My Spontaneous Beer Evolution

Intro

Jean Van Roy of Cantillon
monitoring the boil kettle
 after adding the hops

During the Spring of 2014, my wife and I had the fortune of spending two weeks in Belgium, a well-deserved early graduation present to myself. One of the highlights of our journey was getting to attend an open brew day held at Cantillon in Brussels. We visited Cantillon about 5 times total during our whole trip, but open brew day was absolutely the most memorable visit. Seeing the turbid mash taking place in that ancient looking mash tun and smelling the aged hops after Jean had added them in the kettle are memories that stand out strongly in my mind. Visitors were allowed to freely explore the facility, which consisted of various floors of reasonably small rooms often connected by a narrow staircase or dark passage used for bottle conditioning. Even having read so much about the brewery before our visit, I was somewhat shocked at how humble their facility was. It appeared as if nothing in the brewery had change for many decades (aside from their bottling line and some of the newer stainless tanks that they use for fruited Lambic). Their facility, their gracious staff, and the beers they produce conveyed a story of modesty steeped in family tradition. Being highly intrigued by the Lambic brewing process, I began my pursuit of adapting these methods to my own brewing. This is where my spontaneous story begins.

Batch 1

Batch 1-July 20, 2014
My first turbid mash and 
attempt at a Lambic-inspired beer

• 70% Belgian Pilsen Malt
• 30% raw red wheat
• 0.67oz/gallon of aged pellet hops
• Turbid Mash
• Wyeast Lambic Blend
• No coolship

Prior to our trip to Belgium, I had read about Lambic and its unique production techniques from various sources and listened to several interviews with the brewers and blenders themselves. From this information I had formed a decent understanding of what the process looked like and how it varied from modern brewing practices. Every visit we made to Cantillon was a learning experience for me as I had many of their beers for the first time and got to ask questions of the staff and brewers. At Cantillon's open brew day, a gentleman had overheard me inquiring about their turbid mashing schedule and he pulled me aside to show me his notebook which contained procedural information from the various breweries he had visited. Several years later I came to realize that this gentleman was actually Dave Jansen of Hors Catégorie Brewing. After experiencing the magnificent sights, smells and suds of breweries across Belgium we returned home where I quickly began planning my first attempt at brewing a Lambic-inspired beer.

I am going to try to be somewhat brief on the exact details of each batch in this post and focus on highlighting the process changes that were made from batch to batch and my thought process behind these changes. I am preparing several upcoming posts that will stem from this one where I will provide more low level details for each batch such as OG, FG, pH, tasting notes, etc. For the first batch, I took the approach of closely following what worked for Cantillon as I had more data on their process than any of the other Lambic brewers. I planned to start from there and adapt my own process once I found out what did and didn’t work for me. Additionally, I will never call what I am brewing "Lambic" out of respect for those brewers who have kept the appellation alive to this day. I may describe my beers as "Lambic-style" or "Lambic-inspired", but never simply "Lambic". I ended up brewing Batch 1 in July and thought it unwise to do a coolship in the middle of a hot Missouri summer. Instead, I chose to use Wyeast Lambic Blend as the source of my microbes and chill the beer with traditional methods. My friend Matt was collaborating on this beer with me and was able to source some aged pellet hops from our local homebrew supply store. The aspect of this batch that intimidated me the most was the turbid mash. A turbid mash is somewhat similar to a decoction in that a portion of the mash is removed and heated in a secondary vessel and later returned to the mash tun. The purpose of this process is to create highly starchy wort that will provide a long term food source for the spontaneous yeast and bacteria. I will detail my turbid mashing process in full on a later post as I feel the topic is involved enough to warrant thorough discussion. 

On brew day, we were amazed at the turbidity of the runnings that came off of the mash. I had previously done decoctions for various lagers I had brewed so I had a little experience in handling a mash where a portion is removed. After the wort was transferred to the kettle the extended boil began and the aged hops were added. After chilling the wort with an immersion chiller, we filled two 6-gallon glass carboys and pitched the Lambic Blend along with the dregs from a celebratory bottle of Cantillon Saint Lamvinus. Fermentation wasn’t quite as vigorous as I had expected, but I had never used this blend so I didn’t have a baseline expectation. The beer was sampled around the 3-month mark and it was noted to be excessively bitter. I recall it tasting somewhat like an unpleasantly funky, potpourri-flavored IPA. The hop bitterness and flavor were overpowering and not reminiscent of the aged hop funk I associated with Lambic. This batch was sampled a couple more times until it was decided that it needed to be dumped around the 12-month mark. While it was a failure in the fact that all 11 gallons of the beer were dumped; Batch 1 was a great learning experience. I had successfully completed my first turbid mash and determined that the aged hops should be selected with much more scrutiny. We did acquire hops that were many years old, but they were not aged appropriately for use in a Lambic-inspired beer. I don’t recall asking about how the hops had been stored, but I am guessing the majority of their shelf life was spent in a refrigerator in a vacuum-sealed bag. Regardless of my rookie mistake, I didn’t call it quits.

