Return to TheElectricBrewery.com
  [ Shop ]   [ Building ]   [ Using ]   [ Recipes ]   [ Testimonials ]   [ Gallery ]   [ FAQ ]   [ About Us ]   [ Contact Us ]   [ Newsletter ]

Log inLog in   RegisterRegister   User Control PanelUser Control Panel   Private MessagesPrivate Messages   MembershipClub Memberships   SearchSearch   MemberlistMemberlist   Photo AlbumsPhoto Albums   Forum FAQForum FAQ


Lagering and FuP Experiment

 
Post new topic   Reply to topic   Printer-friendly view    TheElectricBrewery.com Forum Index -> Brewing Techniques
View previous topic :: View next topic  
Author Message
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Sat Sep 01, 2018 11:04 pm    Post subject: Lagering and FuP Experiment Reply with quote


        Register to remove this ad. It's free!
For those interested in Fermentation Under Pressure (FuP), my local brew club has sponsored an experiment being conducted by myself and a fellow club member. The experiment will test to see (at least among our members) if there is a sensory, if not statistical, difference between a split batch of Czech Pilsner Lager - one using a traditional lagering method and one using FuP.

A triangle testing method will be used to taste test the samples – 3 samples (2 of the lager and 1 of the FuP lager). The participants will be asked to identify the FuP sample. Right and wrong answers will be tabulated for statistical analysis. For example, if most of the participants can't identify the different beer, the stats will determine if the results are meaningful or statistically significant (i.e. there's no sensory difference between the two beers). This is a same method used by food companies when they reformulate their recipes.

Also hoping to get a local university to run gas chromatograph tests on the beers to determine the difference in ester concentrations (should be lower in FuP beer). Ester suppression is one of the reasons for traditional lagering.

Attached is the set up.

Results in November...



DRAFT-FuP-5.pptx
 Description:

Download
 Filename:  DRAFT-FuP-5.pptx
 Filesize:  491.63 KB
 Downloaded:  656 Time(s)

Back to top
perogi




Joined: 12 Feb 2012
Posts: 850
Location: NH

Drinking: Perogi Pale, NEIPA, Nutter's Crossing Nut Brown Ale, Edmund Fitzgerald Porter Clone

Working on: Max's Maibock


PostLink    Posted: Tue Sep 04, 2018 1:37 am    Post subject: Reply with quote

Interested in hearing your results
Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Sun Sep 09, 2018 3:15 pm    Post subject: Reply with quote

Update: In-progress fermentation under Pressure

Attached is a chart from the past two days. Fermentation activity started in less than 12 hours from when I pitched the yeast (Wyeast Bohemian Lager) on Friday. Looks like peak activity is already complete. Pressure reached as high as 20 psig and is currently at 15 psig (set pressure). Took a Brix reading this morning @ 5.9 Brix ~ 2.9% ABV still in progress.

Pressure readings were discrete so I used a moving average (BCS-460 logger doesn't accommodate pressure input - bummer). Pressure dropped from 20 to 15 psig overnight.



FuP_CHART-2.pdf
 Description:

Download
 Filename:  FuP_CHART-2.pdf
 Filesize:  114.73 KB
 Downloaded:  576 Time(s)


czech lager.pdf
 Description:

Download
 Filename:  czech lager.pdf
 Filesize:  71.52 KB
 Downloaded:  1051 Time(s)



Last edited by mjo2125 on Tue Sep 18, 2018 1:49 am; edited 1 time in total
Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Wed Sep 12, 2018 12:41 am    Post subject: Reply with quote

Next Step : Closed Loop Transfer

Terminal gravity was reached in less than a week. Pressure setting backed down to 15 - 17psig. Will let it sit for a day or two. Next step is a Closed Loop Transfer (CLT) to mini kegs in preparation for clarifying and conditioning. Attached is the closed loop schematic. No need for a spunding valve in CLT with pressure-rated equipment (and keg PRVs) The attached procedure should result in an O2-free transfer with no foaming.

