Because you can kind of see how it works, but there's a part which you can't see that changes the nature of the finished product over time. (That's the magic)
This article (?) is the product of my brainchild that I've been tinkering with over the last few years. In a different era, when I ran a forensics laboratory for a relatively unknown military unit, I had to recycle some really expensive (and fairly volatile) solvents. I ended up writing the SOP for the filtration and distillation unit, and it's become etched in my grey matter ever since. I mean, c'mon, just because I retired doesn't mean my brain shuts down, folks! So when the idea is gnawing at you...
As an amateur vintner, I had a 6-gallon batch of apple wine go bad a while back - within a couple weeks of bottling. I think it was cork taint, but the results were unsightly and definitely less than appetizing. I could dump the whole 6 gallons down the sink, or do something to salvage all that labor. I chose the latter approach, but most certainly had my work cut out for me.
NOTE: The standard disclaimers apply, all we can do legally in these United States is distill water and fragrances, anything else is verboten without a license and plenty of excise tax being paid.
Great. Now that I have that legalese out of the way, let's talk about a hypothetical way to refine a given solution's composition, shall we? Having that old USAF SOP in hand, with vivid photographic memories of the setup and operation, I proceeded to find assorted parts and pieces. Some were used, some new, some off-the-shelf, others kitbashed. I've actually got hose clamps, picture wire, and duct tape in there for good measure. As for sources, the local state university has a retail surplus store, and of course, there's always eBay. The system I've assembled is currently at Version 4.0, if only because I've learned over time to improve things, either by studying up on the process, or by simple trial and error. I'll illustrate the earlier versions in another blog post - the current system warrants discussion in the here and now. Ready? Here goes the illustration of the
Model 188 Batch Rectifier...
|
The Boiler |
To the untrained eye, the photo reveals what appears to be a power controller, an ancient laboratory stirring hotplate (850 watts, to be specific), and a 4 liter borosilicate boiling flask. There's some kind of sweet corn and barley derived liquid in there, too, along with a white teflon stirring magnet. This is the "boiler" section, and the stirring magnet is one of only two moving parts in this system. In a different blog installment, I'll discuss where that liquid began life, too.
Boiling a liquid mixture of mostly ethanol and water is kind of a running compromise, because as you boil out the more volatile component, you run the risk of sending the stuff you don't want out the top into your collection as the former dwindles in concentration. There are different peaks on the temperature curve where stuff comes out of solution, and you have to pay attention to the timing and characteristics to make proper "
cuts". This allows you to discard stuff you don't want, and keep the stuff you do. Learning how to make those cuts is a major part of making something worthwhile vs. something just plain awful.
Once the vapor you're looking for is heading northwards out of the boiler, you've got a couple of options. You can collect it and condense it right away, which is called "
pot still" mode. That guarantees more of the parent liquid's flavor will remain, but also makes for a much lower proof for a given run. To bump the proof higher, you have to re-distill the output of a previous run, and keep doing that until you reach the target proof. Pot stills are the favorite of the whiskey and bourbon industry, hence the "Double Distilled" and "Triple Distilled" labels on your favorite bottle of hooch. They're looking to retain the grain flavors in the finished product, be they corn, barley, or rye.
If you want to take things a step further, another option is to introduce some quality control to the process, by way of packing the column above the boiler and returning the majority of the vapor back to the boiler through that packing. This is called "
reflux still" mode, because it refluxes the vapors continually, letting the purified vapor escape out the top, while sending the impurities back down the column for another go at it. This provides a much higher proof (188 proof, or 94% ABV in the case of this particular setup), but also a more neutral spirit, very clean and with minimal residual flavor.
|
The Column |
How does this work? The packing in the distillation column is porous to a degree, and provides a huge amount of surface area for the ascending vapors to transfer their heat, condense, boil again, and generally interact with the condensed vapors coming from above as they head back down to the boiler below. (Google "
Theoretical Plates") In a glass Hempel column like the one above, that packing can be glass beads, Raschig rings, or a metal mesh of some type. For the metal mesh, the most popular materials are stainless steel and copper. Stainless steel is popular because it's inert, doesn't corrode, and lasts virtually forever. Copper usually wins because of the unique chemical capability of the sulfur-bearing compounds in the vapor to bind with the metal, making for a considerably better finished product. That's also why one sees so much copper in the construction of the
professional grade units found deep in the Appalachian hills. This particular column in the photo above is packed with pure copper kitchen cleaning pads - cheap, yet very efficient.
