The promotional campaign is now in full swing with full-page

advertisements in major brew magazines. But dried yeast? No

way! After all, many of us in the American craft brewing

industry arrived at the professional ranks via the homebrew

route. We remember the bad old days when our dried yeast

pitches gave us randomly slow or stuck fermentations at best

and infected brews at worst, all for reasons that were largely

inexplicable to us then.

In those not too long ago times, we had no choice but to use

dried yeast. In fact, the selection in the brew stores was sparse

in general. There were perhaps a dozen different cans of

extract, a few bags of grain for steeping, a handful of hop vari-

eties (usually of undisclosed alpha-acid content) and two

yeast choices—ale and lager, both dried, of mysterious lineage

and packed in 7-gram pouches. We made beer with that stuff,

and, for an extra kick, we often threw a bag of corn or cane

sugar into the brew pot. The beer that emerged from these raw

materials was usually drinkable but also unremarkable, at

least by today’s standards.p

In 2

004,we

heard

“throu

ghthe

hop v

ine” th

ata F

rench

yeastmake

r called

S.I.

Lesaff

rewa

s goin

g to s

tart m

arketing

a sele

ction

ofdri

edbre

wersyea

stunder

thebra

ndname o

f Ferm

entis

toAm

erican

craft

brewe

rsand eve

ntuall

y to

homeb

rewers

aswe

ll.

In 2

004,we

heard

“throu

ghthe

hop v

ine” th

ata F

rench

yeastmake

r called

S.I.

Lesaff

rewa

s goin

g to s

tart m

arketing

a sele

ction

ofdri

edbre

wersyea

stunder

thebra

ndname o

f Ferm

entis

toAm

erican

craft

brewe

rsand eve

ntuall

y to

homeb

rewers

aswe

ll.[Note: Ferm

entis currently h

asfive

yeast strains

availab

le in 1

1-gram

homebrew-siz

e pack

ets.]

Editor’s

Note:Thisarticle

originallyappea

redin

the

Septem

ber/October

2005

issue

ofThe N

ewBre

wer.

Fortunately, when the craft and home-brew revolution gained critical mass,things improved dramatically. By theearly 1990s, only the brewer’s imagina-tion—and not the range of ingredients—spelled the limits of what could bebrewed at home or in a craft brewery.Importantly, brewers gained access to liq-uid yeast strains—lots of them—and thebrew literature was full of articles debat-ing the relative merits of dried versus liq-uid yeasts. Rightly or wrongly, most of ushave since developed a bias in favor offermenting agents of the liquid kind.Dried stuff, we have all internalized, isjust for amateurs and sissies. Real brewersuse liquid pitches.

Many of us from the “old school” of dried-yeast skeptics, however, have also learnedthat liquid yeasts come with their ownproblems. We are well aware of the difficul-ties involved in maintaining one or morehealthy yeast strains in a brewery, and wehave experienced the existential crampsthat come with not knowing if the pitch athand—after three weeks in a keg in thecooler—is still viable. It is against this atti-tudinal backdrop that we decided to takethe “Pepsi challenge,” that is, to check forourselves if the new yeast kid on the blockcan make beer of commercial quality.

TEST SET-UP

S.I. Lesaffre happens to be the largest yeastmanufacturer in the world, producingyeast for applications from baking to wine-making to industrial alcohol production.In the United States, the company sellseight commercial varieties of dried brew-ers yeast—three bottom- and five top-fer-menting strains—through its distributorCrosby & Baker of Westport, Mass.

Fermentis’ commercial yeasts are shippedsterile in vacuum-packed 500-gram bags,each containing enough yeast for 4 to 6hectoliters (roughly 3.5 to 5.1 barrels) ofwort, depending on the tank temperatureat pitching time. Each bag costs about$28, much less than an equivalentamount of liquid yeast. S.I. Lesaffre claimsthat its dried products are easier to usethan liquid yeasts yet produce beers ofequal quality. It is these claims of conven-ience and great taste at a lower price thatwe set out to test.

3 2 ZYMURGY July/August 2006 www.beertown.org

Artwor

k©

2006

www.clip

art.c

oman

dIm

ageClub

We conducted the test at the PortsmouthBrewery, a brewpub in downtownPortsmouth, N.H. on February 17, 2005.We diverted portions of a mildly hopped(17 IBU) cream ale production wort of 12.5°P (OG 1.050) gravity at a temperature of58° F (14° C) into four 5-gallon carboys,each pitched with a different yeast, driedand not. To compare the results from driedand liquid yeasts, we selected the sameyeast strain, the so-called “Chico” yeast,from three different sources. From WhiteLabs, we used a homebrew-size vial ofWLP051 California Ale V liquid yeast.From Wyeast we used a smack-pack of1056 American Ale liquid yeast. From S.I.Lesaffre we used 12.5 grams (0.44 ounces)of Fermentis US56 Dry Ale yeast.