Note: A corny keg (fermenter) spare outlet dip tube was shortened by about 3" and used to avoid pulling trub.



Closed Loop Transfer.pdf
 Description:

Download
 Filename:  Closed Loop Transfer.pdf
 Filesize:  121.21 KB
 Downloaded:  616 Time(s)

Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Fri Sep 14, 2018 1:27 pm    Post subject: Reply with quote

Next Step: Conditioning and Cold Crash

Closed Loop Transfer to mini keg went well - no foaming or O2 pick-up. Once flow was established, hands off until keg was filled. Will naturally carbonate with the spunding valve. attached are pics.

It has been observed on various blogs about the difference in carbonation with pressure fermentation - the carbonation bubbles tend to be smaller and finer with pressure fermentation and natural carbonation. My observation in taking samples is the same.



Condition.pdf
 Description:

Download
 Filename:  Condition.pdf
 Filesize:  91.23 KB
 Downloaded:  598 Time(s)


CLT.pdf
 Description:

Download
 Filename:  CLT.pdf
 Filesize:  125.09 KB
 Downloaded:  592 Time(s)

Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Wed Sep 19, 2018 1:50 am    Post subject: Reply with quote

Timeline Comparison: attached

After the closed loop transfer to the secondaries was completed, I observed a phenomenon – after the transfer, I depressured the corny keg. There was about 3” of beer and trub at the bottom.

About 10 seconds after I opened the lid, there was a violent eruption (much like a volcano) at the bottom of the corny – I can only guess that the yeast cell walls ruptured after the pressure was released – much like a scuba diver ascending too fast from the bottom… my takeaway – don’t plan on harvesting yeast if you ferment under pressure…



Days to FG-5.pdf
 Description:

Download
 Filename:  Days to FG-5.pdf
 Filesize:  125.03 KB
 Downloaded:  588 Time(s)

Back to top
perogi




Joined: 12 Feb 2012
Posts: 850
Location: NH

Drinking: Perogi Pale, NEIPA, Nutter's Crossing Nut Brown Ale, Edmund Fitzgerald Porter Clone

Working on: Max's Maibock


PostLink    Posted: Thu Sep 20, 2018 1:12 am    Post subject: Reply with quote

Looking forward to your tasting step.
Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Thu Sep 20, 2018 1:56 am    Post subject: Reply with quote

Next post will be related to the triangle taste test. Full experiment protocol in powerpoint available off-line due to file size for anyone interested.
Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Thu Oct 11, 2018 12:02 am    Post subject: Reply with quote

Next step: Package and perform triangle test.
Attached is the sample form for each judge.

The Statistics behind the Triangle Test:

This test is a popular method in sensory evaluation. It is used to determine whether there is a perceivable difference between two samples (in this case, a traditional lager and a beer fermented under pressure). In a triangle test, the probability of one judge picking the (correct) odd beer by chance is 33% (1 out of 3 sample cups). What we want to determine, statistically, is whether the result of this experiment (identification of the odd beer) could have occurred purely by chance or by some cause (e.g. the method of fermentation used). To do this, we need to develop a hypothesis which will either be rejected or not rejected.

A hypothesis is typically stated as the norm or status quo. In this case, the hypothesis (Ho) is stated as p = .33 (the true proportion of the judges that can correctly identify the odd-beer-out strictly by chance) meaning that all three beers are indistinguishable, except by chance.

The alternate hypothesis (Ha) is that the true proportion of the judges that can correctly identify the odd-beer-out is greater than .33. If (Ha) turns out to be true, it doesn't bode well for pressurized fermentation as an indistinguishable substitute to lagering.

By taking samples (taste tests) and analyzing the results, we can determine a sample "p-value" which is the probability of having observed at least as many correct tasters if (Ho) is true. Statisticians establish a threshold or cut off p-value (alpha) before testing typically equal to .05 or 5%. If our sample p-value turns out to be greater than the 5% cut-off, our hypothesis cannot be rejected - meaning that the beers are the same (indistinguishable from a sensory standpoint). If our sample p-value turns out to be less than the 5% cut-off, our hypothesis can be rejected - meaning that the beers are NOT the same (from a sensory standpoint).