|
The Dephlegmator (aka, cold finger) |
The column packing does a good job of cleaning and purifying vapors, but works considerably better if there's a bit of a temperature difference between the bottom and top of the stack, with the cooler portion on top. This forces a certain percentage of the rising vapor to condense into liquid and drip down on the column packing, preferably traveling all the way back down into the boiler. This is where it gets ticklish, because if you send too much back down, you won't collect anything on the output. Likewise, if you allow too much out the top, you won't be in reflux mode - you'll be in pot still mode and your quality will suffer. The answer is actually pretty elegant - a "
Dephlegmator", or reflux condenser (It's also why I've graduated to Version 4.0 now). The dephlegmator can be tuned to introduce the precise amount of vapor cooling required to get a good reflux cycle going, without killing either output or letting everything just blow by. I used a laboratory cold finger condenser, with a needle valve for precise adjustment of cold water flow to regulate things. Done properly, you'll see vapor condensing and dripping from the cold finger's tip into the copper packing, and also a trickle of condensate heading back into the boiler. In the distilling world, this design is known as the "
CM", or "
Cooling Management" mode of operation. There's also "
VM" (
Vapor Management), and "
LM" (
Liquid Management), but I'll leave the reader to research those on their own nickel.
|
Reflux heading back down into boiler |
|
Unfortunately, as the run commences from start to finish, the ratios of ethanol vs. water and other volatile products change. This means you can't just start things and let it run to the end unattended, or you'll have something in the collection flask which you really aren't going to enjoy. So you carefully manage the flow of cold water to the dephlegmator upstairs, watching the drip rate of the condensate going back down the column, and looking for the rivulets seen above as they head into the boiler. By way of comparison, when running the dephlegmator at 100% reflux (everything going back down, no collection), it'll look something like this:
|
Sending it all back down the column! |
That's the other part of the equation - keeping the hot side hot, and the cold side cold. The good ol' boys in Tennessee usually park their contraptions not too far from a cold stream, and divert the water to their condensers. I use gallon milk jugs full of water, which reside in the freezer portion of my garage refrigerator. When completely frozen several of them fit nicely into a wooden patio cooler 1/2 filled with water, and when done I just dry them off and re-freeze them. Sometimes the best solutions are the simplest. With as small a setup as I've assembled here, I don't need more than about 4 gallon jugs of ice per given 4 liter run - well within the capabilities of my freezer to sustain. There's no mountain stream required, and I don't need to continually run cold tap water down the drain!
|
Ice-cold mountain stream, micro version... |
This cold water has to move from the cooler to the parts of the still that need cooling and back again. It doesn't have to be a fire hose, but it does have to extract heat and send it back to the cooler to get chilled by those frozen milk jugs. To keep things simple, I found a smallish Japanese koi pond pump, rated at 100 volts AC, that I simply rewired to 115 volts AC. The little impeller pump could care less about the 15 volt difference. It moves a LOT of ice water, doing a wonderful job at cooling things down and staying quiet in the process. It's activated by the "water" switch on the power controller.
|
How do you keep that ice water moving? |
Great - we have plenty of cold water, and have used some of it to increase the efficiency of the reflux column. What about the rest of the cold water and vapor path? This is where we get into the "cold side" of the system. Up to the dephlegmator, we were trying to manage heated vapor, with a given amount being allowed to condense and drip down only to be re-heated back in the boiler for the continual reflux purification cycle. Once the dephlegmator is dialed in via needle valve for an optimum reflux ratio, the purified portion of the hot vapor that's allowed to escape the top of the stack continues on as vapor through an elbow into a bigger cold water "product" condenser, as seen here:
|
Both condensers and their plumbing... |
Earlier iterations of this system used a Graham-type 300mm spiral condenser, which worked great but really took up a lot of space. The main condenser seen on the left here is a very compact version that has an inverted double-wall envelope "cup" that is directly in the path of the incoming vapor. As the hot vapor travels under the vapor pressure of the boiler, it impacts that condenser envelope, which has been chilled by ice water. It immediately condenses into liquid ethanol, which then rains down inside to be collected underneath. In the photo above, you can see the upper needle valve that controls the ice water flow to the dephlegmator/cold finger, and the return water line to the cooler. The lower green fitting on the product condenser is the cold water feed as controlled by another needle valve. The upper green fitting and tubing is the return water line to the cooler. Both feed water and return water lines are combined into their respective tee fittings to keep the total number of water lines to and from the cooler minimized. You'll see some stainless steel braided wires in these photos - they prevent the still from getting knocked over accidentally, and also let the top end of the still hang safely when I remove the boiler and column for cleaning and refilling after each run.