We started the smack pack the daybefore brew day. We hydrated theFermentis yeast in accordance with theproducer’s instructions for a little morethan an hour in 10 times the yeast’sweight in sterile wort (in approximately0.25 liter or 8.5 fluid ounces of wort).Because the three brews had identicalspecifications (except for the yeastsources) and were being treated identi-cally, we reasoned that any noticeablequantitative or sensory differencesshould be directly attributable to differ-ent Chico strains.

The theoretical assumption, of course, isthat there ought to be no sensory differ-ences in the finished beer, because theseyeasts are supposed to be geneticallyidentical.

Merely for contrast, we also pitched a 5-gallon carboy of cream ale with a hydrat-ed half-and-half mixture of FermentisUS56 dried ale yeast and Fermentis W-34/70 dried lager yeast. The rest of thewort went into a stainless steel conical fer-menter, where it was pitched with thePortsmouth Brewery’s house ale yeast,which happens to be White Labs WLP051California Ale V liquid yeast. We hopedthat the hybrid test brew and the regularproduction batch would give us an addi-tional qualitative and quantitative frameof reference to help us interpret any vari-ations we might discover among the threeChico brews. All batches were pitchedsimultaneously.

We selected cream ale for the test becauseits grain bill makes for a fairly neutral brewand the low bittering rate would ensurethat the hops did not overpower the rest ofthe flavors. This, we reasoned, would giveus a better chance of detecting sensoryvariations (if any) that we expected wouldbe present in the beer as a result of the dif-ferent yeast strains. For the record, wecomposed the cream ale’s 387.5-poundgrain bill from 310 pounds (82 percent) oftwo-row lager malt from Cargill (2.8-3.5°L); 10 pounds (>2 percent) of caramelmalt from Cargill (20 *L); 10 pounds (> 2percent) of Weyermann Carafoam® malt(1.5-2.4 °L); 7.5 pounds (> 2 percent) ofWeyermann acidulated malt (1.7-2.8 °L);and 50 pounds (13 percent) flaked ricefrom Briess (0.7 °L). We mashed the gristby single infusion at 149° F (65° C). Atmash-out, we raised the temperaturethrough sparging to 165° F (74° C). Thisgrain bill yielded a net kettle volume of224.75 gallons (7.25 barrels) at a colorvalue of 4.4 SRM. We employed a 75-minute boil with additions of Centennialhop pellets for bittering five minutes intothe boil, of Ahtanum leaf hops in a steep-ing bag for flavor 30 minutes into the boil,and Saaz pellets for aroma in the whirlpool.

TEST OBSERVATIONS

The White Labs strain had by far thefastest start, showing the first signs of fer-mentation within 12 hours of pitching.The Wyeast smack-pack took twice as

long, 24 hours. The Fermentis Chicobatch was in the middle of the two liquidyeasts, starting to crank within 18 hoursof pitching time, while the combination ofdried ale and lager yeasts kicked in after16 hours. By day 12, however, all batcheswere at roughly the same attenuationlevel, 4.1 °P ± 0.2 °P.

www.beertown.org July/August 2006 ZYMURGY 3 3

least amount of scum, while the WhiteLabs batch showed the most.

We left the test brews undisturbed in a cooldark corner of the fermentation cellar for atotal of three weeks and then racked theminto Cornelius kegs. The gravity of theFermentis dry Chico brew had dropped atthat point to 2 °P; the White Labs liquidChico brew to 2.2 °P; the combination ale-lager brew to 2.2 °P; and the Wyeast liquidChico brew to 3.2 °P. All test brews and theproduction batch showed good floccula-tion and finished equally clear. None of thebeers were filtered. Table 1 shows the grav-ity drops of the test brews.

SENSORY EVALUATION AT END OF

FERMENTATION

For us, the big question, after three weeksin the fermenters and before transfer intothe conditioning Cornelius kegs, was ifthere was a noticeable difference in tastebetween the beers made with dried com-pared to liquid yeast.

NOSE

We found no severe off-aromas in any ofthe brews’ bouquet, just slight variations.The nose of the White Labs batch wasslightly fruitier than the others and showeda slight touch of DMS, but no signs of sul-fur. The Wyeast batch had perceptibleDMS components and a slight vegetal note,probably the effect of the longer lag timebetween pitching and the start of fermenta-tion. (This would dissipate after condition-ing.) The Fermentis ale batch had a fewsulfur notes, as did the batch with the driedFermentis ale and lager yeasts combined,but the latter less so. The Fermentis alebatch had fairly perceptible DMS compo-nents, while the combination ale-lagerbatch was free of DMS. The combinationbatch also had a slightly “lagerish,” greenapple, acetaldehyde bouquet.