For example, if we take 5 judges and ask them to take the triangle test, the odds that all 5 will randomly pick the odd-beer-out by chance is 0.004 or 0.4% (which is below our threshold of 5%). I could go through the math, but suffice it to say that it is a binomial (apprx bell curve) distribution and 0.4% is the calculated number (attached is a graph).

Let's say, in fact, all 5 judges correctly picked the odd beer. Then it must be inferred that there is in fact a difference between the two beers not attributed to chance (because the odds are very low that all 5 judges would correctly pick the odd beer by chance alone). In this case, (Ho) would be rejected (and Ha would be accepted) and it can be assumed that pressurized fermentation does NOT result in a beer with the same flavor profile as one fermented under traditional lagering.

The triangle test will be conducted in November and will result in a calculated “p-value” based on number of participants and correct choices. The beer bottles will be randomly numbered and randomly placed in each case row – e.g. A-B-A, A-A-B, B-A-A, etc where B is the beer fermented under pressure.



bionomial.pptx
 Description:

Download
 Filename:  bionomial.pptx
 Filesize:  41.92 KB
 Downloaded:  689 Time(s)


test form.pptx
 Description:

Download
 Filename:  test form.pptx
 Filesize:  42.44 KB
 Downloaded:  589 Time(s)

Back to top
Ozarks Mountain Brew




Joined: 22 May 2013
Posts: 737
Location: The Ozark Mountains of Missouri


PostLink    Posted: Thu Oct 11, 2018 4:09 am    Post subject: Reply with quote

funny this could be phycology or just plain sweet tooths but when you create a blind test I've found that most non brewers or inexperienced drinkers like the sweeter beer, so using them in the taste test can throw the results off
_________________
"Imagination is more important than knowledge. Knowledge is limited. Imagination encircles the world."
Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Thu Oct 11, 2018 10:30 am    Post subject: Reply with quote

Ozarks, interesting observation- i did not know that. However, the prime objective of this triangle test is to determine if there is a discernable sensory (taste) difference between the beers - good or bad. the choice of which beer someone likes is of secondary importance for this particular test. it is assumed that if someone perceives one of the beers as sweeter, then that person will select that beer as the odd one - which is perfectly fine for our objective. it is conceivable that a judge dislikes all three samples equally which is fine as well - perhaps that particular judge doesn't like lagers as much as ales. That judge's observation is still valid. Since the results could be applied to the approximately 54 million beer drinkers in the US ( with an appropriate margin of error) it may actually be beneficial to have inexperienced beer drinkers in the mix. Hope this clarifies the objective.

Last edited by mjo2125 on Fri Oct 12, 2018 11:51 am; edited 1 time in total
Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Thu Oct 11, 2018 12:15 pm    Post subject: Reply with quote

Ozarks, your point is so well taken that I’m thinking about removing the second question (favorite beer selection) from the survey form so as not to confuse the judges...
Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Thu Oct 11, 2018 9:40 pm    Post subject: Reply with quote

Attached is a revised sample form - selection for favorite beer removed to avoid confusion and conflict with the overall objective of the test.


test form.pptx
 Description:

Download
 Filename:  test form.pptx
 Filesize:  40.57 KB
 Downloaded:  574 Time(s)

Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Thu Oct 25, 2018 1:11 am    Post subject: Reply with quote

Sensory test announcement:

http://daytondraft.org/draft-meeting-november-19th/
Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Sun Oct 28, 2018 1:47 pm    Post subject: Reply with quote

The Molecular & Cellular Biochemistry Dept. at Indiana University (The Bochman Lab) has agreed to run Gas Chromatography analysis of our samples. Looking to compare levels of diacetyl, ethyl acetate and isoamyl acetate.

