Monday, January 30, 2017

LODO Festbier: Split Batch Experiment

Running off the HighDO wort.My first attempt at low dissolved oxygen (LODO) brewing was lackluster... generously (I dumped the last gallon of that Pilsner). I wanted to try the intensive process again, replicating both wort production and fermentation as close as I could to what is prescribed in the second version of On Brewing Bavarian Helles. Rather than brew ten gallons of LODO festbier, I split the batch pre-boil aerating half of the wort as a control. I boiled this "HighDO" wort harder and with a copper immersion chiller. The LODO half I gently simmered and then chilled through a stainless-steel Blichmann plate chiller (although it was brazed with copper). Post-chilling I treated the beers identically from cold fermentation through spunding.

At the start of the January meeting I roped 20 members of DC Homebrewers into a blind triangle tasting of these two beers (poured from growlers I had counter-pressure filled 90 minutes prior). Only 7 of 20 (P=.52) correctly selected the odd (LODO) beer out of the three samples. That is a number perfectly consistent with random chance, suggesting that my LODO and HighDO beers were indistinguishable to the average beer nerd. Of the seven who correctly identified the aberrant sample, only one preferred it (four preferred the HighDO, and two had no preference).

After a month of drinking the two beers, I was able to select and identify the beers in my single attempted triangle test. They are similar, but the LODO does have an ever-so-slightly maltier aroma to my nose. Flavors are nearly identical.

As a disclaimer, I intended this test to explore whether my kluged-process LODO made an incremental improvement to this pale lager. I'll say "yes" if you know what to look for, but in the barest of terms. What I’d love to see is someone with a dialed-in system try the same experiment!

A single experiment can’t prove or disprove anything. That's why replication is an essential (if unsexy) part of science. Even under rigorously controlled conditions statistics like this only provide a confidence interval that suggests that the results are not due to chance. Compound this with variability introduced by brewer, brew house, tasters, conditions etc. and you sometimes produce false positive and negatives. That said, blind taste tests are the best way to insulate results from expectation and bias. Triangle tests are a pain to conduct, and put a target on you from people who can swear they can taste the difference. I have a lot of respect for what the Brulosophy folks put themselves through for data (especially after participating a couple times)! Looking forward to hanging out with Marshall in New Zealand in a couple months between talks at NZHC 2017!

Tube ringer for White Labs.As a side note, Audrey got me this tube wringer for toothpaste, but it works perfectly to extract the last few billion cells from PurePitch packages. White Labs should probably license it and sell an official version!

LODO Festbier

Smell – Clean bready malt aroma. Pleasant waft of sulfur, although a few tasters felt it considerably stronger than I do. Faint grassiness of noble hops.

Appearance – Slightly-hazy deep yellow. Dense white head sticks around until the bottom leaving patches of lacing down the sides of the glass.

Taste – Malt flavor is well rounded. Crisp, but the 5% crystal malt adds a mild honey-like sweetness. Pleasant herbal hop flavor in the finish. Clean balancing bitterness. Retro-nasal brings the appropriate lager-light-egginess back. It had a flavor that reminded me of the doughiness of a no-boil Berliner weisse when it was young, but that has faded.

Mouthfeel – Medium body with moderately prickly carbonation.

Drinkability & Notes – One of the better lagers I’ve brewed. Malty without being heavy. The sulfur is at the top of the my range, likely a result of the cold fermentation and spunding rather than the sulfite.

Changes for Next Time – Unlike my first attempt where the primary issue was double-dosing metabisulfite, this is a pleasant beer! Next time I’d reduce or eliminate the Carahell, and save the effort and brew it with the standard wort production (and warm up the fermentation towards the end)!

The LODO FestbierFestbier Recipe

Batch Size: 10.00 gal
SRM: 5.3 SRM
IBU: 17.5 IBUs
OG: 1.055
FG: 1.013
ABV: 5.5%
Brewhouse Efficiency: 65%
Boil Time: 65 Mins

Grain
-------
75.5 % - 17.1 lbs Weyermann Pilsner
17.5 % - 4 lbs Weyermann Vienna
5.4 % - 1.2 lbs Weyermann Carahell
1.7 % - .4 lbs Weyermann Acidulated

Mash
-------
Sacch Rest - 30 min @ 152F

Hops
-------
4.40 oz Hallertauer Mittelfrueh (Pellets, 2.00 % AA) @ 60 min
1.60 oz Hallertauer Mittelfrueh (Pellets, 2.00 % AA) @ 10 min

Yeast
-------
White Labs WLP833 German Bock Lager

Notes
-------
Brewed 11/14/16

Recipe above is the ingredients for the entire 10 gallon batch.

Made a 5L stir-plate starter. Fermented at room temperature for 36 hours, then crash chilled.

Boiled 18 gallons of water (8 distilled, 10 filtered DC) added 12 g of CaCl. Preboiled water, then added 7 crushed sodium campden tablets. Underlet mash after purging with CO2.

Mash pH 5.28.

Collected 7.5 gallons of wort as is, 4.5 aerated and left in an aluminum pot until the remainder came to a boil.

Adjusted 2.4% AA hop pellets down to 2%. Bagged.

LODO, 2.75 oz @ 60 min. 1 oz @ 10 min. Plate chiller. 1.056. Slightly sweeter, maybe could pass for maltier. A shade lighter.

Aerated, 1.65 oz @ 60 min. .6 oz @ 10 min. Immersion chiller. 1.054.

