It wasn’t too long into making beer that I found that the original gravity or final volume was always lower than expected. The OG and volume should be pretty predictable, especially with extract brewing. By taking better measurements and computing additions and losses of sugars and water at each step, I’ve been able to get much closer to my predicted targets. It turns out that in my case the main culprit was how much wort was being absorbed by the hops and then discarded… and I was adding a lot of hops!
For those who might be interested, I present a bunch of numbers intended to help with predicting the effects of different brewing techniques.
|water absorbed by 80% 2-row, 20% specialty malts||0.147 G/lb = 0.59 qt/lb
0.210 G/lb = 0.84 qt/lb
(see text below; the larger number is probably the better estimate)
|wort left in bottom of 52-quart Coleman Xtreme cooler||0.21875 G (3.5 cups)|
|evaporation during steeping of grains at 160°F, uncovered||0.35 G/hr|
|evaporation during low rolling boil, uncovered, targeting ~10% total evaporation, with 10G pot and 7G of wort||about 1.0 G/hr|
|evaporation during uncovered hop stand (170°F to 180°F)||0.35 G/hr|
|water absorbed by hops during boil and hop stand||0.063 G/oz|
|water absorbed by hops after dry-hopping (in weighted mesh bag)||0.006 G/oz|
|PPG of crystal 20 malt||10 PPG|
|PPG of crystal 40 malt||15 PPG|
|wort and break lost during transfer from 10G pot, using “auto siphon” racking cane with no spacer.||0.30 to 0.45 G (extract)
0.15 G (all-grain)
|wort and break lost when racking from 6G “better bottle” carboy, using “auto siphon” racking cane with no spacer.||0.30 G|
|wort and break lost when racking from 5G glass carboy, using “auto siphon” racking cane with no spacer.||0.15 G|
|weight of 1 tsp of Calcium Chloride anhydrous (CaCl2)||4.462 g = 0.157 oz|
|weight of 1 tsp of Calcium Chloride dihydrate (CaCl2·2H2O)||5.035 g = 0.1776 oz|
|weight of 1 tsp of Calcium Carbonate (CaCO3)||2.825 g = 0.996 oz|
|weight of 1 tsp of Calcium Sulphate dihydrate (CaSO4·2H2O)||3.266 g = 0.115 oz|
|weight of 1 cup of Briess Pilsen Light Dried Malt Extract (DME)||5.1 oz|
Water Absorbed by Grains: The first set of numbers for the absorption rate of grains was computed from three measurements: (a) the amount of wort that dripped out of the cooler after I stopped collecting wort for the boil, (b) the weight of all of the wet grains, and (c) the dry-to-wet grain ratio, obtained by drying 2.0 lbs of the wet grains and measuring the dry weight of the spent grains. I collected 9 ⅓ cups of wort just letting it drip out slowly into a container, or 0.5833 G. The 15 original pounds of grain weighed 21.625 lbs (346 oz) when spent and wet. The 2 lbs of wet, spent grain weighed 0.575 lbs (9.2 oz) after being dried in the oven for two hours. So, the dried, spent grain weight is 28.75% of the wet, spent grain weight. For the total of 15 lbs, that would yield 6.217 lbs of dried, spent grain. The amount of water in the grains was therefore 21.625 lbs – 6.217 lbs = 15.408 lbs of water. With the weight of water at 8.34 lbs/G, that’s 1.8475 G of water absorbed by the spent grains. (This means that 0.49 qt/lb was retained by the grains (and didn’t drip out), which may be where Palmer obtained his 0.5 qt/lb estimate.) If we add 1.8475 G that was retained by the grains to the 0.5833 G that dripped out from the cooler, and subtract the 0.21875 G of water that simply collects at the bottom of the cooler without being absorbed, we get 2.212 G of water that were absorbed by the grains when I stopped collecting wort. This translates into 0.59 qt/lb or 0.147 G/lb absorbed by the malt (2.212 G x 4 / 15 lbs). This is in between the 0.5 qt/lb reported by Palmer in “How to Brew” (p. 184) and the 0.8 qt/lb reported by Daniels in “Designing Great Beers” (p. 64).
Note that the dried, spent grain weight, 6.217 lbs, is 41.44% of the original grain weight of 15 lbs, which accords well with Daniels in “Designing Great Beers” on page 64, in which he says “the postmash grain mass is about 40% of the weight of the grain you added.”
I obtained a second estimate for the absorption rate of grains in a much more direct manner. I added 6.2 G of water to 13.75 lbs of grains and measured the runoff at slightly less than 4.10 G. Assuming that the value for the amount of wort left in the bottom of the cooler is correct (~0.22 G), that might mean that 1.89 G remained in the wort, or approximately 0.55 qt/lb. However, this method of computation is flawed. When you add 6.2 G of water to grains, you don’t get 6.2 G of wort; you get a larger volume of wort because of the sugars extracted from the grain contribute to the volume. I’ve measured 0.059 gallons worth of dissolved sugar per pound of dried malt extract, and 0.074 gallons worth of dissolved sugar per pound of grain. This would add about 1 G of liquid, resulting in about 7.2 G of wort in this example. With runoff of 4.10 G and known loss of 0.22 G, this would mean grain absorption of 0.84 qt/lb, which is close to the value reported by Daniels. Why did the two methods yield such different results (0.59 qt/lb and 0.84 qt/lb)? The lower number represents the amount that is retained by the grains if all of the excess wort is drained off. The higher number represents the amount that is retained by the grains with some amount not quite absorbed but not drained in a reasonable time. Therefore, I think the 0.84 qt/lb is the better number to use in brewing.
