2017 Coolship Experiment - Hopping Rate

 Introduction

This post will be updated regularly as the experiment progresses so it may be possible that my opinions change as time goes on and I research the topic further. This experiment is scheduled to run from January 2018 to January 2021.

Turbid wort being pulled from mash

2017 proved to be a year of thorough planning in my little home brewery. I had spent the better part of a year planning for my first 3-year blend of spontaneous beers and in tandem had been planning the coolship experiment for that year as well. If you are new to my blog or have not read my posts on spontaneous brewing, I highly recommend reading at least one of the following posts for background information on my spontaneous brewing process and experiences up to this point:

My Spontaneous Beer Evolution (This is the best single-source summary. If you only read one background post, read this one)

2015 Coolship Experiment - Cooling Rate (This sets a baseline for coolship experiments going forward)

Spontaneous Beer Blending 2017 (This summarizes how my spontaneous beers turned out for my first 3 coolship seasons)

Turbid wort recirculating over grain bed

As I have already stated in my summary of Batch 4, I had noticed over time that as my spontaneous beers were aging the hop bitterness was fading, but the acidity still seemed low for what I expect in a spontaneous beer. For that reason, I chose to reduce my hopping rate on Batch 4 from 0.6oz./gallon (4.48g/L) to 0.3oz./gallon (2.24g/L). Shortly after brew day for Batch 4, I was able to visit Black Project Spontaneous & Wild Ales in Denver, CO where I spoke to James Howat about my overly-hoppy spontaneous beers. My biggest question for him was the following. Up to that point I had been using thoroughly aged hops at the rates that Lambic brewers were using so why weren't my beers developing more acidity? James explained to me that the IBUs in the spontaneous wort I was producing were probably much higher than the IBUs in the spontaneous wort of authentic Lambic. He also suggested that I have a lab test my wort or finished beer to determine the exact level of IBUs present. This conversation provided the inspiration I needed to begin planning my next coolship experiment, Hopping Rate.

Approach

For my 2017 Coolship Experiment, I knew I wanted to incorporate IBU testing, but I wanted to do it in way that maximized my learning experience and produced as many meaningful data points as possible. After several months of contemplation, I decided that I would produce two different worts with two different hopping rates, 0.6oz./gallon (4.48g/L) and 0.3oz./gallon (2.24g/L). These two hopping rates represent an approximated average maximum and minimum rate as reported by Lambic brewers. Therefore, I expected these hopping rates would produce two uniquely distinguishable beers and two unique sets of data.

3.6oz. of aged hops for 0.3oz./gallon batch

7.2oz. of aged hops for 0.6oz./gallon batch

Prior to brew day, a sample of the aged hops to be used will be sent to Sweetwater Science Labs to determine their alpha and beta acid content. This information will not be used to influence the experiment itself, but will serve as an additional data point from which to draw conclusions from at the end of the experiment. Both worts will be produced and coolshipped at the same time to minimize introducing additional experimental variants. The worts will be produced from a single turbid mash and split between two identical 34-gallon, electric boil kettles. One boil kettle will receive aged hops at a rate 0.6oz./gallon (4.48g/L) while the other will be hopped at 0.3oz./gallon (2.24g/L). At flameout, the kettles will then be moved outside where they will both ambiently cool. For this experiment, the kettles will serve double-duty as both the boil kettle and the coolship. The hops will remain in both kettles until the wort is transferred to a fermentation vessel. While transferring the wort from each coolship to their respective fermentation vessels, a sample of each wort will be collected and sent to Sweetwater Science Labs to perform IBU analysis on. IBU analysis will be done at the beginning and the end of the 3-year experiment (and possibly at the 1 and 2-year marks as well) to determine how much the IBUs change over time, if at all. During the 3-year experiment, gravity and pH will be checked on an interval of every 3 months for the first year and every 6 months for the remainder of the experiment. These data points will be plotted on a graph that will be updated on this blog at every measurement interval. Additionally, a data logger will be used to record the temperature of each wort as they cool. This data will be used to generate a temperature versus time plot that will be appended to the Coolship Weather page.

Recipe Specifications

Beer Name: MOambic 2017/2018 (Spontaneous Batch #5)

Brew Date: 1-27-2018

Batch Size:  gal (L) 24gallon (90.8L)

Estimated OG: 1.050

Estimated: Color: 3.4 SRM

Estimated IBU: <10 IBUs

Brewhouse Efficiency: 70%

Boil Time: 180 minutes

Grain:

62.5% - 30lbs (13.6kg) - Belgian Pilsen
37.5% - 18lbs (8.16kg) - Raw Triticale 

Hops:

Boil: 180min - 3.6oz Aged Leaf Hops [0.5%] - 3 IBUs (0.3oz./gallon)
Boil: 180min - 7.2oz Aged Leaf Hops [0.5%] - 6 IBUs (0.6oz./gallon)

Yeast:
No yeast pitched. Wort was coolshipped outside for 12 hours to inoculate.

