Posting this up here to see if anyone else has had similar issues, and what they have done about it. I was ripping my hair out over this issue for the last six months, and finally have the data to back up my conclusions. I apologize for the Novel like length of this post, but being a man of science, data is always required to support my conclusions, if i have missed something, someone please tell me.
We have had some recalls over the last 6 months or so related to over-fermented product in cans and bottles. This has only been an issue for some of our brands of beer, but not all them. It was not until recently that we were able to definitively pinpoint the cause of these issues. This document will hopefully serve as reassurance that we had (at the time) done everything in our power to ensure quality product out to market, and that, ultimately, this issue will cease moving forward. The underlying cause was very difficult to pinpoint without specialized equipment and/or laboratory analysis.
Background:
Upon startup of our new facility every batch of beer was tested in house AS WELL AS through a third party laboratory to ensure the beer was free of microbial contaminants. We have copies of these analysis’ for every batch of beer sold to market. Each and every single batch tested that went out to market showed that the beers were FREE AND CLEAR of the most common microbial contaminants found in the brewery using ASBC or EBC standard methods of detection. (American Society of Brewing Chemists & European Brewing Convention)
Upon startup of our new facility we switched house yeast strains and suppliers to find strains of yeast that would create a more desirable profile, had better flocculation (easing the strain put on by our filter during filtration), and were much more efficient at fermenting than our previous house strains.
Very early on this year we began to a notice an issue we had never before encountered in our canned products. Cans (and some bottles) had begun to referment to the point where the cans were exploding/becoming dangerous. It was at this point that we started down the long road of looking for the root cause of this issue.
Analysis:
Tank Cleaning/Sanitation Procedures and Validation
Transfer Lines/Filter Sanitation Validation
Exterior and Interior Testing for Microbial Contaminants
We use a variety of different microbiological media to look for spoilage organisms in our beers.
Packaging Line Cleaning and Sanitation Validation
Yeast Health and Management
Analysis of Re-fermenting Beers
We have had some recalls over the last 6 months or so related to over-fermented product in cans and bottles. This has only been an issue for some of our brands of beer, but not all them. It was not until recently that we were able to definitively pinpoint the cause of these issues. This document will hopefully serve as reassurance that we had (at the time) done everything in our power to ensure quality product out to market, and that, ultimately, this issue will cease moving forward. The underlying cause was very difficult to pinpoint without specialized equipment and/or laboratory analysis.
Background:
Upon startup of our new facility every batch of beer was tested in house AS WELL AS through a third party laboratory to ensure the beer was free of microbial contaminants. We have copies of these analysis’ for every batch of beer sold to market. Each and every single batch tested that went out to market showed that the beers were FREE AND CLEAR of the most common microbial contaminants found in the brewery using ASBC or EBC standard methods of detection. (American Society of Brewing Chemists & European Brewing Convention)
Upon startup of our new facility we switched house yeast strains and suppliers to find strains of yeast that would create a more desirable profile, had better flocculation (easing the strain put on by our filter during filtration), and were much more efficient at fermenting than our previous house strains.
Very early on this year we began to a notice an issue we had never before encountered in our canned products. Cans (and some bottles) had begun to referment to the point where the cans were exploding/becoming dangerous. It was at this point that we started down the long road of looking for the root cause of this issue.
Analysis:
Tank Cleaning/Sanitation Procedures and Validation
- Tank cleaning/sanitation procedures were validated through ATP Bioluminescence and found to be within the very tightest allowances for cleanliness and sanitation. Zero lumens were found, which means there was an undetectable amount of biomaterial on the inside of the tanks.
- This combined with visual inspection of the inside of each tank after every cleaning validated the issue was not from tank cleanliness/sanitation.
- Every time a tank or piece of equipment is cleaned or sanitized in the brewery chemical concentrations and strengths are validated with test strips/titrations to ensure proper concentrations within industry standards.
Transfer Lines/Filter Sanitation Validation
- All transfer lines and filters that are used to pump beer/wort from one place to another are pasteurized with hot water to 175F for 20 minutes just prior to use. When beer is being transferred to a bright beer tank or other aging vessel the transfer lines are also blown out with sanitary CO2 from our bulk tank. When we begin to transfer wort from our brewhouse to the fermenter we pack the lines/heatex with uncooled 200F and sit for 2-3 minutes to ensure everything is as sanitary as possible. (this is in addition to a pasteurization we run from our WP vessel through our Heat ex through our xfer lines to fermenter that runs between 170-180F for AT LEAST 20 minutes)
- This process was validated in house and externally by checking microbial stability of water inside transfer lines both during pasteurization and post blowdown of the lines/filter (there is some residual water in lines post blowdown, enough to pull a sample for microbial stability). No contaminants were found.