UP-FRONT TASTE IMPRESSION

Though all batches had the same IBUlevel, there was a noticeable difference inperceived bitterness. The ale-lager combi-nation had the least bitter profile, whilethe White Labs batch revealed some bitter-ness. The Wyeast liquid and Fermentis

34 ZYMURGY July/August 2006 www.beertown.org

Yeast WLP 051-V (liquid) Fermentis US-56 + W-34/70 (dried) Fermentis US-56 (dried) Wyeast 1056 (liquid)

Lag time from pitchingto start of fermentation 12 hrs 16 hrs 18 hrs 24 hrs

Gravity at day 4 8.0 °P 7.3 °P 7.9 °P 8.4 °PDay 6 5.0 °P 6.6 °P 5.5 °P 7.5 °PDay 8 4.9 °P 5.7 °P 4.7 °P 5.9 °PDay 10 4.6 °P 5.0 °P 4.0 °P 4.9 °PDay 12 4.0 °P 4.3 °P 3.9 °P 4.2 °PDay 14 3.4 °P 4.0 °P 3.6 °P 4.0 °PDay 16 3.1 °P 3.1 °P 3.1 °P 3.9 °PDay 18 2.9 °P 2.8 °P 2.9 °P 3.7 °PDay 19 2.4 °P 2.5 °P 2.4 °P 3.5 °PDay 21 (FG) 2.2 °P 2.2 °P 2.0 °P 3.2 °PABV 5.3% 5.3% 5.4% 4.8%

TABLE 1 : GRAVITY READINGS OF TEST BREWS

BREW DATE: FEB. 17, 2005; OG: 12.5° P (1.050); SRM: 4.4; IBU: 17

The relatively slow start of the Wyeastbatch was clearly an abnormality and wasprobably attributable to some improperhandling of the smack pack before we gotit. We have both worked with Wyeastyeasts before and never had such difficul-ties with lag times. The Wyeast batchremained “behind” the others throughoutthe trial, but after about four weeks, thefinished beer tasted smooth and com-pletely free of defects.

After primary fermentation, there werenoticeable differences in the residue collarin the carboy headspace. The two batchesfermented with dried yeasts showed the

dried pure ale batches had the most hopcharacter. The parallel production batchgave us a slight sulfur sensation. None ofthe batches showed any buttery flavors.There were no noticeable levels of diacetyl.

MIDDLE FLAVOR AND MOUTHFEEL

True to the cream ale style, all four testbatches and the production batchturned out light-bodied, with the WhiteLabs batch having the most roundedmouthfeel.

FINISH

The three pure ale batches all finishedequally dry, while the batch pitched withboth ale and lager yeast finished very dry.The Wyeast batch had slightly more hopsthan malt in the finish, while the WhiteLabs test batch finished clean with a nicebalance between maltiness and hoparoma. Interestingly, there was a touch ofpleasant sweetness in both the White Labstest and production batches at this stage,but none in the other batches. Both theproduction and Fermentis ale batches hada slightly viniferous component in the fin-ish, which left a refreshing aftertaste.Finally, the most noticeable component inthe finish of the ale-lager combinationbatch was a touch of hop aroma.

SENSORY EVALUATION OF

CONDITIONED BREWS

After two weeks of conditioning the testbatches in the cooler at about 38° F (3° C)in Cornelius kegs and the productionbatch in the unitank at 35° F (2° C) ± 2°F (1° C) at approximately 10 psi, thebatches all seemed to converge in nose,taste and finish. There were still a few sen-sory differences, but they were not asgreat as at the end of fermentation.

The Wyeast batch especially had madegreat strides. It was now smooth, roundedand pleasantly balanced. It had a slightlyfruity nose, no DMS, no sulfur notes andno off-flavors. It finished with a delicioustouch of sweetness and a mild lingeringhoppiness. It had slightly more body thanthe other batches. In our judgment, thisbrew was late out of the starting gate, butmade up ground in the stretch and edgedthe other batches by a short nose in over-all balance. It became a particularly good,hearty cream ale.

great and refreshing summer quaffingcream ale.

CONCLUSION

We set out to prove or disprove a preju-dice. We wanted to know if dried yeastcan stand up to liquid yeast under con-trolled conditions. We found that none ofthe yeasts we tested singularly or in com-bination were universally better or worsethan any others. They all had differentstrengths and weakness. Obviously, ourtest, though fair in approach, is only asingle experiment and a single data point.