Mash:

Dough In - 8 gallons of water (0.66qts./lb. or 1.39L/kg)
Rest -  0min @ 118°F (47.7°C)
Pull - 1 gallon of turbid runnings
Add - 4 gallons of water to mash tun
Rest -  15min @ 133°F (56.1°C)
Pull - 4 gallons of turbid runnings
Add - 6 gallons of water to mash tun
Rest -  45min @ 152°F (66.6°C)
Rest -  45min @ 162°F (72.2°C)
Rest -  10min @ 172°F (77.7°C)

Sparge:

Add - heated turbid runnings back to mash tun
Sparge with 180°F (82.2°C) water

Misc.: 

Notes:
1-27-2018
At beginning of 152°F rest, pH was 5.96pH. 30mL of 88% Lactic acid was added.
Ten minutes later pH was 5.30pH.
At beginning of 162°F rest, 10mL of 88% Lactic acid was added.
Ten minutes later pH was 5.20pH.
29 gallons of 1.054 wort was collected and split between two the two identical 34-gallon kettles.
5 gallons of water was added to each kettle to bring pre-boil gravity to 1.040.
After two hours of boiling gravity was 1.043.
After three hours of boiling gravity was 1.049.
Brew system rolled outside at end of boil where kettles were left to open cool for 12 hours.
1-28-2018
Wort was transferred from each kettle into a horny tank to ensure wort was inoculated homogeneously.
OG and pH of each wort was measured in horny tank.
0.3oz/gallon wort was 1.054SG and 5.5pH.
0.6oz/gallon wort was 1.053SG and 5.3pH.
Samples of each wort collected in 12oz. bottles and shipped to Sweetwater Science Labs for IBU analysis.
The 0.3oz/gallon wort was placed into a 6-gallon and 3-gallon glass carboy.
The 0.6oz/gallon wort was placed into a 6-gallon and 3-gallon glass carboy.
An additional 6-gallon glass carboy was filled with a 50/50 blend of the two worts.
All five carboys were placed in a fermentation chamber set to 68°F (20°C).
1-30-2018
Fermentation chamber set to 70°F (21.1°C).
1-31-2018
First signs of light fermentation lasted a few days
2-10-2018
All carboys removed from fermentation chamber and placed in my basement.

29 gallons of wort was collected into a single kettle.

Wort was split between two identical kettles and boiled for 3 hours.

Results

The first data point from this experiment was the hop analysis performed by Sweetwater Science Labs prior to brew day. It was found that the alpha acid content was 0.5% and the beta acid content was 0.2%. With an alpha content of 0.5%, the IBU's for the 0.3oz./gallon and 0.6oz./gallon batches were calculated to be ~3IBUs and ~6IBUs respectively. However, when the worts were analyzed post-coolship, the IBUs for the 0.3oz./gallon and 0.6oz./gallon were measured to be 71.9IBUs (StdDev: 0.6) and 127.3IBUs (StdDev: 0.8) respectively. These measurements were taken using ASBC standard method for IBU analysis. I found these numbers quite shocking as the worts did not taste all that bitter to my palette. I have brewed several IPAs and Double IPAs over my 6+ years of homebrewing and those worts generally tasted somewhat aggressively bitter prior to fermentation. However, these aged-hop worts did not have that same level of aggressive bitterness even though they had measured IBUs in the range of a Double IPA. In my mind, this makes me question what the IBU really means in the grand scheme of beer. For one, I believe it means that the International Bitterness Unit (IBU) is not an adequate unit for measuring perceived bitterness across all applications of hop usage. In the case of this experiment, I am using hops in a very unconventional manner by aging them for several years, something you wouldn't/shouldn't intentionally do for something like an IPA. I am assuming that similarly deceiving IBU values might be acquired from something like a New England style Pale Ale given the unusual hop usage in that style of beer. For the duration of this experiment, I am planning on measuring the IBUs again at the 1, 2, and 3-year marks to see how the IBUs change over time. 

