The idiosyncrasies of a nano-scale craft distillery

While my usual daytime work revolves around liquor retail, for the last few months my schedule has squeezed in something else a little more interesting - a bit of craft distilling here in Australia.

While I’ve worked in production at a couple of breweries already, I’ve never had the chance to take that delicious base that is unhopped beer and turn it into new-make spirit.

That was until a few months ago when a good friend who’s currently building his own brewery/distillery gave me the opportunity to come along and play with his shiny new toys. We’ve been collaborating to produce some pretty diverse liquids, ranging from fairly typical things in their categories to peculiar juices bordering on the outright whacky. I won’t give too much away at this point, but there are a few things the SWA would take issue with if we were trying to do them in Scotland.

Because the friend I’m experimenting with makes a bit of everything, the approaches we take tend to be lateral and holistic. The beers are largely mixed-culture sours in the vein of old world styles like lambics and gueuzes, oud bruins, Flemish reds and saisons, though there are a few playful new-world sours in the mix too. The new brewery also makes everything from light European styles like decoction pilsners and goses to incredibly viscous, heavy styles such as barleywines and barrel-aged imperial stouts.

The wines he makes - in collaboration with a local producer who usually sources the fruit and lends equipment - are a blend of old-world stylings while embracing some of the more modern minimal intervention methods, accentuating some of the fruit-and-funk interplay that many are now dubbing as natural wines.

The spirits umbrella will encapsulate many projects; gin and vodka recipes are already in production with more varieties on their way, meanwhile there are active experiments in whisky, rum and brandy with potential for agave to be added down the track.

There are also various experimental projects for bitters, aperitifs/digestifs and other similar oddities under way, all of which will borrow elements of production and flavour from other streams. One of the overarching themes that connects all of these categories and substrands of drinks is an idea my friend passionately upholds; “Everything relates to everything else”

For instance, as brewers, we consider some of the microbiological research and methodology that’s come out of the wine industry from bodies such as the Australian Wine Research Institution (AWRI). Likewise when working with grapes, rather than treating brettanomyces as an inherent issue, using mixed culture fermentation data available from the brewing realms helps to guide process implementations which shape its role in the final wine, bringing out positive attributes while controlling or eliminating the possible negative ones.

Even raw ingredient use is considered; the waste streams from some products can be recycled and transformed by process into an ingredient for a future product in another category. As a completely contrived hypothetical, botanicals used in some gin stage might later be recycled in a bitters project, and a fraction of the bitters might be blended into an upcoming beer, some of which might eventually be distilled into an essence used for blending into a new aperitif. The possibilities are almost endless.

The learning that’s come about from all this has been considerable - odd little things most people don’t expect one would ever think about. For instance, which types of receptacles are better for use in short-term spirit storage as a function of their lid’s seal material. Some can leach plastic/rubbery flavours into the spirit just from interaction with the vapours in the headspace. Other considerations such as how to control wash distillation foaming with the measured application of an agitator. Then even stranger and more niche things like the difference in various qualities between coriander seeds with a spherical vs oblate spheroid shape (ie Aussie rules football or rugby ball) .

Possibly there’s been as much learning coming from the actual building of the brewery and distillery as there has been from the literature reviews, mad experimentation and resultant data collection. Things like how to jerry rig seemingly unrelated fittings together that have no earthly right being connected, or how to use a forklift, wooden pallet, straps and tensioners to turn an audio system into an impressively improvised worker’s lounge on top of a refrigerated storage room. Yes, it’s load rated- we checked.

Anyway, this is a whisky blog so let’s talk about the distillery and the whiskies to come.

Raw ingredients

While the majority of whisky production is likely to be focused on malted barley, we are open to making a bit of everything. Corn, wheat, rye, oats, spelt and whatever else catches our fancy. Being a brewery, one of the advantages is many of these ingredients are already going into various beer recipes anyway, so they’re largely on hand to begin with. What’s more, there are certain beer styles which lend themselves particularly to being made into a whisky equivalent. For instance, the new-make I’ll be reviewing below was from an imperial stout mash bill. Other beer styles utilising stewed, kilned and/or roasted malts will feature as mash bill templates in future distillations.

