Most homebrewers make their beer on their stovetop, boiling 2–3 gallons (7.5–11 L) of wort then diluting it with water in their fermenter to make 5 gallons (19 L) of beer. This method has been used for decades and has two main advantages — it’s quick and requires very little specialized equipment. The problem with this method is that boiling a concentrated wort causes sugars in the wort to caramelize and brewing a light-colored beer is almost impossible. In addition, because hop utilization decreases with increasing wort gravity, extract beers made this way frequently are not bitter enough.
One way to solve the problems of the standard method is to perform a full-wort boil. In other words, boil enough wort so that you end up with 5 gallons (19 L) after boiling and cooling without having to add any water. In order to do this, however, you need a kettle large enough to hold the full volume of wort plus have enough space to accommodate foaming. You also need a heat source capable of bringing this wort to a boil and a method of cooling it once the boil is over.
Over the past two years, Brew Your Own has introduced two new methods of extract homebrewing — the extract late method and the Texas two-step method. These methods do not require any equipment beyond that used in the standard method, but purport to offer better results.
The extract late method was presented by Steve Bader in the October 2002 issue (“Boil the Hops, not the Wort”) and involves withholding half or more of the malt extract in the recipe and adding it at or near the end of the boil. Liquid malt extract has already been boiled in its production, so why boil it again? In his article, Steve claimed that homebrewers could make lighter, more hoppy homebrew than with the standard method of boiling a concentrated wort.
The Texas two-step method was presented by me in the October 2003 issue and involves making two half-sized batches of wort and combining them. In my article, I claimed that the Texas two-step afforded stovetop brewers all the benefits of a full-wort boil, including the ability to make lighter-colored, more hoppy homebrew. The tradeoff in the Texas two-step is that it takes longer to brew your beer compared to the standard or extract late methods.
Since the publication of these articles, the response from homebrewers has been very positive. Stovetop brewers have been able to brew lighter, hoppier homebrews and have given the thumbs up to both methods. However, until now, nobody has tested all the various methods head to head (to head to head) and compared the results. (See the Step by Step instructions on page 38 for how, exactly, all the beers were brewed in this experiment.)
What I Did
In my experiment, I drew up an extra pale ale recipe and brewed it twice using each of the four methods mentioned (for a total of eight 5-gallon (19-L) batches of beer). The recipe was designed to make a light-colored, hoppy beer. I did not try to make the lightest, most hoppy beer possible. Rather, I tried to make a drinkable beer in which the differences between the methods would be obvious, if they existed. The recipe was the basically the same for all eight batches, with one exception. I did the first set of four brews using an “extract with grains” formulation of the recipe and the second set of four brews with a “partial mash” formulation. The only difference between the two was a small amount of extract was swapped for some pale malt in the “partial mash” formulation. (See the recipes on page 38 for the exact details.)
The ingredients were bought in bulk and all the malt extract — an extra-light liquid malt extract — came from the same drum at my local homebrew shop. Most of the hops came from two bulk bags of pellet hops, although I did need to buy two smaller bags to get the total amount. The smaller bags were of the same type of hop and rated at the same alpha acid rating. The crystal malt was the same for all the beers. All of the yeast tubes purchased were fresh and had the same manufacture date except for one that was a week older than the rest. I used this tube to make a yeast starter for one of the full-wort boil beers.
I weighed the grains on my kitchen scale. For the hops, I used a laboratory three-beam balance. The amounts of each ingredient that went into each batch were well controlled. I also calibrated each bucket and marked the 5-gallon (19 L) fill line to ensure the volume of beer was the same for all beers (within reason). The fact that all the beers registered the same original gravity demonstrated that errors in ingredient measurements did not likely influence the results.
For all four methods, the specialty grains were crushed by my homebrew mill immediately before brewing. The grains were steeped in the same soup pot at the same temperature and volume of water. The three stovetop beers were all boiled at the same volume in the same 5-gallon (19-L) stainless steel brewpot on the same burner of my stove. The full wort boil was done outside using a propane burner and 10-gallon (38-L) stainless steel brewpot. (I tried to eyeball the wort as it boiled and adjust the full wort boil to match, as best as I could, the boil vigor of the stovetop methods.) The amount of hops added and the timing of the hop additions were the same for all beers. All of the beers were cooled with the same copper immersion chiller. (Since there was more wort to chill in the full wort boil beers, cooling time for this beer was slightly longer.)