Coolship Data - January 27, 2018

The weather for my brew day was exactly what I typically target for doing a coolship, a daily high of 50-60°F (10-15.55°C) and an overnight low of ~32°F (0°C). However, with my batch size (typically 12gal or 45.42L), I do have to insulate the coolship vessels to some degree to prevent grossly undershooting my target fermentation temperature of 70°F (21.11°C). Having now successfully produced at least one batch that was coolshipped at slightly warmer temperatures, I will probably target an overnight low of 40-50°F (4.44-10°C) and a daytime high of less than 70°F (21.11°C). The two worts were open cooled from ~6PM to 6AM for a total coolship time of 12 hours. Both worts cooled at the same rate which was expected due to both kettles being identical in size and containing the same volume of liquid.

UPDATED 11-4-2018
pH of all fermentation vessels over time

UPDATED 11-4-2018
SG of all fermentation vessels over time

Given the 24-gallon total batch volume and number of carboys I had available at the time, I chose to fill a 3-gallon and 6-gallon glass carboy with each of the worts then blend the remaining wort into a single 6-gallon glass carboy. The specific gravity and pH of these 5 vessels over time is detailed in the graphs above. The pH of the 0.3oz./gallon wort was measured to be 5.5pH, the 0.6oz./gallon wort measured to be 5.3pH, and the blended wort calculated to be slightly less than 5.4pH. It has been shown that the addition of dry hops to a beer raises the pH so one might assume that the beer with more hops would have a higher pH. However, the opposite was true in this case. I have not yet found a resource that proves hot-side hop additions affect pH in the same way that cold-side additions so if anyone is aware of one please let me know. Otherwise, that may make for an interesting experiment in the future.

It is observed in the pH graph above that, while the 0.3oz./gallon beer starts out with a higher pH, it's pH does appear to drop more quickly that the 0.6oz./gallon beer and at the 6-month mark all of the 0.3oz./gallon beers are lower in pH than the 0.6oz./gallon beers. The blended beer appears to be tracking pH very similarly to the 0.6oz./gallon beer. I have observed similar behavior during split batch experiments in the past as well. When creating a blend from two unique worts, the result is that the blended wort behaves and tastes very similar to one of the worts. I have not yet seen a blended wort whose behavior was somewhere in between the behavior of the two unique worts it was comprised of.

The specific gravity of the 0.3oz./gallon wort was measured to be 1.054, the 0.6oz./gallon wort measured to be 1.053, and the blended wort calculated to be 1.0535. The most noticeable difference on the specific gravity graph at this point is the gravity difference between the 3-gallon and the 6-gallon carboys. The gravity of all three 6-gallon carboys appears to be dropping linearly while the 3-gallon carboys reached a lower gravity mush faster and are now beginning to taper the rate at which gravity is dropping. I did not expect this result, but will use this information to guide future long-term aging projects. At the 6-month mark, all beers are below 1.0045 as measured by a final gravity hydrometer. I did not expect all of the beers to attenuate so quickly as I had not frequently measured gravity in my three previous seasons of brewing spontaneous beers. It is known that Lambic producers typically rely on the residual sweetness of their 1-year old Lambic to help provide carbonation to their Gueuze blends. If I wish to stay true to traditional lambic production techniques, then I believe this suggests that I may need to modify my turbid mash procedure and/or the temperature that I allow the beers to ferment at in order to decrease the rate of attenuation.

Conclusion

Based on the sensory results of my first three years of brewing spontaneous beers, I believed that I was over-hopping those beers and based on the IBU measurements from this experiment I am now confident that I was. As stated in the introduction, it was my conversation with James Howat that provided my inspiration for this experiment and it was some advice he gave me that I believe provides the solution to my over-hopping problem. James told me about a method he uses for approximating the  virtual IBUs in his spontaneous beers that allows him to dial in his hopping rates before brew day. The method involves brewing a test batch using the same aged hops that will later be used in the spontaneous beer. After the wort is produced, a sample of the wort is sent to a lab for analysis. The resulting IBU measurement can be used to determine if the hopping rate for the upcoming spontaneous beer needs to be increased or decreased and a scalar can be applied to get closer to the brewer's target. For example, let's say a pilot batch is produced with a hopping rate of 0.6oz./gallon and the resulting beer measures 60IBUs. If the brewer was actually targeting 30IBUs it can be approximated that by cutting the hopping rate in half, the IBUs would also be cut in half. This is not a guarantee that the IBUs will be precisely 30, but it can help the brewer get much closer to their target IBU level. The IBU results from this experiment will serve this very purpose for my own spontaneous beers. Given that the 0.3oz./gallon beer resulted in a wort of 71.9IBUs, I plan on using a hopping rate of 0.15oz./gallon in my next beer in hopes of achieving something closer to 35IBUs. Also worth noting is that different lots/batches/varieties of hops would require repeating the above mentioned virtual IBU approximation.