The mill

The design of our milling system is fairly simple, but it’s an elegant use of space that suits the capacity of production we’re looking at in a fairly condensed workspace. The mill is below shoulder height and can accommodate manual loading pretty well. That said, having a forklift available, it tends to make more sense for us to fill the grain into a bag or some other vessel which we can lift using the forklift prongs to sit above the mill.

The grain is then fed into the mill and turned into grist, the composition of which is largely determined by the mill gap. This is the adjustable distance between the mill’s rollers (in our case a single pair, though the larger scale ones used for greater production volumes in brewing tend to have two or three pairs) which determines how finely the grain will be milled into grist. Selecting an appropriate gap size is thus important for several reasons; it determines the grist’s ratios of flour, grits and husks, as well as the coarseness of said grits and the integrity of the husks.

Ideally, the distiller will usually want the husks to be kept whole, simply split open to allow the interior of the grain to be ground into grits and flour. This is so that when it comes time for mashing, the husks can provide a better, more porous filter medium for the wort during recirculation, or in brewer’s terms; vorlauf. It is this recirculation which will dominantly determine the wort clarity before fermentation; longer/more effective vorlauf through a healthy filter bed generally leads to a higher clarity wort.

Having a poor filter bed where too many of the husks have been shredded into fragments can lead to a filter bed which restricts wort flow and can lead to a stuck mash. It’s important to ensure a sufficiently fine milling too, such that starch granules are fine enough to be properly gelatinised, dissolved and acted upon by enzymes for saccharification, beta-glucan hydrolysis and proteolysis. Typically, most malt distilleries look for about a 7:2:1 grits:flour:husk ratio, though it depends on individual process and desired outcome too.

Auger and grist casing

Once the malt has been milled into grist, it’s fed into an auger which utilises interspersed plastic discs attached to an internal pulley loop to collect discrete parcels of grist, then ferry them up to our grist casing. The advantage of this auger system is that it’s gentle on the grist; many conventional screw augers generate significant friction on the grit and husk portions of the grist, shredding some of the husks and increasing the flour fraction.

It also has the advantage of being able to move around bends and corners, which makes it more suitable for our space than a conventional auger, in which case we’d have to find space for the diagonally ascending shaft to transport the grist directly from the mill to the grist casing.

The grist casing is basically just an intermediate storage vessel for the grist which helps to regulate the pace of mashing. The grain-in speed is altered by adjusting how far open the grate is.

The mash tun

The mash tun is a fairly standard semi-lauter design; it’s a combined mash-lauter vessel, meaning that the grain is both mixed and recirculated/run off in the same vessel. This is pretty standard fare for most small craft breweries running a three vessel (hot liquor tank, mash/lauter tun and kettle) kit these days, with possibly the only bit of extra excitement built in being the mash hydrator.

This is essentially just a bit of extra plumbing that allows for water to be sprayed into the grist chute directly before entering the mash tun, thus increasing the surface area of hydration compared to simply dumping the grain into the water already in the mash tun. This decreases the likelihood of grist forming anhydrous lumps known as dough balls in the mash. Dough balls decrease mash extract efficiency by reducing the total amount of grain sugars extracted into the wort.

Fermentation

This is one of the areas where vagueness is somewhat unavoidable. Given that my friend makes so many beverages, we have a scary number of yeasts, bacteria and other oddities in the fridges at our disposal; many various saccharomyces strains, brettanomyces, a handful of lactobacillus cultures, pediococcus, aceto/glucano-bacters, koji (yes, rice will be on the cards) and a handful of other things which are in development.

Because the distillery is built parallel to the brewery, we also have the opportunity to sanitise wash production by boiling it in the same way one would make beer, meaning that only the cultures we deliberately add to the fermenters would have a significant role in fermentation, if so desired.

