Date post: | 20-May-2018 |
Category: |
Documents |
Upload: | nguyendung |
View: | 225 times |
Download: | 1 times |
Hopsteiner PanelTesting Results
Comparison of Simcoe with(9 panelists)
Difference detected
No difference detected
Significant difference(at 0.01 level)
60% Eureka40% Calypso
8 panelists 1 panelist Yes
50% Eureka40% Apollo10 % Cascade
4 panelists 5 panelists No
Method: Triangle Test (MEBAK – Sensory Analysis, 2014, Method 3.1.3)Results: Recommended alternative to Simcoe = 50% Eureka + 40% Apollo + 10% Cascade
Comparison of Citra with(10 panelists)
Difference detected
No difference detected
Significant difference(at 0.01 level)
40% Eureka35% Calypso25% Apollo
9 panelists 1 panelist Yes
70% Calypso30% Bravo
7 panelists 3 panelists No
Method: Triangle Test (MEBAK – Sensory Analysis, 2014, Method 3.1.3)Results: Recommended alternative to Citra = 70% Calypso + 30% Bravo
Citra alternative - dry hopped
Simcoe alternative - dry hopped
Comparison of Amarillo Difference detected
No difference detected
Significant difference
80% Cascade20% Lemondrop
No
Method: Panel Tetrad Test (Sensory Analysis, 2016)Results: Recommended alternative to Amarillo = 80% Cascade + 20% Lemondrop
Amarillo alternative - late kettle addition
The application of science
to hop blending
Dr Chris Smart
Head of Brewing Services
Campden BRI
Overview
• Hop blending
• Our approach
• Hop tea data and insights
• Beer data and insights
• Hop blending and results – ‘show me the way to Amarillo’
• Conclusions
Hop blending
• More art than science
• Based on experience – hop merchant
and brewer
• Difficult to predict
• Several recommended substitute hops
Our approach
• Identify a range of aroma hops
– New and traditional varieties
– From several geographical regions
• Assess hops - sensory profiling and analytically
• Pilot brews – single hop brews, late addition
• Analyse data and use it to see if it can predict the
effect on blending
• Brew blends to match a specific aroma hop variety
• Assess blended hop beer using sensory panel
Hop choice • Identify a range of hops to investigate: 14 selected in total
• New and established varieties
• From various regions in the world (UK, Europe, US, Aus)
Hop variety
Amarillo hop pellets Hersbrucker Spat hop pellets
Bramling Cross hop pellets Lemondrop hop pellets
Citra hop pellets Mandarina Bavaria hop pellets
East Kent Golding hop pellets Sovereign 2014 hop pellets
Fuggle hop pellets US Cascade hop pellets
Galaxy hop pellets Saaz hop pellets
Hallertauer Mittelfruh hop pellets Simcoe 2013 Harvest US hop pellets
Hop tea sensory data
• Hop teas prepared using 5g hop pellets
in 2L boiling water left to cool to room
temperature
• Expert panel agreed the sensory
attributes to be used
• 22 attributes identified and used to
assess each hop, scale 0-9
All 14 hop teas
UK grown hops
European grown hops
US grown hops
Geographical differences
Geographical differences
• Why are the hops in each region similar to each other but
different from other regions?
– Genetically similar through breeding programmes or natural populations?
– Consumer/brewer preferences pushing selection/breeding in one direction
in a region?
– Growing conditions/climate?
– What are the sensory characteristics of a single hop variety grown in
different regions?
Fuggle vs Saaz
Analytical data
• Hops analysed by GCMS QToF (x3)
• Over 100 volatile or semi-volatile
compounds identified
• The majority have known sensory
properties
• Wide variation found for many compounds,
as expected, in each hop variety
GCMS QToF: Simcoe hop pellets
Examples
1. Amarillo, 2. Bramling Cross, 3. Citra, 4. East Kent Golding, 5. Fuggle, 6. Galaxy, 7. Hallertauer
Mittelfruh, 8. Hersbrucker Spat, 9. Lemondrop, 10. Mandarina Bavaria, 11. Sovereign,
12. Cascade (US), 13. Saaz, 14. Simcoe
Beer sensory data
Beer sensory - taste
Beer sensory - aroma
Hop tea aroma vs beer aroma
Hop tea aroma Beer aroma
• Similar in top right quadrant
• Green beans, cooked veg, green leaves, herbal medicinal <- disappeared! Why?