Wort being transferred to horny tank after coolship outside

So why do we care about IBUs if they don't accurately represent perceived bitterness? In the case of aged-hop usage for bittering, the resulting beer has a much smoother bitterness than one would anticipate simply based on the measured IBUs. I believe this experiment demonstrates that with the 71.9IBU and 127.3IBU worts that are not perceived as being nearly as bitter as expected. However, if a brewer desires to produce something that is similar to lambic, having a knowledge of the measured IBUs in authentic lambic gives the brewer an extra data point to use in designing their own beers. After my conversation with James, Dave Janssen published a post on his blog, Hors Catégorie Brewing, about IBUs in Gueuze. In it he shares FG and IBU measurements for various commercially available Gueuzes and helps explain why it is that IBUs from aged hops aren't perceived as being as bitter as IBUs from non-aged hops. Perhaps the most shocking information from the post is the IBU measurement from Cantillon's 2013 Gueuze, which was measured at 30IBUs. How can a beer with 30 IBUs be so sour? It doesn't initially make sense, but Dave's explanation helps shed some light on why. I will expand on this further in future updates to this post.

Hops and trub in bottom of kettle

Quite honestly, the IBU measurements from this experiment had me completely stumped. The first and most obvious question is why are the measured IBUs from these worts so far off from the calculated IBUs? Secondly, why are the measurements so high? From my early batches of spontaneous beer, I learned that hops needed to be thoroughly aged and I had done so ever since. The hops from this batch were purchased as "debittered" (I don't remember what variety) and I aged them for nearly two years in a brown paper bag at room temperature. They smelled very similar to what I recalled of the hop bales in Cantillon's attic and I had used the hopping rate's that Cantillon had claimed to use so why were my IBUs so high? After pondering this for the last 7 months (and a few conversations with some intelligent friends) I believe that the hop variety of the aged hops may have a significant impact on the resultant measured IBUs. Different varieties of hops exhibit different characteristics due to each variety having varying levels of the hop oils including myrcene, humulene, caryophyllene, and farnesene. As hops age, these oils oxidize and are transformed into humulinones and hulupones. It is possible that some amount of individuality from each hop variety may carry over to producing aged hops that are also unique in some way. If this is true, using aged hops of different varieties could produce varying levels of hop aroma, flavor, and perceived bitterness in spontaneous beers. I believe this needs further explored and am currently crafting an experiment to do so.

Being a homebrewer who is heavily focused on the art and science of spontaneous fermentation, I have listened acutely to Lambic brewers when they have described their process and their recipe. I used to think that if a certain hopping rate worked for them that it would surely work for me as well. However, it does not appear that relying on Lambic hopping rates alone will consistently produce a beer similar in character to Lambic. Going forward, I will strongly encourage anyone who is serious about spontaneous fermentation to use the Virtual IBU method described above. Speaking from experience, it is quite disheartening to wait 3-4 years for a beer to be ready to drink only to find out that you were too heavy-handed with the hops.

Update: 

11-4-2018 - pH and SG measurements taken of all (5) vessels and plots updated.

The updated plot of specific gravity shows that attenuation of all vessels has slowed considerable and all are now at or below 1.001. I believe this may indicate that I need to work on my turbid mashing techniques to ensure there are more fermentables remaining in the young spontaneous beer for doing three-year blending. I can always add sugar (and did so for my first three-year blend), but relying on the sweetness of the young spontaneous beer for re-fermentation is a technique used by authentic Lambic producers so I have attempted to comply with those methods when possible.

The updated plot of pH shows that the pH of the 0.3oz./gallon wort is decreasing more rapidly than the pH of the 0.6oz./gallon wort and the 50/50 blended wort is somewhere between the two. Historically, the 3 gallon vessels have always remained lower in measured pH than the same wort in a 6-gallon vessel, but this time the 3-gallon vessel of the 0.3oz./gallon wort was 3.46pH while the same wort in a 6-gallon carboy was 3.41pH. Both of these measurements are in the range of what I typically see in my spontaneous beers when they have reached terminal acidity. Therefore, I wonder if the measurement differences of the different vessel sizes will become less obvious as the beers get closer to terminal gravity and acidity? I guess time will tell.

Data up to this point has been strictly numerical, but at the next check in I will add tasting notes as well as a second IBU measurement for the two different worts.