Each set of four beers were siphoned to identical buckets (with identical stoppers and fermentation locks). For the three stovetop beers, I pitched one tube of White Labs yeast. For the full-wort boil, I pitched the yeast from a ~2 qt. (2L) starter. As each set of four beers was brewed in a single weekend, they were all fermented at the same time and in the same chest freezer for temperature control. Likewise, each set of four beers was racked to secondary and kegged at the same time.
Experiment in a nutshell
So basically, all eight beers were made with the same recipe, the same ingredients, the same equipment (where possible) and by the same brewer (me). I did have some help on the second brewing weekend and I would like to thank Charles and Theresa Culp and Marc and Janet Martin of the Austin ZEALOTS for their assistance.
The only differences in procedures were the differences in the four methods. And, since each method was tried twice, I could judge if the differences between the methods were repeatable.
I suppose some people might question if I was biased and — consciously or unconsciously — favoring “my” method (the Texas two-step). All I can say to that is I believe I did the experiment fairly and encourage other interested homebrewers or homebrew clubs to try to replicate, and perhaps expand upon, these results.
What I Found
I took a variety of data for this experiment, but the main variables of interest are color, bitterness and flavor. I will discuss these in some detail. Table 1 (on page 38) lists some of the other results from the experiment.
One of the easiest aspects of the experiment to judge was the color of the resulting beers. As you can see in the photo above, there are obvious differences in color between the four methods. The two extract late beers were the lightest colored beers. The standard method produced the two darkest beers and the four other beers were all fairly similar in color. The Texas two-step beers were a bit lighter than the beers made with a full wort boil, but not by much.
So, as expected, adding your malt extract late in the boil does lead to lighter colored homebrew. In fact, this method produced the lightest two beers of the bunch. Performing a full wort boil — either all at once in a large brewpot or by brewing two stovetop batches as in the Texas two-step — also produced lighter-colored homebrew than the standard method. Why the full-wort boil batches and Texas two-step batches varied in color (if only slightly) is a bit of a mystery. In both cases, the entire wort was boiled at the same density. However, the Texas two-step brews were boiled on the stove and the others by a propane burner. I suspect that the boil intensity was a bit higher for the full-wort boil beers.
Given the results of this experiment, I think it’s reasonable to suggest that if you are interested in brewing the lightest possible homebrews, you should use the extract late technique.
The level of hop bitterness was the second major variable in which the methods were supposed to differ. Comparing two beers and determining which is more bitter can be difficult if the beers are of different strengths or styles. However, if all the beers sampled are of the same style — or, as is the case here, made from the same recipe — comparing relative levels of bitterness is not that difficult.
The most obvious difference among the beers was that the two beers made with the standard method were far less bitter than the others. The level of bitterness in the remaining six beers was fairly close. The extract late beers seemed the most bitter and the full-wort boil beers in turn seemed to have slightly more bitterness than the Texas two-step. Again, it is somewhat puzzling that the full-wort boil beers and Texas two-step beers would be different given that the details of the boil were the same except that the Texas two-stepper were made in two halves. However, I again suspect that the boil vigor is responsible for this small difference. Also, the full-wort boil beers took a bit longer to cool and this may have led to a few extra IBUs being extracted from the hops.
The reason that the extract late beers were the most bitter of the bunch may lie in the details of how the extract was added. In the extract late beers, the extract was added at knockout and the wort sat for 15 minutes before it was cooled. Thus, the hops were exposed to 15 more minutes of heat compared to the other methods. In retrospect, I should have let the full-wort and Texas two-step worts sit an equivalent amount of time before cooling. I suspect the differences in bitterness among the extract late, Texas two-step and full wort boil brews would disappear if the knockout procedure were the same for all the methods.