• Hop teas are not a good predictor of hop aroma in beer (at least not under these
brewing conditions)
Beer analytical data
• From 120 hop compounds identified in hop teas only 9 are
found in the beer namely:
– 2-methyl-1-butanol (roasted wine onion fruity)
– isobutyl butyrate (sweet, fruity, candy, berry, cherry, tutti frutti, over ripe
and bubble gum-like)
– beta-pinene (woody, piney, turpentine-like, minty, eucalyptus,
camphoraceous, spicy peppery and nutmeg)
– alpha-phellandrene (citrus, terpenic, slightly green, black pepper)
– linalool (citrus, orange, floral, terpy, waxy and rose)
– geraniol (floral, sweet, rosey, fruity and citrus)
– humulene (woody)
– beta-cadinene (green woody)
– cis-calamenene (herb)
Beer analytical data
• Why so few hop cpds detected in the beer? And what happens
to the sensory aroma notes?
– Below limits of detection (analytically)
– Masked by other compounds (analytically and sensory)
– Chemically transformed
– Biochemically transformed
– Too volatile
• Some cpds are lost due to heat/CO2 stripping etc
• Some cpds biotransformed e.g. reduction (NADH recycling?) or
esterification
– 2-undecanone (hop cpd) -> 2-undecanol (beers)
– Decyl acetate (in beers) <- ester possibly formed from decanol/acetic acid
• Question: late/dry hopping – how much difference does this
make?
Hop blending
Matching Amarillo – beer data
Matching Amarillo – beer data
Matching Amarillo – beer data
Panel Tetrad Test result: no significant difference
Conclusions • 22 fixed sensory attributes can be used to differentiate all 14
hop varieties tested
• For most varieties hop teas are a poor predictor of the sensory
attributes in the final beer product
• Different hop varieties grown in the same country have very
many sensory attributes in common as hop teas
• The aroma characteristics of some specific hop varieties, even
those grown in different countries, are almost identical (e.g. UK
grown Fuggle and European grown Saaz)
• Certain aroma notes in all of the hop varieties tested are
reduced, and often disappear, in the final beer especially
green/herbal notes
• Blending can be assisted by sensory data but it is best done in
beers not hop teas
• Amarillo can be matched in finished beer with late hop addition
using Cascade and Lemondrop at a ratio of 5:1
Acknowlegdements
• Ed Wray, Project brewer
• Gill Fisher, Senior sensory scientist
• Eung Lee, Project maltster
• Campden BRI members
Questions
Dry hopping to match a target
hop profile
Progress Report
30th January 2017
Outline
• Objectives
• Brewing
• Analytical
• Sensory
• Hop Blend modelling
• Next steps
• Sensory results from Hop blends
Objectives
The key objective is to assess and blend a range of hops (dry hopped in
fresh beer) to match the sensory character of a target hop. The
deliverables will be descriptive profiles of 10 dry hopped beers which
will be used to produce beers with hop blends to match the target hop.
Objective 1. Propose hop blends to match Citra without using Simcoe
Objective 2. Propose hop blends to match Simcoe without using Citra
• Recipe
25 kg 100 % pale malt
5 kg Maltose syrup to achieve gravity required (for 200 litres of wort)
•Mashing
64 ⁰C for 60 minutes
•Kettle
Boil 60 minutes
Alpha extract to achieve approx 20 BUs (65 g)
•Fermentation
Nottingham Ale yeast (dried)
2 x 100 litre FVs
18 ⁰C for 6 days
13 ⁰C for 3 days (diacetyl rest)
3 ⁰C 3 days (removal of yeast)
•Dry hopping
10 x 16 litres in Cornelius kegs for the 10 hop varieties
5 g hops per litre of beer
Condition at cellar temperature (approx 13 ⁰C) for 7 days
•Bottling
No filtration – bottle conditioned
Priming sugar addition
Crown
Condition at 18 ⁰C for 7 days prior to sensory evaluation
Analytical results
ze
ro
Ap
oll
o
Bra
vo
Ca
lyp
so
Ca
sc
ad
e
Cit
ra
De
na
li
Eu
rek
a!