- For each batch we pull a wort stability sample to check for contaminants between the brewhouse and the fermenter. To date none of the samples have tested positive for common spoilage organisms.
Exterior and Interior Testing for Microbial Contaminants
We use a variety of different microbiological media to look for spoilage organisms in our beers.
- We currently test for the most common strains of Wild Yeast, Lactobacillus and Pediococcus found in most breweries using HLP, UBA, NBB and LWYM
- Our exterior lab tested for the same organisms using different media, MRS, UBA+A, LWYM & Lysine.
- All tests performed for all beer released to market came back negative for Microbial Contamination PRE AND POST PACKAGING.
Packaging Line Cleaning and Sanitation Validation
- Interior and exterior cleaning and sanitation procedures were validated through ATP Bioluminescence and found to be within the very tightest allowances for cleanliness and sanitation (again zero lumens).
- Packaging lines are cleaned and sanitized immediately before every use.
- The entire exterior of the packaging lines are foamed with a foaming chlorinated caustic cleaner, which simultaneously cleans the exterior of the line while the chlorine sanitizes the cleaned surfaces.
Yeast Health and Management
- We validated our oxygenation procedures via DO Meter to ensure proper oxygenation.
- We perform cell counts and adjust pitching rates for every batch of beer, and plot the ferm data out on a graph to look at the fermentation curve data
- We began performing forced wort fermentation samples to determine proper finishing gravity of Flagship beers. We found that the data collected from the forced fermentation samples lined up perfectly with the brew log data collected over the last year. This continues to be a vital QA check for us and ensures our yeast are happy, healthy and viable so they can fully ferment beers as well as they ought.
- We evaluate (via microscopy) that yeast is free of pediococcus and lactobacillus because their shapes make them easy to differentiate from conventional yeast strains. This method DOES NOT work for wild yeast strain detection.
Analysis of Re-fermenting Beers
- Microbial testing after refermentation - the tests revealed the same results as before the beer went to market, the beer was clean of the microbial contaminants we were testing for. (These tests were conducted inhouse AND third party.)
- Refermented beers exhibited finishing gravities up to 1Plato below final finishing gravity prior to packaging, even though in some cases beers sat for three weeks at terminal gravity @ fermentation temperature, and theoretically the beer should have finished fermenting.
- Every beer that went out to market underwent refermentation testing (not just ones that exhibited refermentation). Method and analysis are detailed below.
- Post packaging, beers sit 1-2 weeks (depending on the beer) in an 85F incubator to simulate the worst possible abuse beers can take out at market. After this beers are gravity tested for refermentation and for microbial contamination. The 85F temperature also falls within the ideal growth range of MOST COMMON spoilage organisms found in breweries.
- All beers tested in this fashion showed no signs of re fermentation or microbial contamination.
Continuous regular testing for refermentation of warm retains (setup to simulate at-market conditions) for the FIRST MONTH of the beers packaged life showed no significant signs of refermentation. - We have since learned that for this particular issue our protocols were insufficient, as the issue took between 2.5-3 months out at market JUST TO BEGIN TO MANIFEST
- Secondly, warm retains kept from the SAME BATCH that were affected displayed varying levels of refermentation with some showing none at all and some exploding. This contributed to the level of difficulty in noticing this problem with our testing regimens and procedures. We assumed that every can would be uniform in showing this issue AT THE SAME TIME, this was not the case, some cans, under the exact same conditions, appeared to referment much more quickly than others.
- pH testing on all of the refermented beers found that the pH was the same as it was when the product was packaged. This indicates that ACID PRODUCING BACTERIA WERE NOT THE CULPRIT.
Had acid producing bacteria been the culprit, even if acid could not be detected by taste, we would have seen a significant drop in the pH. - Taste panel on all batches of refermented beers revealed no significant taste difference other than standard flavor degradation due to warm storage as well as dryness and over carbonation when compared to the cold retains. No off flavors such as diacetyl, phenolics or acidity were present.
MOST spoilage organisms will show the above off flavors in the beer if they are present, providing an easy qualitative measure for determining source/cause of issue to a trained palate. - Filtration had no impact on whether a batch would referment or not, one batch that was recalled had been filtered, another one had not, but the same issue manifested in both. We use a fairly loose filter sheet that is designed to remove 95% of yeast in solution. This looser filter set allows us to remove the majority of yeast while still maintaining flavor profile pre and post filtration.
- One beer that had this issue used 1st generation yeast STRAIGHT FROM OUR SUPPLIER. This was our first red flag.
Comment