different growth behavior that cannot be simply approximated by adding
a constant scaling factor"), which is why this is a Record of
measurements and not a multiplier on MAX_GROWTH_PER_GRAM.
The crossing is above 2.5 B/g, which is a starter pitched roughly twice as
thick as anything planStarter will build. It's a quirk at the edge
of the model, not a reason to switch the stir plate off.
Growth rate for the two unstirred methods, in new billion cells per gram of extract. These are flat across the whole 0–3.5 B/g range, and aren't a fraction of the stirred curve.
This gap is the case for a stir plate. At any ordinary inoculation rate a stirred starter sits on the 1.4 plateau and makes three and a half times the yeast a still one does from the same DME.
Why these are flat, and why the bands cross
Held to 3.5, a flat 0.62 crosses the stirred fit at (2.33 − 0.62) / 0.67 = 2.55 B/g, and 0.4 crosses it at 2.88. Above those rates the model says an unstirred starter grows more yeast than a stirred one.
That isn't a transcription error, and we don't "fix" it. Kai both states the flat range explicitly ("a growth rate of 0.4 Billion per gram should be assumed over the full initial cell density range (0 – 3.5 B/g)") and gives the mechanism for the shape: "This effect of dropping yeast yield is not as pronounced for a non agitated culture … because in such a culture not all yeast cells have the same access to wort nutrients due to sedimentation." A stirred starter has every cell competing for the same wort, so crowding has a bigger effect. He explicitly says the modes can't be related by a scaling factor ("stirred and still starters are expected to show different growth behavior that cannot be simply approximated by adding a constant scaling factor"), which is why this is a
Recordof measurements and not a multiplier on MAX_GROWTH_PER_GRAM.The crossing is above 2.5 B/g, which is a starter pitched roughly twice as thick as anything planStarter will build. It's a quirk at the edge of the model, not a reason to switch the stir plate off.