Evaporation Rates: The evaporation rates were obtained with a measuring stick (i.e. a long aluminum rod) calibrated to my pot with ¼- and ⅓-gallon markings. I’ve found the strength of the boil a little difficult to “dial in” with my propane burner and outdoor brewing. Several authors recommend a “vigorous” boil, going so far as to say “boil it as hard as you can” (Mosher, “The Brewer’s Companion”, p. 136). Lewis and Young say that “efficient boiling requires a ‘full rolling boil’ meaning intense and rapid motion as well as evolution and removal of steam (“Brewing”, 2nd edition, p. 272). Miller says to “maintain a vigorous rolling action” (p. 145), but also cautions that “overboiling can create harsh bitterness in your beer.” (p. 186). Others also caution about too vigorous a boil, noting for example that this will cause “more color development, melanoidin production, … [and] can also cause the hop bitterness to take on additional harshness” (Strong, “Brewing Better Beer”, p. 61). Fix is the only source I know of who has quantified this, saying that “volume reduction should be at least 7%. … Evaporation rates above 12% may produce level 2 [of non-enzymatic browning], leaving vegetal malt tones that are accompanied with some astringency. A wide range of [non-enzymatic browning] is possible once evaporation rates exceed 15%.” (Fix, “Principles of Brewing Science”, p. 78). He summarizes this as “The best general recommendation is an evaporation rate of 9 to 11%. This can usually be achieved with a ninety-minute boil” (Fix and Fix, “An Analysis of Brewing Techniques”, p. 54). I’ve found it quite easy to go beyond a 15% reduction in volume in a sixty-minute boil with my propane burner and an uncovered pot, and in trying to keep the volume reduction to around 10% in sixty minutes, I aim for a “low rolling boil”, in which there are many small bubbles but few big bubbles.
I’ve found that, other than the strength of the flame, the biggest impact on evaporation rate is the size of my pot and amount of wort. In my 10G pot with about 7G of wort, I observe an evaporation rate around 1.0 G/hr. In my smaller 5G pot with 4 G of wort, I observe an evaporation rate around 0.8 G/hr. In my smaller 5G pot with 1 to 2 G of wort, I observe an evaporation rate of around 0.55 G/hr.
I’ve developed a model for evaporation rates at sub-boiling temperatures, available at my web site by clicking on the “Cooling and Evap.” button on the left-hand side.
Water Absorbed by Hops: The absorption rate of hops during the boil was computed by weighing the wet hops after removal from the wort, converting weight to gallons (approximating the weight of wort as about that of water, 8.34 lbs/gallon), and dividing the total number of gallons by the total ounces of hops.
The absorption rate after dry hopping was computed in the same way, but using the hops after removal from the secondary fermenter. This absorption rate is quite low because of the technique I use that removes much of the beer from the hops, described in Dry Hopping in a Weighted Mesh Bag.
Yields of Steeped Malts: The yields of crystal 20 and crystal 40 are much lower than the values reported by Palmer in his third edition. The steeping was performed by soaking the grain in 2 gallons of water per pound (see http://beersmith.com/blog/2009/03/22/steeping-grains-for-extract-beer-brewing/) for 30 minutes at 160°F. I’m not sure why my measured yields are so much lower than Palmer’s.
Wort and Break Loss: These numbers are the volumes of wort, hot break, and cold break left in the bottom of a vessel after racking with a 1/2” “auto siphon” without the spacer at the bottom (in order to minimize the loss). I have noticed that brewing with the combination of extract (Briess liquid malt extract) and steeped grains leaves much more break in suspension, compared with all-grain brewing. The all-grain brews tend to settle more quickly, leave a denser layer on the bottom, and produce a clearer beer in the end. As a result, I feel that with extract+steep brewing I need to leave more wort behind during the first transfers. (I have brewed with Briess LME and no steeped grains, and in those cases the wort has settled nicely and a clear beer was produced; the extract itself doesn’t seem to be an issue, but the addition of steeped grains.)
Weight of Salts: I used salts marketed by LD Carlson Company and MoreFlavor. (As determined from the Material Safety Data Sheets (MSDS) available online, the CaCl2 from LD Carlson Company and Brewcraft USA are anhydrous, i.e. CaCl2 instead of the CaCl2·2H2O used by Palmer and sold by MoreFlavor, which changes the amount of each ion contributed by weight of salt. The CaSO4 from all three companies is CaSO4·2H2O, the same as used by Palmer.) Using a jewelry scale, I weighed 5 teaspoons in ¼-teaspoon increments (since I add to the mash in ¼-tsp units), and divided by 5 to get a weight per teaspoon. I repeated the measurements by measuring 5 teaspoons in 1-teaspoon increments and then dividing by 5. (All measurements were level teaspoons or ¼-teaspoons.) The resulting per-teaspoon values were close. I then averaged these two measurements per salt. The values reported here are, for reasons not entirely clear to me, quite different from those reported by Palmer in “How to Brew” (p. 166). If you’re trying to predict mash pH or know how much of a salt you’re adding, you may want to weigh your salts yourself for greater confidence. (A wine-making website that is no longer available gave 2.6 g/tsp for CaCO3, which is an in-between value that might be good to use as a compromise.)
Weight of DME: I measured out Briess Pilsen Light DME into a 1-cup measuring cup three times, and took the average. Each cup was filled and level with malt extract. A weight of 5.1 oz translates into 3.1 cups per pound, which is in between the 4⅛ reported by homebrew4less (no longer available online), the 2.4 reported by brewersdirect, and the 2.75 reported by BYO.
Many measurements in this table were not only averaged over a few sessions, but also adjusted in order to make the entire set of measurements be in accordance with predicted measurements.