Likewise, our palates, though experi-enced, are subjective. Our sensory evalu-ations reflect not only what was in thebrews but also what we like and dislike.Our somewhat mixed conclusions, there-fore, do not represent law-like generaliza-tions about all characteristics of theseyeasts, nor should they be the basis forpassing judgment about the relative mer-its of different yeast brands. That was notour purpose. Rather, we wanted to focus

www.beertown.org July/August 2006 ZYMURGY 3 5

BECAUSE THE THREE BREWS HAD IDENTICAL SPECIFICATIONS

(EXCEPT FOR THE YEAST SOURCES) AND WERE BEING TREATED

IDENTICALLY, WE REASONED THAT ANY NOTICEABLE QUANTITATIVE

OR SENSORY DIFFERENCES SHOULD BE DIRECTLY ATTRIBUTABLE TO

DIFFERENT CHICO STRAINS.

The White Labs test and productionbatches showed just a touch of sulfur upfront, but both finished very clean with anice balance between gentle tartness andmellow sweetness, which made for a verydrinkable cream ale.

The Fermentis pure ale batch showedslightly more bittering hop notes than didthe other batches, both up front and inthe finish. It also revealed a slight hint ofDMS and sulfur, but so faint that thesecompounds can be detected only byexpert palates. There was some residualsweetness in the finish to make the brewreverberate pleasantly on the palate. Thisended up being perhaps the most com-plex and ale-like of the cream ales.

The combination ale-lager batch contin-ued to show true “lagerish” characteris-tics. The flocculation was perfect, therewas no nose and the finish was excep-tionally clean and dry, with the Saazaroma more clearly identifiable in thefinish than in the other batches. It was a

narrowly on the flavor difference, if any,produced by dried versus liquid yeasts.

We found the White Labs test batch andthe two Fermentis batches (ignoringslight notes of sulfur) almost ready to beserved to brewpub customers after fer-mentation, while the Wyeast batch andthe original production batch neededmore mellowing. After two weeks of con-ditioning, however, the Wyeast batch hadclearly caught up in flavor and mouthfeel.All sulfur notes had largely dissipated,too. Much to our surprise, we liked thehybrid combination test batch of cream

ale made with both dried ale and driedlager yeast the best, mostly because it hadthe cleanest and most appealing flavorfrom start to finish.

We also observed that, along several vari-ables, the batches with the two liquidyeasts showed greater differences amongthemselves than they did in comparisonwith the dried-yeast batches. For exam-ple, both the dried Fermentis ale yeastand the liquid Wyeast ale yeasts (in spiteof the latter’s relatively high final gravity)showed virtually no residual sweetness inthe finish, while the liquid White Labs

yeast did in both the test and productionbatches. Likewise, there was no pattern inthe lag times of dried and liquid yeasts.

In the end, all our batches resulted inexcellent, commercially viable brews. Thechoice between liquid versus dried yeasts,therefore, should be based on such crite-ria as brewing objectives, brew houseworkflows and brewery economics.

Also, because packages of dried yeast havea long shelf life if kept unopened in therefrigerator, they can function as relativelyinexpensive backups for situations whenthe brew schedule becomes unhinged andthe yeast harvested from a previous batchhas reached a questionable age. The onedrawback we see with the current offeringof dried yeasts is the relatively limitedrange of yeast types compared to thehuge array of specialty yeasts available inliquid form.

POST-TEST POSTSCRIPT

For us, perhaps the most significant findingfrom the test was the slightly stronger sulfurcomponent, especially before conditioning,in the batches fermented partially or entire-ly with dried yeast—a difference thatdiminished, however, with extended tanktimes. We decided to verify this conclusionin two production follow-up batches. Thefirst was an all-dried Fermentis yeast creamale. After fermentation it did indeed showsulfur notes, which, however, amelioratedwithin two weeks of conditioning. A secondhybrid batch of cream ale, fermented with acombination of White Labs liquid ale andFermentis dried lager yeast, also had a hintof sulfur initially, but tasted clean, “lagerish”and truly delicious after a total of five weeksfrom brew day.

Tod Mott is the head brewer of thePortsmouth Brewery in New Hampshire.He has won several medals at the GreatAmerican Beer Festival, including a 1997bronze for wheat bock, a 1999 gold forScotch ale and a 2001 gold for amberale. Horst Dornbusch is a frequent con-tributor to several brew magazines. Heowns Cerevisia Communications, a PRagency for the international beverageindustry. At the 2000 Great AmericanBeer Festival, he won a bronze medalfor altbier.

3 6 ZYMURGY July/August 2006 www.beertown.org