Below is a link to The Bochman Lab – looks like they’re bioprospecting for wild yeasts for brewing…

https://www.indiana.edu/~helicase/research.html#fermentation


Last edited by mjo2125 on Wed Oct 31, 2018 10:01 pm; edited 1 time in total
Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Wed Oct 31, 2018 9:58 pm    Post subject: Reply with quote

Next Step: Sampling

Samples (50 ml each) sent to the lab for GC analysis. May have results next week to compare with sensory test in a couple of weeks. Note: lab doesn't require carbonated samples for analysis. Minimum volume per sample required is 20 ml for GC equipment. Deliverables include chromatograms for each beer.



IMG_0092.jpg
 Description:
 Filesize:  109.02 KB
 Viewed:  20048 Time(s)

IMG_0092.jpg


Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Fri Nov 16, 2018 12:54 am    Post subject: Reply with quote

GC Raw Data:

The lab conducted head space analysis on the sample groups "A1 & A2" - traditional lager and "B1, B2 & B3" - pressure lager. Attached are preliminary chromatogram results. The numbers are time, the abundance peaks need to be converted to ppm to provide meaningful comparisons with published sensory thresholds for each major ester. The lab will conduct a liquid extraction next also to identify the levels of the Diacetyl keytone.

Preliminary results indicate that the abundance of esters are consistently higher in the pressure lager than traditional lager samples - the critical questions are - What is the corresponding ppm levels and how does this compare with sensory thresholds for each ester?

Over 16 esters were identified. More info to come.



GC Raw Data.pdf
 Description:

Download
 Filename:  GC Raw Data.pdf
 Filesize:  159.08 KB
 Downloaded:  493 Time(s)

Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Sun Nov 18, 2018 1:29 am    Post subject: Reply with quote

The GC data was converted to the attached flavor abundance chart. The quantitative results were unexpected as the general consensus is that pressure fermentation inhibits yeast growth resulting in lower ester concentrations than by traditional lagering.

The quantitative results could mean that different yeast strains have different pressure tolerances (1 Bar, 2 Bar even 3 Bar)and that different pressures can be dialed in for specific strains to achieve the desired results (an experiment for another day).

The next step is to see if the differences between the two beers are detectable by taste and if the sensory results are statistically significant.



DRAFT-GC Flavor Chart.pptx
 Description:

Download
 Filename:  DRAFT-GC Flavor Chart.pptx
 Filesize:  74.64 KB
 Downloaded:  398 Time(s)

Back to top
mjo2125




Joined: 27 Feb 2017
Posts: 248
Location: Dayton, OH


PostLink    Posted: Tue Nov 20, 2018 10:44 pm    Post subject: Reply with quote

Last Step: Sensory Triangle Test

15 judges were selected to sample the beers.

• 12 out of 15 taste testers correctly identified the odd-beer-out (pressure lager). The calculated p-value (probability of the sensory test results being attributed to chance alone) is 0.028% which crosses the standard 5% level for statistical significance. The results are statistically significant.

• The hypothesis that the odd-beer-out can only be identified by chance is rejected. The alternative hypothesis that the beers are statistically different via fermentation method is accepted.

• The GC lab data and sensory taste tests support each other. The GC analysis predicted higher than expected ester levels in the pressure lager which was later confirmed with the triangle test results. Note: The taste testers were not made aware of the GC data until after the triangle test was completed.

A complete PowerPoint Slide deck is available. It provides some background into pressure fermentation, the test recipe, procedures and controls used in the testing from start to finish. It includes both quantitative and qualitative results as well as thoughts and observations for future consideration.

Below is a link to the slide deck:

https://www.facebook.com/groups/DAYTONDRAFT/permalink/2411460145547665/

Forgot to mention that it was not all bad news for pressure fermentation - based on comments made on the score sheets, there was a taste preference for the pressure lagers (smoother taste).

This completes the experiment.
Back to top
Display posts from previous:   
Post new topic   Reply to topic   Printer-friendly view    TheElectricBrewery.com Forum Index -> Brewing Techniques All times are GMT
Page 1 of 1
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum



Forum powered by phpBB © phpBB Group