On top of all this, we have what’s known in Belgium as a koelschip; a fairly shallow but wide and long vessel used for spontaneous fermentation by natural inoculation. Hot wort is pumped into the koelschip to cool down fairly rapidly via convection, after which the wild microflora of the environment have a large surface area of wort to settle on and begin their spontaneous fermentations before transfer to a dedicated fermentation vessel. This gives us the potential to harvest wild microflora in a sense, and wash made using this method might be blended with other washes made using more controllable methods to add particular flavours in a desired ratio.

For some of our upcoming rum projects, we also have a muck pit which is fed intermittently with bits and pieces to maintain biodiversity.

What I’m saying is we have almost too many options when it comes to fermentation, and that compared basically all of the traditional distilleries we have an incredible degree of control. I could probably spend days describing all of the other methods and control parameters we can vary to tweak our fermentations, but some of it might be IP and it almost certainly goes way beyond what the majority of you want to be reading about. Suffice to say, we have options.

Distillation

I won’t bore you all with the minutiae of how we run our distillations for two reasons; one, you’ve not long ago endured a deep-dive on simple pot distillation mechanics from me already. Two, we don’t have a standard approach to distilling yet, especially given the wide variety of distillates we’re hoping to create. In lieu of those details then, I’ll surmise the important bits of the kit; the copper pot still is heated by an external jacket powered by three elements, and depending on desired functionality the jacket can be filled with either industrial grade oil (operating pressures and temperatures up to about 400 degrees Celsius) or water (operating temperatures/pressures up to about 140 degrees Celsius).

For the moment we operate the still with water in the jacket, as this gives us a much more gentle heating allowing for better fractionation over the course of distillation. The still has a charge volume of roughly 500L, a neck featuring a pretty hefty reflux bulb of sorts and an agitator which is used both to control foaming and to keep lees from settling/burning on the bottom of the still. The lyne arm is horizontal, feeding into a series of plumbing and valves which allows for the vapour path to include or bypass a fairly small column (five plates, each of which are also bypassable) and a botanical infusion basket for gin and other similar spirits. The advantage of this system is, much like the rest of the brewery and distillery’s designs, maximal flexibility while maintaining a small setup.

The condenser is a standard shell and tube design built for efficiency, but thanks to the easy variability of input counterflow water, the fairly short path length and the relatively slow rate of distillation, we can alter the temperature of condensation pretty drastically, maintaining the condensed spirit’s temperature while observing output water temperatures between about 40 and 75 degrees Celsius. In other words, we have the option of cooling the condenser much more quickly if we want to decrease the condenser’s copper interactions for a heavier, more sulfurous spirit.

Right, enough of that, shall we talk about some new-make?

New-make

When my friend made his last batch of imperial stout, the goal was for the beer to be as heavy, viscous and voluptuous as possible. To achieve this, he used a massive grist:liquor ratio (ie lots of malt, not much water) to extract an incredibly high concentration of wort sugars. After sending this wort to the brewhouse’s kettle for boiling (more than four hours of boiling at that; about three times the industry standard) rather than leave the grain in the mash tun with a high amount of unextracted sugars, he mashed the grains in a second time with another batch of hot liquor.

This process is ubiquitous in Scotland too, where three consecutive batches of liquor are used to extract a maximal degree of sugars from each mash, but in that instance all three waters/worts are obviously used for whisky rather than beer. Anyway, I digress; this second run and the following spargings of wort were then extracted into a pair of IBCs and set aside to cool down for later yeast pitching. As an early prototype batch, M1 distiller’s yeast was pitched for primary fermentation to ensure a high attenuation of sugars for good yield, as well as a typical broad spectrum of congener production.

This was likely aided by warmer free-rise fermentation temperatures compared to the Scottish average thanks to the South Australian climate. To give added interest, after primary fermentation the IBCs were left in a dark, quiet corner of the shed for approximately eight months to undergo secondary fermentation with whichever bacteria and other heat-resistant bugs had survived mashing.