Chilled both to 46F, shook to aerate, pitched the decanted starter. Left at 48F to ferment.

11/20/16 Down to 1.032 (43% AA).

11/21/16 Started dropping 1F per day. Until it reached 43F.

11/24/16 1.024, still pretty yeasty with a small krausen.

11/27/16 1.019 (66% AA). Kegged into quadruple-purged kegs. Purged and pressurized head space. Left at 48F to carbonate. Both have some sweetness, so hopefully drops below 1.015 (73% AA).

12/3/16 Good pressure on both, 15 PSI. Removed spunding valve and began dropping 2F per day.

12/18/16 Attached to gas and dumped yeast from both. FG 1.013 (76.8% attenuation) on the LODO, 1.012 (77.8%) on the HighDO.

I get a commission if you click the links to MoreBeer/Amazon and buy something!

10 comments:

Techbrau said...

Just to be perfectly clear, you didn't test LoDO vs HiDO. You brewed a single LoDO beer, and split it. We corresponded over email, and I specifically told you not to do it this way for the following reasons:

1) The "HiDO" beer still had significant sulfites from the mash carry over through the boil, which would have gone a long way to protect it against any intentional aeration.

2) The mash is where the worst oxidation happens because the oxygen solubility is highest on the hot side at this point, and the oxidizing enzymes are most active in the mash because they are deactivated in the boil

3) The "HiDO" beer was still given the LoDO treatment throughout the cold side (i.e. spunding), which isn't reflective of what the average homebrewer does.

Please don't call this LoDO vs HiDO, because it's misleading.

GWA said...

I think that by the nature of the low DO beers (and brewers) it will be assumed that there was an experimental error if the results show that there is no statistical significance of low DO brewing. There are too many opportunities for potential oxygen ingress to appease everyone that everything was done in a low DO setting.

It would be informative to have Techbrau (or ideally multiple recognized low DO brewers) and Michael or the Brulosophy crew to brew up the same beer at the same time, but under normal DO conditions (i.e., not high DO). Perhaps it could be coordinated with a national homebrew conference. There will obviously be a number of variables in addition to DO involved (e.g., equipment, process), but if there is no statistical difference between the beers then we can cross off low DO as a variable of concern (and possibly several other variables). If the beers are statistically different, then more experimentation might be needed to isolate the variable(s) responsible.

The Mad Fermentationist (Mike) said...

There was no perfect solution, but I went with a split batch for two reasons:
My system is 20 gallon pots, so minimal empty space in the mash-tun for a no-sparge 10 gallon batch.

1. Water profile. I could try to mimic the sodium and sulfite contributed by the campden with salt and gypsum, reducing the CaCl, but it seemed confounding.

2. If it wasn’t clear, I aerated the wort pre-boil when the enzymes were still active (no mash-out). If running the wort from the mash into the kettle from a few feet up and then splashing for five minutes with a mash paddle wasn’t enough to cause damage with the sulfites present, it doesn’t seem like all the other precautions are necessary.

This test was solely testing the hot-side process. I don’t think there is any argument that pristine cold-side process is beneficial. If I had done my standard process vs. the hot and cold-side process I wouldn’t have been able to make any conclusion about what the significant points were.

GWA, likely too many confounding variables having the beers brewed by different brewers.

Bryce said...

Wow, TechBrau feels pretty strongly about this. RDWHAHB.
Anyway, nice job Mike as usual. I learned a lot. : )

Anonymous said...

My understanding is that the LODO process is meant to emulate flavors that only some of the largest German lager producers--Spaten, Pailaner, Weihenstephaner--are able to achieve, since even smaller German brewers don't have the equipment to brew beer under such stringently oxygen-free conditions. There's something ironic about homebrewers trying so hard to perfectly emulate macro-brewed lagers--even Bavarian ones. It's like an amateur chef investing lots of time and money in an effort to make the perfect Big Mac.

The Mad Fermentationist (Mike) said...

I get your drift, but I wouldn't compare Weihenstephaner to McDonald's. Maybe more like trying to mimic the character of Heinz ketchup at home? I look at any process or technique as a tool. I try them out and figure out if it has a place in my arsenal, either for the originally intended beer or another.

Interestingly McDonald's posted a video of their Executive Chef making a Big Mac at home. I guess safe in the knowledge most of their customers won't spend their time and money for a homemade version (but might appreciate that the ingredients are "real"). Sort of like breweries that are kind enough to post homebrew recipes.

Anonymous said...

You're right: I should've used Bitburger or Warsteiner in my analogy, especially since one of the most vocal (and defensive) online LODO proponents often holds them up as paragons of that Teutonic "it" flavor. If I desire that particular flavor myself, I find it easier simply to reach down to the bottom shelf when I'm shopping at Trader Joe's.

Anonymous said...

Haha... the LODO myth. What a joke!

Unknown said...

Came across this since I'm planning to brew a Festbier/Oktoberfest soon and noticed something that surprised me ....

Do you typically use so many campden tablets (i.e. 7 for a 10G batch in this case)?!? Why?

The Mad Fermentationist (Mike) said...

It is intended to mimic the ultra-low oxygen exposure that large breweries can achieve with nitrogen or steam purging, closed transfers, and other equipment and techniques outside the purview of homebrewing. I'd suggest reading through my previous attempt a "LODO" and the original paper that sparked my interest (both linked from the first paragraph).