SRM Formula
- Thread starter ryandlf
- Start date
Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:
Pretty much everything uses the Morey equation. Last time I played with ProMash (which was, admittedly, some time ago), I believe that's what was being used. Are you sure you're doing the math properly?
How would you know if one formula is more accurate than another? The only way would be to measure the SRM of each beer you brew and compare the measured value against what each of the various formulae predicted and choose the best. You would not be impressed with the accuracy of even the best in which the parameters are adjusted to minimize the mean square error. This is because all the formulae (that I know of) are some function of the sum to the products of the masses of grains of various colors multiplied by the "color" of the grain. The color of the grain is the Lovibond color of a Congress mash made with the grain. A Congress mash is appreciably different from a mash tun mash. For example, it uses distilled water. It is well known that bicarbonate in the mash water results in darker beer so the Congress mash underestimates the color of every grain used relative to what you will experience with those same grains if your water is at all alkaline. The next thing to consider is that Lovibond color is not the same as SRM color though it is, in modern Lovibond Tintometers, at least measured photometrically as opposed to the old method which utilized visual comparisons to a set of tinted glasses. The mapping between Lovibond and SRM is a mystery though the SRM was originally defined in order to make SRM numbers and Lovibond numbers the same but the original work defining SRM was based on an ensemble of beers none of which was darker than about 7 SRM. Thus °L an SRM may compare for light colors but all bets are off when the colors are more intense.
The sigma(wt*°L) concept assumes that Beer's Law applies. The evidence suggests that where the beer has "average spectral characteristics" it does but not all beers do(in fact none do though many approximate the average quite closely).
The function of MCU [ SRM = f(MCU) ] model does not take into account things like the aforementioned bicarbonate content of the water, the mashing method and the length of the boil, all of which have considerable effect on the color of the beer. A triple decoction mashed beer with kettle time of 2 hours will be a lot darker than a Kölsch which has been made with a single step infusion mash and been boiled 1 hour.
So you might as well just use the Morey model because it is the one that most other home brewers use. There is a certain element of robustness in it in that I believe he took photometer measured SRM values and estimated the values of the parameters in the model that best matched the data. I did the same thing and came up with different parameters because my brewing methods and equipment are different from his. The Morey model with the parameters estimated from measurements on my beers give me the best predictions but as I noted earlier they are not good enough to be taken seriously.
The sigma(wt*°L) concept assumes that Beer's Law applies. The evidence suggests that where the beer has "average spectral characteristics" it does but not all beers do(in fact none do though many approximate the average quite closely).
The function of MCU [ SRM = f(MCU) ] model does not take into account things like the aforementioned bicarbonate content of the water, the mashing method and the length of the boil, all of which have considerable effect on the color of the beer. A triple decoction mashed beer with kettle time of 2 hours will be a lot darker than a Kölsch which has been made with a single step infusion mash and been boiled 1 hour.
So you might as well just use the Morey model because it is the one that most other home brewers use. There is a certain element of robustness in it in that I believe he took photometer measured SRM values and estimated the values of the parameters in the model that best matched the data. I did the same thing and came up with different parameters because my brewing methods and equipment are different from his. The Morey model with the parameters estimated from measurements on my beers give me the best predictions but as I noted earlier they are not good enough to be taken seriously.