The results were intriguing; smelling and tasting the wash before distillation, it was apparent that some lactic acid bacteria had soured the wash considerably, and indeed there was some low level funkiness that had probably arisen from protein catabolism or some other late stage metabolic pathways. The main feature though was the malt bill; the high proportions of stewed, kilned and roasted malts had brought an almost Reece’s Peanut Butter Cup tone- sans the butyric acid that ruins almost all American chocolate though. As a starting point, we were quite happy.

Distillation was pretty straightforward; the heating elements were set to keep the water in the jacketing at around 110 degrees Celsius, which gave a total hearts run time of about two hours. Hearts were kept pretty conservative on the heads side (cut in at about 73%) however tails were run to a slightly lower ABV than usual- around 60% to capture more of the roasty mocha tones with lower volatility, as well as adding some oiliness and texture.

Cut points were decided principally based on organoleptics, although viscosity measures (ie rubbing distillate between our hands to check for relative degree of oily slickness) and demisting tests were used as rough guides toward late hearts and throughout tails. Towards the end of hearts we began to run into smaller collection containers to make a set of what were essentially micro-cuts.

After distillation, we went back through the micro-cuts and made sample blends with the bulk of the hearts to determine how deep/low ABV we were happy to go, with the aim to maximise the roasty mocha/peanut notes while occluding the undesirable tails characters. This is how we arrived at the tails cut at around 60% ABV.

Nose

Given the high strength (and the fact that I’d mucked around with dilution for misting tests during production already) I took this down to about 50% ABV as a starting point for nosing.

Typical new-make volatility including acetic acid and ethyl acetate plus various white orchard fruits; all acceptable characteristics which will help contribute more diverse esters later in maturation. Big espresso heading toward mocha, Reece’s Pieces Peanut Butter Cups (sans any weird American chocolate butyric acid nonsense) and moderately toasted malt loaf.

There is just a touch of tailsy character (think cheddar cheese with a tinge of sweaty socks) but this will likely age out with time in oak. The same is true of essentially all the high quality peated new makes; to get the less volatile phenols, you have to go deeper for the tails cut. As the saying goes; if the new-make’s too clean, the whisky will be no good.

Palate

For the palate I started neat to get a sense of the initial texture, then diluted straight down to about 50% ABV.

Neat, all the same elements as from the nose, but with an emphasis on the Peanut Butter Cups and mocha, plus a little of the perceived sweetness one finds in most new makes; not sugar obviously, just the effect of ethanol and congeners. The texture here is delightful; incredibly coating, very silky with a nice viscosity.

Dilution brings out some tart/unripened peaches, more of the roasty elements and more of the malt, as well as a touch of the cheesiness found on the nose.

The Dregs

Obviously there’s just no way that I can be objective about this distillate, even having had only a pretty small role to play in its creation. That said, I like where it’s at, and with some time in cask (we’re thinking some relatively active American oak, maybe refill 100L ex-bourbon casks) this could go in some rather cool directions.

For those of you familiar with the Glenmorangie Signet, or even some of the Westland whiskies made with roasted malt, there are parallels here derived from the mash bill. In our opinion though, the ridiculously long time on lees in the IBC coupled with tertiary bacterial fermentation has given some components to the new-make that ought to evolve beautifully. Whether this reaches the market as a single cask offering down the line or is used as a blending component remains to be seen; it’s almost certainly a decision that will be made after the cessation of my involvement with production. In any case, I guarantee this is only the beginning of some utterly mad experimentation.

Watch this space.

TK

Dramface is free.
Its fierce independence and community-focused content is funded by that same community. We don’t do ads, sponsorships or paid-for content. If you like what we do you can support us by becoming a Dramface member for the price of a magazine.
However, if you’ve found a particular article valuable, you also have the option to make a direct donation to the writer, here: buy me a dram - you’d make their day. Thank you.
For more on Dramface and our funding read our about page here.

A Day In The Life Of A Distiller