Ha
lle
rta
u
Mit
ellfrϋ
h
Le
mo
nd
rop
Sim
co
e
Ethanol (% v/v) 4.58 4.75 4.70 4.67 4.56 4.70 4.59 4.62 4.58 4.61 4.78
Bitterness (BU) 21.3 31.2 29.2 30.9 22.5 24.6 24.4 23.8 23.3 22.9 27.9
CO2 (g/L) 5.06 5.78 6.05 6.13 5.93 6.49 5.64 6.05 5.45 5.86 7.02
NIBEM10 (s) 84 63 89 75 87 65 84 58 80 61 50
NIBEM20 (s) 174 141 176 151 185 143 176 125 171 142 106
NIBEM30 (s) 265 222 259 227 271 227 264 207 269 237 167
Flavour Volatiles
Th
res
ho
ld
ze
ro
Ap
oll
o
Bra
vo
Ca
lyp
so
Ca
sc
ad
e
Cit
ra
De
na
li
Eu
rek
a!
Hall
ert
au
Mit
ell
frϋ
h
Le
mo
nd
rop
Sim
co
e
Acetaldehyde
(mg/L)25 3.2 4.5 3.3 3.2 4.2 3.6 3.8 4.0 4.2 3.7 3.9
Ethyl Acetate
(mg/L)33 9.6 11.6 9.7 10.3 10.8 9.9 11.1 11.6 10.8 10.8 10.8
Iso-Butyl Acetate
(mg/L)1.6 <0.06 0.06 <0.06 0.06 <0.06 <0.06 <0.06 <0.06 <0.06 <0.06 0.07
n- Propanol
(mg/L)800 14.6 22.0 27.4 26.0 18.7 30.0 18.6 27.7 25.4 27.4 28.3
Iso-Butanol
(mg/L)200 31.3 35.6 39.9 38.4 32.7 42.1 32.3 41.2 41.5 41.1 40.3
Iso-amyl acetate
(mg/L)1.6 0.42 0.64 0.58 0.76 0.62 0.63 0.59 0.95 0.081 0.088 0.66
Iso- Amyl
alcohol (mg/L)70 85.9 83.9 85.7 83.9 83.6 86.0 83.6 85.7 86.8 85.1 87.8
Ethyl Hexanoate
(mg/L)0.2 0.04 0.07 0.04 0.06 0.06 0.05 0.04 0.08 0.06 0.06 0.04
Dimethyl
Sulphide (μg/L)50 21.3 27.6 34.1 38.2 23.3 34.6 14.9 24.3 21.0 <12 35.8
Diacetyl (mg/L) 0.15 0.03 0.05 0.03 0.03 0.05 0.04 0.05 0.05 0.06 0.07 0.05
2,3 –
Pentanedione
(mg/L)0.90 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.02 0.01
Sensory Analysis
• Check All That Applies using a defined hop
glossary based on literature and past experience
• Generation of contingency tables, PCA
• Analysis of attribute desirability and penalty
using similarity scores to target hop profile
• Hop profile comparison
• Modelling of hop blending using PLS regression
CATA contingency tableF
lora
l
Sp
icy
Tro
pic
al F
ruit
Gra
pefr
uit
Lem
on
Pin
e
Gra
ss
y
He
rbal
Re
sin
ou
s
Su
mm
er
Fru
its
Wo
od
y
Sw
eet
So
ur
As
trin
ge
nt
Lin
ger
Apollo 7 7 4 10 7 13 8 12 13 3 10 7 4 10 9
Bravo 16 7 10 9 14 4 13 4 5 10 5 11 9 5 5
Calypso 8 9 6 10 10 7 11 13 7 7 6 5 9 13 8
Cascade 9 5 7 7 10 3 6 5 6 7 2 9 9 9 6
Citra 11 5 9 16 10 10 12 14 13 3 6 7 9 11 10
Denali 12 6 7 10 9 5 10 6 6 7 6 8 8 10 6
Eureka 7 3 9 16 14 6 10 13 13 7 4 6 12 11 10
Hallertau 9 6 3 3 8 4 10 11 6 10 5 9 8 10 8
Lemondrop 10 5 2 8 10 2 12 2 6 9 5 8 9 11 6
Simcoe 11 5 7 16 10 7 11 15 11 6 5 7 9 10 8
Note: only attributes with frequency ≥ 15% are included in the analysis
Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbal
Resinous
Summer Fruits
Woody
Sweet
Sour
Astringent
Linger
Apollo
Bravo
Calypso
CascadeCitra
Denali
Eureka
Hallertau Mitt.