Batch 2

Batch 2-November 2, 2014
My first attempt at a coolship 
and 100% spontaneous fermentation 

·         70% Belgian Pilsen Malt
·         30% raw red wheat
·         0.67oz/gallon of aged pellet hops
·         Turbid Mash
·         Half spontaneous, half Wyeast Lambic Blend
·         Kettle coolship

I had realized my mistake on Batch 1 after pulling the first sample at the 3-month mark and I quickly ordered some aged hop pellets from an online source. They were listed as a “Lambic Hop Blend”, noted to be an aroma variety, and an alpha acid content was defined. As a result, I had a higher level of confidence in how thoroughly the hops had been aged. When I received the new batch of aged hops they smelled much more like the aged hops I had encountered in Belgium than the hops I had used for Batch 1. Well-aged hops have a smell that I find quite pleasant. To me it is very earthy and herbaceous, somewhat like a funky tea. On the other hand, hops that are only partially aged (maybe more accurately described as “old” or “stale” hops) generally seem a little less enjoyable to me. They have a harsher character that I would assume comes from a declining but still present alpha acids.

For Batch 2, I left the general recipe alone, but decided to incorporate a coolship into this batch. A coolship is a large shallow vessel that is used by Lambic brewers to ambiently cool the wort over an extended period of time. During this open cooling, microbes are allowed to fall into the wort and inoculate it spontaneously. For Lambic production, pitching any yeast is prohibited and the brewers rely on their natural environment to provide the necessary yeast and bacteria for fermentation. For my coolship I used my 20-gallon stainless kettle as the coolship vessel. I boiled the 11-gallon batch of wort in a 20-gallon kettle and then moved the kettle out into my yard where I placed a coarse mesh bag over the top of the kettle to prevent bugs from falling in. I assumed the cooling would occur rather rapidly in such a small vessel so I stacked cinder blocks around the kettle to help insulate it a bit. 

The kettle coolship (kettleship?) was left outside until the following morning (~12 hours) when the wort was racked from the coolship directly into two glass carboys. Just as a geographical/agricultural reference point, we lived several miles outside of town, surrounded by cattle farms and hay fields. One glass carboy was left to spontaneously ferment and the other was inoculated with a single smack pack of Wyeast Lambic Blend. I was a bit nervous to commit the entire batch to 100% spontaneous fermentation, as I had never tried it before so I figured I would increase my odds of success by inoculating half the batch with lab cultures. Finally, there was 1oz. each of new and used medium toast French oak cubes added to each carboy for a total of 2oz. per carboy. The cubes had been boiled prior to adding to the carboy to kill any present yeast or bacteria that might be present from being used in previous beers. Both carboys were checked around the 1, 3, 6, 9 and 12 month marks. Initially, the spontaneous carboy seemed to have less acidity and more sweetness than the inoculated carboy. Unlike Batch 1, both carboys had a moderate bitterness that wasn't overwhelming. The quality of bitterness seemed to me to be very reminiscent of what I had experienced in Brett beers with high hopping rates. I would describe this as a somewhat phenolic, almost burnt plastic character, which at low levels in Brett IPAs I find intriguing. As time progressed the two beers became more and more similar and even became difficult to distinguish when I didn't properly label the glassware I had used for sampling. I will save final tasting notes for Batch 2 for a later blending post, but will go ahead and say that the spontaneous carboy produced a much more interesting beer than the inoculated carboy.

Overall, I considered Batch 2 a success as neither the spontaneous carboy or the inoculated carboy was worthy of dumping. Also, I believe I verified that my failure in Batch 1 was indeed caused by the poorly aged hops.

Batch 3

Batch 3-November 21, 2015

Coolship experiment, same 

volumes in insulated 

and un-insulated vessels

•  66% Belgian Pilsen Malt
• 34% raw, locally grown triticale
• 0.60oz/gallon of aged pellet hops
• Turbid Mash
• 100% Spontaneous
• Insulated and Uninsulated Coolships