The results of this experiment show that if you are trying to brew a bitter beer, avoid the standard method. You are better off using either the extract late, Texas two-step or full wort boil method. In addition, logic suggests that if you are trying to brew a very bitter beer that you should try the Texas two-step or full-wort boil. This latter suggestion, however, was not tested in this experiment.
The most important result of the experiment was how the beers made with the various methods tasted. This is also, of course, the most subjective. I tasted the beer along with a few of my homebrew club members and there was no clear consensus. None of the methods seemed to yield any off flavors or aromas and all the beers tasted like fairly decent homebrew. (If I brewed any of these again, I would make some tweaks to get a better balance in some of the beers, but nothing major.) The extract late, Texas two-step and full wort boil beers were preferred over the standard method beers, mostly (or entirely) because they were more hoppy. The partial mash formulations were generally favored over the extract with grains formulations for each beer type.
The most interesting finding was that the extract late beer had a slightly different malt character than the other beers. The different character wasn’t good or bad, just different. Some people preferred it to the Texas two-step and full wort boil; others preferred the other two methods. If I had to describe the character, which showed in both extract late batches, I’d call it very slightly “worty.” I thought the character would work well in malty, full-bodied beers, but less well in drier beers (But of course, this experiment didn’t test this.)
The results of the taste testing show that none of the methods produces faulty homebrew. The differences in preferences between the beers was mostly due to differences in the malt/hop balance in the beers, and this can be adjusted (within limits) with any of the methods. This suggests that you should use color and target bitterness as your main concerns when deciding on which extract method to use for any given beer. If you can hit the color and bitterness you want, all the available methods will produce palatable beer. See Table 2 at the bottom of this story for a set of recommendations regarding which methods to use for various types of beer.
Table 1 at the bottom lists some of the other results of the experiment. The specific gravity the wort is boiled at is given as well as the amount of IBUs required to hit the target IBU value when the wort is diluted (or not) to 5 gallons (19 L).
The Texas two-step was the most time consuming of the methods, with its two separate brewing sessions. The standard and extract late were the least time consuming. The full wort boil took a little more time than these methods due to the time it took to cool the larger volume of wort. (It took a little longer to heat the water, too, although I could have turned up the propane burner.)
There are any number of other variables I could have collected data on, clarity and head retention being two of these. Although I didn’t formally look at these (and other) variables, I kept an eye out for differences that jumped out between beers. However, nothing really jumped out that separated one method from the others. For example, all the beers seemed about equally clear and yielded a reasonable head. As I said in the section on flavor, all the methods made some fairly decent homebrew.
What I Should Have Done
If I could go back time in time and redesign the experiment, I would add a lot more bittering hops. I believe that, at high hopping rates, I would have found a difference in bitterness between the extract late method and the full wort boil method. In both cases, the hops would be boiled in wort of the same density. However, in the extract late method, the alpha acids would be diluted when the wort was added to the fermenter. For example, let’s say I was trying to make a beer with 80 IBUs. Using the full wort boil method, the 5 gallons (19 L) of wort would need to reach 80 IBUs by the end of the boil. In the case of the extract late method, however, the 2.5 gallons (9.4 L) of wort would need to reach 160 IBUs by the end of the boil and then be diluted to 80 IBUs in the fermenter. The problem is, the theoretical maximum level of IBUs you can obtain by boiling hops is thought to be around 120 IBUs. Thus, for beers over 60 IBU, you would need to do a full wort boil (or use the Texas two-step and boil half of the hops in each 2.5 gallon (9.4 L) step). With the low hopping rate of this experiment, however, I wasn’t able to confirm this suspicion.
This experiment shows that the method with which you brew your extract beers influences their color and bitterness. Although all of the methods produce drinkable homebrew without any off flavors or off aromas, some methods may be more appropriate to use when light color or higher levels of bitterness are a priority.
The standard method of extract homebrewing produces decent beer, but frequently that beer is too dark and not hoppy enough. For some styles of beer — such as Scottish ale or brown ale — it may work well. For any beer that needs to be light in color or show much bitterness, however, there are better choices.
For most beer styles, the extract late method should work well. This method requires no equipment beyond what is needed in the standard method and should allow extract brewers to brew a wide range of beer styles. Like the standard method, the extract late method is also very quick. Even for beers that are not particularly light or bitter, the method has the benefit of exhibiting better hop utilization.