Lemondrop
Simcoe
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
-0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 -1E-15 0.1 0.2 0.3 0.4 0.5 0.6 0.7
F2 (
19
.43
%)
F1 (51.50 %)
Symmetric plot(axes F1 and F2: 70.94 %)
Attributes Products
Analysis of attributes based on ideal
profile (Citra)
Similarity score based on Euclidean distance using 5 principal components
(90% difference explained)
Hop Similarity
Apollo 5.1
Bravo 2.8
Calypso 5.6
Cascade 3.7
Citra 10.0
Denali 5.5
Eureka 6.5
Hallertau Mitt. 4.3
Lemondrop 4.0
Must have Nice to have Does not influence Does not harm Must not have
Grapefruit Floral
Herbal Spicy
Resinous Tropical Fruit
Lemon
Pine
Grassy
Woody
Sweet
Sour
Astringent
Linger
Summer Fruits
0.0
2.0
4.0
6.0Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Apollo Citra
Similarity: 5.1
0.0
2.0
4.0
6.0Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Bravo Citra
Similarity: 2.8
0.0
2.0
4.0
6.0Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Calypso Citra
Similarity: 5.6
0.0
2.0
4.0
6.0Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Cascade Citra
Similarity: 3.7
0.0
2.0
4.0
6.0Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Denali Citra
Similarity: 5.5
0.0
2.0
4.0
6.0Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Eureka Citra
Similarity: 6.5
0.0
2.0
4.0
6.0Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Hallertau Mitt. Citra
Similarity: 4.3
0.0
2.0
4.0
6.0Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Lemondrop Citra
Similarity: 4.0
Blend modelling Citra
0123456
Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
Grassy
HerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Citra Pred(Citra)
60% Eureka + 40% Calypso
0
1
2
3
4
5
6Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
Grassy
HerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Citra Pred(Citra)
60% Eureka + 40% Apollo
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
Grassy
HerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Citra Pred(Citra)
40% Eureka + 30% Calypso + 30% Apollo
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
Grassy
HerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Citra Pred(Citra)
50% Eureka + 30% Calypso + 20% Denali
Blend Modelling Citra (II)
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Citra Pred(Citra)
70% Calypso + 30% Bravo
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Citra Pred(Citra)
70% Apollo + 30% Bravo
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Citra Pred(Citra)
40% Apollo + 40% Calypso + 20% Bravo
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Citra Pred(Citra)
50% Calypso + 30% Denali +20% Bravo
Analysis of attributes based on ideal
profile (Simcoe)
Similarity score based on Euclidean distance using 5 principal components
(90% difference explained)
Hop Similarity
Apollo 4.9
Bravo 4
Calypso 6.4
Cascade 5.2
Denali 7
Eureka 7
Hallertau Mitt. 5.8
Lemondrop 5.6
Simcoe 10
Must have Nice to have Does not influence Does not harm Must not have
Herbal Grapefruit Floral
Spicy
Tropical Fruit
Lemon
Pine
Grassy
Resinous
Summer Fruits
Woody
Sweet
Sour
Astringent
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Apollo Simcoe
Similarity: 4.9
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Bravo Simcoe
Similarity: 4.0
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Calypso Simcoe
Similarity: 6.4
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer…
Woody
Sweet
Sour
Astringent_
Linger
Cascade Simcoe
Similarity: 5.2
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer…
Woody
Sweet
Sour
Astringent_
Linger
Denali Simcoe
Similarity: 7.0
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Eureka Simcoe
Similarity: 7.0
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Hallertau Mitt. Simcoe
Similarity: 5.8
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer…
Woody
Sweet
Sour