Up to this point I had continued with the mindset of keeping my process as close to Cantillon's as possible. Having finally incorporated all of their major process variables (other than fermenting in a full size wine barrel) and achieved success with a spontaneous fermentation, I felt comfortable beginning to experiment a bit with my recipe and process. Due to the way in which it is produced, Lambic is a beer that tells a unique story about where it came from. The populations of ambient microbes that end up falling into the coolship are influenced to some degree by the weather and surrounding environment and provide a unique terroir to the beer that can possibly vary from one brew day to the next. I decided that I wanted to expand this terroir mindset in my beer a little further. I have a friend that plants triticale as a grazing food source for his cows. After telling him about the spontaneous beers I had brewed up to that point, he encouraged me to incorporate some triticale into the beer. I had dismissed the idea as I assumed triticale would taste more like rye than wheat and I didn't know how it would impact the resulting beer several years down the road. During one visit to my friend's house I noticed some grains that he had thrown out behind his barn, so I did as any insane person would do. I picked a few kernels of the grains up off the ground, likely only feet away from a fresh pile of cow manure, and popped them in my mouth. They were much softer than the raw wheat I had been using and actually tasted more wheat-like to me. I asked my friend why there was a pile of wheat on the ground next to his barn and he informed that was actually a pile of triticale seed he had thrown out because he found some worms in the storage bin. He also told me that he sourced the triticale seed from a neighbor of his who grew and harvested the triticale himself. After this experience, I decided to replace the raw wheat I had been using with locally grown raw triticale as an additional source of Southwest Missouri terroir. In addition to this major grist adjustment, I reduced the hopping rate from 0.67oz/gallon to 0.6oz/gallon.

After my casual attempt to insulate the coolship in Batch 2, I began to wonder what the effects of insulating vs. not insulating (at the homebrew-scale) might have on the resulting beer. Would a longer stand at above 86F/30C result in larger populations of lactic acid producing bacteria and a lower final pH? Would a longer stand below 86F/30C result in larger populations of yeast and more brettanomyces forward character? To test this, I decided to split my coolship for Batch 3 between two different vessels. One vessel was a 10 gallon insulated beverage cooler and the other was a 10 gallon aluminum stock pot, both seen in the photo above. The idea being that the insulated vessel would spend more time above 86F/30C than the non-insulated. The vessels selected had nearly identical internal dimensions so that there would be a comparable surface to volume ratio.

Following turbid mashing and boiling, the wort (1.049 at this point) was transferred to the two sanitized vessels and the hops were removed. Both vessels were sat directly on the ground and left outside overnight for 10 hours with no mesh covering the wort this time. The insulated coolship reach a terminal temperature of 100F/37.7C and the un-insulated reached a terminal temperature of 50F/10C. I could have left the insulated coolship longer so that it could get closer to my target fermentation temp of 70F/21.1C, but I felt that keeping the inoculation time identical between the two vessels was important to keep the numbers of experimental variants limited. We had moved to a new home between batches 2 and 3 and I hadn't yet realized that the sun would be shining right where I had set my coolships the night before. Therefore, I felt the need to limit the coolship to 10 hours to prevent the rising sun from beaming down on my highly hopped wort. As an updated geographical/agricultural reference point, we at this point lived within a mile of town, surrounded by cattle farms, forests, and hay fields.

I am finding it difficult to not include all of the low level details for Batch 3 in this post, but I have decided that I am going to do a full writeup for Batch 3 as I feel it produced some really interesting data points and some really meaningful results that will likely influence further coolship experimentation. However, I will go ahead and share that the insulated and uninsulated worts produced two uniquely identifiable beers and I believe this is absolutely a result of the cooling rate of each wort.

Batch 4

Batch 4-November 19, 2016
First year to reduce hopping
rate to 0.3oz/gallon (2.24g/L

·         66% Belgian Pilsen Malt
·         34% raw red wheat
·         0.3oz/gallon of aged pellet hops
·         Turbid Mash
·         Half spontaneous, half wood inoculated
·         Coolship

I must say that I got a little lazy for Batch 4 and my horrible note taking for that brew day supports this. There had been some discussion online regarding how thoroughly a barrel should be cleaned and sanitized prior to being used for spontaneous fermentation. This seemed like a great opportunity for a split batch experiment as the experimental variable was a cold-side process variable. To simulate using a poorly sanitized barrel, I took 1oz. of used oak cubes and soaked them in a slurry of carboy dregs and beer that I acquired from Batch 2. After soaking the cubes for at least a month, they were briefly rinsed with hot tap water to simulate a used barrel being inadequately heat sanitized.

I also made one major recipe adjustment for Batch 4. Having now seen what a hopping rate of 0.6oz./gallon does over the course of a few years, I was wondering if I had been hopping my turbid wort too heavily. At this point I was aging some two-year-old beer hopped at this rate and it wasn't achieving the acidity that I expect to find in Gueuze or Lambic. The hop flavor and bitterness was fading over time in all of my batches, but I determined that reducing the hopping rate was needed to achieve the level of acidity I was targeting. At this point I had just run out of aged pellet hops that I had been using for the last two years and switched over to some aged leaf hops that I had acquired from a different source. I purchased several pounds of the leaf hops as "choice debittered" and upon receiving them in January 2016 loosely packed them into a large brown paper bag that was held at room temperature. I figured cutting the hopping rate in half would produce a beer with more acidity if that was the limiting variable. Therefore, 0.3oz./gallon was the hopping rate used for Batch 4.