The Texas two-step and the full-wort boil method can also produce light-colored, bitter homebrew. The Texas two-step, however, takes more time to brew and the full-wort boil method requires equipment beyond what it takes to brew a stovetop batch of beer. However, both of these methods may be capable of producing the most bitter extract brew (although this experiment does not test this idea).
This experiment focused mainly on color and bitterness in a pale ale. However, there may be other differences between beers made with these methods that did not show up in this experiment. It’s always possible that different recipes, beer styles, ingredients, accompanying procedures or equipment may cause different results for some brewers. So, always keep in mind that you have options when you brew extract beers.
Take good notes when trying something new while brewing and maybe you’ll find the perfect way to brew your favorite beer on your system.
Experimental Extract Extra Pale Ale
(5 gallons/19L, extract with grains)
OG = 1.051 FG = 1.013
IBU = 48 SRM = 7 ABV = 4.9%
7.25 lbs. (3.3 kg) extra pale liquid malt extract
0.66 lbs. (0.30 kg) crystal malt (20 °L)
12 AAU Willamette hops (60 mins)(2.4 oz./68 g of 5.0% alpha acids)
0.5 oz. (14 g) Cascade hops (15 mins)
1.5 oz. (42 g) Cascade whole hops (dry hops)
1 tsp. Irish moss (15 mins)
1/4 tsp. yeast nutrients (15 mins)
White Labs WLP001 (California Ale) yeast
Partial mash formulation:
Replace malt extract above with 6.6 lbs. (3.0 kg) extra pale liquid malt extract and 1.0 lb. (0.45 kg) 6-row pale malt.
Step by Step
Preparation For the two full-wort boil brews, a 2.0 qt. (1.9 L) yeast starter was made three days before brewday.
Brewday For the extract with grains formulation, the crystal malt was steeped in 1 qt 11 oz. (1.32 qt) of water at 150 °F
(66 °C) for 30 minutes for all the methods. For the partial mash formulation, the crystal malt and 6-row was steeped in 2.25 qts. (2.1 L) at 150 °F (66 °C) for 30 minutes. For all three stovetop methods, 2.5 gallons (9.5 L) were heated in my 5-gallon (19-L) brewpot. For the full-wort boil beers, 5 gallons (19 L) was heated in my 10-gallon (38 L) brewpot. In all cases, the “grain tea” from the steep or mash was added to the water in the brewpot and brought to a boil. The heat was then shut off and the extract was added. In the standard method beers, all the malt extract was added to the (just over) 2.5 gallons (9.5 L) in the brewpot. In the extract late and Texas two-step methods, half of the malt extract was added to the (just over) 2.5 gallons (9.5 L) in the brewpot. In the full wort boil, all of the malt extract was added to the (just over) 5 gallons (19 L) in the brewpot. In all cases, the hop and other additions were the same and done at the times listed in the ingredient list. All the beers were chilled with the same immersion chiller, which was submerged in the wort 20 minutes before the end of the boil. At knockout, the remaining malt extract was added to the extract late batches. The wort allowed to sit for 15 minutes before cooling. All other beers were chilled immediately after knockout. After cooling, the wort was allowed to settle for 15 minutes before siphoning to a bucket fermenter. If needed, distilled water was added to make 5 gallons (19 L) of wort — 2.5 gallons (9.5 L) in the case of the Texas two-step. In the standard, extract late and full wort boil methods, all 5 gallons (19 L) were aerated with oxygen for 60 seconds. For the Texas two-step batches, 2.5 gallons (9.5 L) of wort was aerated with oxygen for 30 seconds. In the standard, extract late and Texas two-step batches, one tube of White Labs yeast was pitched. In the full wort boil method, the sediment from the yeast starter was pitched. All buckets were placed in a chest freezer with the temperature set at 66 °F (19 °C).
Day Two The next day, the second half of the Texas two-step worts were made by boiling 2.5 gallons (9.5 L) of wort for 15 minutes. No hops were added. The wort was aerated again for 30 seconds with oxygen.
Later All beers were racked, kegged and carbonated the same way.