Astringent_
Linger
Lemondrop Simcoe
Similarity: 5.6
Blend modelling Simcoe
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Simcoe Pred(Simcoe)
60% Eureka + 40 % Calypso
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
Grassy
HerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Simcoe Pred(Simcoe)
45% Eureka + 35% Calypso + 20% Apollo
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
Grassy
HerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Simcoe Pred(Simcoe)
70% Eureka + 30% Apollo
0.01.02.03.04.05.06.0
Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Simcoe Pred(Simcoe)
50% Eureka + 40% Calypso + 10% Hallertau
Blend Modelling Simcoe (II)
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Simcoe Pred(Simcoe)
90% Calypso + 10% Bravo
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Simcoe Pred(Simcoe)
80% Apollo + 20% Bravo
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Simcoe Pred(Simcoe)
60% Calypso + 30% Apollo + 10% Bravo
0123456Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
GrassyHerbalResinous
Summer Fruits
Woody
Sweet
Sour
Astringent_
Linger
Simcoe Pred(Simcoe)
70% Calypso + 20% Hallertau + 10% Bravo
Conclusions from Blend Model
• Simcoe and Citra have the closest sensory profiles
• Eureka and Calypso are the nearest sensory match to both Simcoe and Citra and they can be used in blends to match both target hops
• Apollo and Denali could be used in blends to match Citra
• Apollo and Hallertau Mitellfrϋh could be used in blends to match Simcoe
Next Steps
• Validate blend models
• Brew 2 hL of beer, split for dry hopping,
bottle and pasteurise
• Descriptive analysis of up to 10 hop
blends and validation by triangular testing
Hop Blends
1. 40% Calypso 30% Apollo 30% Denali
2. 40% Eureka 35% Calypso 25% Apollo
3. 50% Eureka 30% Calypso 20% Denali
4. 50% Eureka 40% Apollo 10% Cascade
5. 60% Eureka 40% Calypso
6. 65% Eureka 35% Apollo
7. 70% Calypso 30% Bravo
8. Citra
9. Simcoe
PCA Blends vs Target Hops
40% Calypso 30% Apollo 30% Denali
40% Eureka 35% Calypso 25% Apollo
50% Eureka 30% Calypso 20% Denali
50% Eureka 40% Apollo 10% Cascade
60% Eureka 40% Calypso
65% Eureka 35% Apollo
70% Calypso 30% BravoCitra
Simcoe
Floral
Spicy
Tropical Fruit
Grapefruit
Lemon
Pine
Grassy
Herbal
Resinous
Summer Fruits
Woody
SweetSour
Astringent
Linger
-8
-6
-4
-2
0
2
4
6
8
10
-10 -8 -6 -4 -2 0 2 4 6 8 10 12
F2 (
15
.91
%)
F1 (48.18 %)
Biplot (axes F1 and F2: 64.09 %)
Agglomerative Hierarchical Clustering Citra
50
% E
ure
ka 3
0%
Cal
ypso
20
% D
enal
i
65
% E
ure
ka 3
5%
Ap
ollo
50
% E
ure
ka 4
0%
Ap
ollo
10
% C
asca
de
60
% E
ure
ka 4
0%
Cal
ypso
70
% C
alyp
so 3
0%
Bra
vo
40
% E
ure
ka 3
5%
Cal
ypso
25
% A
po
llo
Cit
ra
0.01
0.11
0.21
0.31
0.41
0.51
0.61
0.71
0.81
0.91
Sim
ilari
ty
Dendrogram Citra
Agglomerative Hierarchical Clustering Simcoe
40
% E
ure
ka 3
5%
Cal
ypso
25
% A
po
llo
Sim
coe
40
% C
alyp
so 3
0%
Ap
ollo
30
% D
enal
i
65
% E
ure
ka 3
5%
Ap
ollo
50
% E
ure
ka 4
0%
Ap
ollo
10
% C
asca
de
60
% E
ure
ka 4
0%
Cal
ypso
-0.27
-0.07
0.13
0.33
0.53
0.73
0.93
Sim
ila
rity
Dendrogram Simcoe
Selected Blends for Validation
• Citra
– 40% Eureka 35% Calypso 25% Apollo
– 70% Calypso 30% Bravo
• Simcoe
– 60% Eureka 40% Calypso
– 50% Eureka 40% Apollo 10% Cascade
– 65% Eureka 35% Apollo
– 40% Calypso 30% Apollo 30% Denali
Sensory Profiles Citra Blends
Sensory Profiles Simcoe Blends
Next Steps
• Validate selected blends by Tetrad testing with
12 trained panellists
– Perceived difference at 95% confidence?
– Estimate degree of difference using
Thurstonian model and calculating d’
Project Management Team
12
Management Support:
Dr. Chris Smart
Head of Department – Brewing Services
Brewing:
Ed Wray
Project Brewer Maltster
Sensory:
Dr. Javier Gomez-Lopez
Sensory and Product Innovation Manager
Analytical :
Ian Slaiding
Analytical Laboratory Manager-Beer and Beverage Analysis