Following my standard brew day and coolship process, the wort was racked into glass carboys with one of them being inoculated with the oak cubes. I only recently realized that I never added any oak to the other half of the batch. Looking back, I think it would have been a more apples-to-apples scenario if both carboys received the same quantity of oak cubes, with one carboy getting poorly sanitized and the other getting thoroughly sanitized, but I was still able to reach a conclusion fairly quickly regarding the differences. Upon sampling both carboys around the 3-month mark, it was noted that the oak-inoculated carboy was already at a much lower gravity than the spontaneous carboy and it exhibited a much cleaner, fermentation character. This batch would be the last one required for me to produce a Gueuze-inspired blend of 1,2 and 3 year old beers in late 2017, but the high level of attenuation observed in the oak-inoculated carboy seemed to indicate that it may not have enough residual sugars in a year to prime the blend without adding additional sugars. For these reasons, I determined that thoroughly sanitizing the wood used for a spontaneous beer is critical to ensure attenuation is not achieved too quickly. If a beer is being designed with the intent of fermenting over a period of several years, introducing microbes that speed up this process is a little counter-productive in my opinion. The hop flavor for this batch was fairly pronounced, but the perceived bitterness was significantly lower than my previous batches. There wasn't yet a noticeable acidity, but from my research, Lambic doesn't usually develop significant acidity until after the second or third summer.

Batch 5-January 27. 2018
Split batch IBU experiment
0.3oz/gallon & 0.6oz/gallon

Batch 5

• 62.5% Belgian Pilsen Malt
• 37.5% raw, locally grown triticale
• 0.6oz/gallon & 0.3oz/gallon
• Turbid Mash
• 100% Spontaneous
• Coolship

Brew day for Batch 5 was only a month ago and I have set this one up to be a 3-year-long experiment so I will attempt to be brief. A few weeks after brew day for Batch 4, I found my self in Denver, CO for work. On this trip I was able to stop by Black Project Spontaneous & Wild Ales and speak with James Howat about some of my spontaneous fermentation experiences. After he showed me their newly expanded barrel room and variable capacity coolship, I commented that I was surprised at the lack of acidity I had noticed in my spontaneous beers. He encouraged me to send a wort sample to a lab to get an IBU measurement and told me about a method he uses to achieve a target "Theoretical IBU." (I will discuss this further in  an upcoming post dedicated to Batch 5)

Following my conversation with James at Black Project, I was motivated and inspired to incorporate IBU testing into my Batch 5 experiment. After several months of contemplation, I decided that I would perform a single turbid mash and split the wort between two identical boil kettles. One boil kettle would receive aged hops at a rate 0.3oz/gallon while the other would be hopped at 0.6oz/gallon. Each kettle would then be move outside and serve double-duty as a coolship. Lab analyses would be performed to acquire an alpha acid content of the aged hops before the boil and IBU testing would be performed on each wort following transfer out of the coolship. I had met Jim Gawenis of Sweetwater Science Labs at the SEC Craft Beer Conference in August of 2017 and coordinated all of my lab analysis needs with him. The plan for this experiment is to acquire an initial IBU measurement immediately following brew day and a final IBU measurement at the end of the 3-year experiment. Additionally, 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. Being the Excel nerd that I am, temperature was logged of both coolships as they cooled and a graph of that has been generated. I now have all of the lab analysis back and am working on gathering my thoughts and doing a little research for the Batch 5 write-up. I was definitely shocked by the results.

Conclusion

Having now completed my 4th year of turbid-mashed, age-hopped, long-boiled, open-cooled, spontaneously fermented beers, I am still no veteran, but I feel like I have had some meaningful experiences that are worth sharing. The learning process for spontaneous fermentation has been slow for me since I am giving the beers 1-3 years to mature and I have only been brewing once per year, but I am definitely learning something new each year and carrying that knowledge forward. I look forward to continuing to learn and share my experiences with you. I hope you have enjoyed reading about my evolution in spontaneous brewing! I would love to hear your thoughts on any of the processes or results I have described so feel free to add comments below. While this post is lengthy, it is missing a lot of low-level details (OG, FG, pH, IBU, time, temperature, etc.) that will be included in many of my future posts. There will definitely be an ebb and flow between art and science in my posts and I believe this post probably falls right in the middle of the two somewhere. I have a few more spontaneous beer posts that will stem off of this one and will likely have a saison post in the coming months as well.