Help mee aan het verbeteren van het forum via een donatie. Klik hier,
Gestart door Hoppenbrouwers, 20-10-2013 07:27 u
0 leden en 1 gast bekijken dit topic.
CiteerMorey's Equation:It has been well documented that beer color is not a linear scale. So why try to relate Malt Color Units to an estimated SRM by a linear equation? I propose a power equation would better approximate the color phenomenon. What I did was: Assume MCU = SRM from 0 to 10.5, Daniel's equation holds in the range of 10.5 < MCU < 37, and Mosher's formula holds for MCU < 37. Then I plotted In (MCU) vs. In (SRM) and performed linear regression. This exercise resulted in a R^2 value of 0.97 (This is better than my linear fit with force zero intercept on the MCU vs. SRM graph - R^2 0.96). My resulting equation is: SRM = 1.4922 [(MCU) ^ 0.6859] - for values of SRM < 50 Now there is a single equation that can be used to estimate the final color of beer. Also, in the age of calculators and computers, it is no more difficult to calculate than a linear equation. Mosher, Daniels Formula:From "Beer Color Demystified" - Part III. Ray Daniels, Brewing Techniques Vol. 3, #6 - November/December 1995 Issue. 1) Calculate MCU, which is basically: MCU = (Lovibond rating x pounds) / gallons. Each grain in the grist gets this number calculated, the totals are added, and we get a final MCU number.THEN:2) Randy Mosher compared commercial beers with known rating to his own, and came to the conclusion that SRM could be approximated using the formula: SRM = (MCU x 0.3) + 4.7Although not perfect, when graphically plotted his scale runs right through the middle of the data set.3) In Ray Daniels 'Beer Color Demystified' (see reference in ProMash), Ray does further refinement on Mosher's equation and stated that SRM is approx: SRM = (MCU * 0.2) + 8.4and when graphically plotted, looks a tad better than Mosher's equation.
Citaat van: Hoppenbrouwers op 20-10-2013 18:26 u@Robin: heb je een idee hoe dat zou kunnen dat we precies het tegenovergestelde ervaren?
Pagina opgebouwd in 0.086 seconden met 30 queries.