|
It's raining in there! |
Here's another view of the product condenser, showing the condensate dripping from the double-walled envelope inside. Note how the ice water flow has formed condensation on the tubing both entering and leaving the condenser - the little koi pump does such a great job maintaining high flow that even the water leaving the condenser is cold! It takes no time at all for the hot vapor at around 180 degrees to condense and rain into the collection flask underneath, at a temperature not much above that of the ice water's. Now, for collection's sake...
|
The whole system in operation, small footprint... |
Save for the water cooler and pump, this is the
entire system. Seriously, that's it! This is on the smaller end of a still, with the column only 1" in diameter. Couple that with a 4 Liter boiler, and you're looking at a whopping pint or so of output each run. That ain't much, but it's enough for the hobbyist. When in operation, the boiler is covered in water heater insulation to maintain better heating. There's a gooseneck CFL lamp for observing the whole thing during a run, and the collection flask sits high enough for a proofing parrot's beak to fit underneath when you're checking the proof of the output. The collection flask is valved at the bottom, but if you look closely enough, you'll see a pressure equalization tube on the near side running from top to bottom. That's very important - the vapor pressure from the boiler would pressurize and eventually explode the still if there wasn't a vent of sorts built into the system somewhere. Having the pressure equalization tube on the collection flask allows the liquid ethanol to be captured, while allowing the boiler to continue pushing hot expanding vapor through the system towards the product condenser. It pushes at a brisk pace, too.
|
One drop at a time. |
When running a distillation from fermented vegetation, it's always a good idea to discard a small portion of what boils off first. This is because methanol, or wood alcohol, boils at a lower temperature than ethanol. If your mash or wine was heavy on stems, seeds, pits, stones, or skins, then you'll have a higher percentage of methanol produced during fermentation. The first 25 ml of each run in my system gets discarded for safety's sake. When I say "discard", I mean it gets repurposed into my truck's windshield washer tank, or used to start burning whatever's in my fire pit, etc. JUST DON'T DRINK IT - EVER! There's more than enough to drink once you hit the ethanol peak in the distillation temperature curve, so do yourself a favor and make appropriate cuts. Once you're firmly in ethanol territory, you can keep the still's output. You can also measure the proof, understanding that it may in fact be too high to drink right away, so do be careful on the "sampling" thereof:
|
That's right - 188 proof! |
The Model 188 Batch Rectifier is very consistent, producing 188 proof distillate (94% ABV) every time. This is both good and bad. Good because, well, it's 188 proof ethanol. Bad because, well, it's also 188 proof ethanol. Granted, it's very pure, lacks any bad flavors, and extracts the most out of a given batch of mash. However, it must be cut way down to 110-120 proof (
cask strength) for aging, and if you're looking for corn, barley, or rye flavors, they just aren't there. Therein lies the rub - you have to rely on the cask to provide basically everything, as you've just distilled a very neutral spirit, aka white whiskey or moonshine. Neither IMHO are very palatable, so they need some help. Here's where you depart the world of laboratory science, and delve into the magic that is cask aging...
|
It takes patience, a lot of patience! |
For those who don't already know, charred or toasted white oak is wonderful stuff. When you construct a cask or barrel out of white oak, then fire it inside to the right color and depth, you've got a perfect receptacle to age your new cask-strength young whiskey within. The solvent qualities of the young whiskey will immediately start extracting color and flavor from the oak, and over time you'll see and taste the transformation. A longer stay in that cask is
usually better, but not always - you can actually over-oak a whiskey if you're not careful. It's best to rotate the cask a bit each month or so for even breathing of the whiskey into and out of the wood's pores, and of course don't forget to sample now and again. The whiskey will extract a nice caramel color, as well as flavors like vanilla, chocolate, pepper, etc. This all happens sight unseen, and benefits from both the cask rotation and the temperature changes that occur each day and night. You will lose a small fraction of the whiskey to evaporation due to the open pores of the oak staves, but this is known as the "
Angel's Share", and is considered paid tribute for the wonderful results later. I use smaller casks with a different surface-to-volume ratio, and they mature whiskey a bit sooner than their bigger cousins. After about one year, I temper the whiskey down to 80 proof with charcoal-filtered rain water (No flourine, chlorine, or dissolved minerals). It's a lot of work, but once bottled, the results to me are well worth the effort.
|
How do you do? |
I wouldn't even want to hazard a guess at what the cost of labor, materials, and equipment is to produce even one pint of finished whiskey. Truthfully, I don't care. I made it myself, and in a recent tasting amongst friends, it's been likened to some of the finer Lowland Scotch Whiskies out there on the market. That says more than enough to me - I must've done something right.