Ian Smiley's Whiskey book and fractionating stills
Posted: Sun Apr 26, 2015 8:33 am
Ian Smiley's book "Making Pure Corn Whiskey" 2nd Ed is a good book, but it has some stuff that is confusing and outdated. But if you are a newbie, it can send you down some strange alleys. The purpose of this thread is to discuss some of the more contentious and inconsistent things, so that people can gain the value while avoiding the pitfalls.
It would be good if someday Ian updates his book, with some ideas from this forum, this thread and the comments I'm hoping for from some of the other masters. His book could then be truly excellent. (A friend, at his first day of a Med degree was told "half of what we are going to teach you is wrong. Unfortunately we don't know which half.")
I'm going to start with two simple points and move on to the most interesting and important chapter.
Leaving your still unattended
Ian mentions that you can do this under certain circumstances. This he probably took from Nixon's book which also discusses this. I, and others in these pages, think this is incredibly stupid. Suppose the (cooling) water stops due to municipal repairs? How will you know if you are out shopping? Where will that flammable alcohol vapour go? Where does your home insurance sit relative to an unattended still?
Is the flavour in the heads or the tails?
At different places he makes different statements. Both contribute, the most important is the tails (I'd much prefer a whiskey without heads to one without tails.) See http://homedistiller.org/forum/viewtopi ... y#p7278767
The smell of the heads is usually dominated by ethyl acetate. It has a nail polish smell. Many people confuse this with acetone, which is similar. Nail polish can contain both acetone and ethyl acetate. Ethyl acetate is produced when certain bacteria combine oxygen and alcohol to produce acetic acid, which then reacts with the ethyl alcohol to become ethyl acetate (so ferment with an airlock and distil straight after fermentation eases to minimise ethyl acetate.)
Ethyl acetate can be detected in many bourbons (e.g. Makers Mark), whiskeys & rums. It is the primary smell component of sugar wash spirits that are purified with carbon - as carbon does not remove it, though most other things get taken out. However a good fractionating column will separate it, even though its boiling point differs from the ethyl alcohol + water azeotrope by just 1 degree C.
Tails have an earthy taste. Although they can be unpleasant, this is often because they are separated from the overall spirit and are out of balance. When drinking a good whiskey (say Laphraoig), I wonder how bitter was the spirit when the distiller ended the tails cut. Ian claims that for more full-bodied spirits the distilleries will end the tails cut quite late, and that makes sense to me.
Using a fractionating column to make whiskey
This gets a whole chapter of this book, and he only describes one way of doing it. Smiley (correctly) observes that you can use a batch fractionating column to make good whiskey, but also tries to stick within the legal definition of whiskey - a spirit collected with an alcohol content not exceeding 80%. But the commercial distilleries use continuous reflux stills or pot (/alembic) stills so this mix of still type and law doesn't make sense to me. Many people making whiskey at home are breaking a much bigger law than this one!
And there are some other very interesting things you can do with a variable reflux fractionating still. A fractionating still sorts liquids by boiling point. In an ideal fractionating still, the temperature would rise in steps and the different components of the wash would come off in boiling point order: acetone 56, methyl alcohol 64.7, ethyl acetate 77.1, ethyl alcohol 78.4, propyl alcohol 97. In practice, compounds azeotrope and because the still is not perfect, get smeared together. But in a column with an infinite number of plates and with an infinite reflux ratio, you would get a stepped temperature and corresponding compounds in the output. And by changing the collection vessel each time the temperature steps, you would sort the components of your wash. With a pot still, the distinction of heads/heart/tails and remaining low wines is a sorting into boiling point categories, but with the components smeared across the collection vessels.
A fractionating column with adjustable variable reflux ratio is a truly strange beast. With the reflux set to 0 (no recirculation) and assuming it is well insulated it is like a pot still. With a little bit of reflux it is more like an alembic still. And with high reflux it is only itself. And you can change this through the run.
So what Ian does is to run his still at a reflux ratio of 2 or so (though he is not fussy about that) but then starts collecting at 80% alcohol. The same wash in a pot still might start out below 80% so it would go into the final whiskey. In a pot still in which there is no fractionation, low wines below 55% will not produce a distillate exceeding 80%. But with a fractionating columnn the product will start with a much higher alcohol concentration, so a lot of alcohol is discarded. And then, as he has compressed his distillation - though while still trying to use the accepted percentages for the tails cuts - he has a tight window as he works out where the end point is. He worries about this a lot, and even has the boiling point changing while filling a hydrometer! So I bet he does not get much whiskey out of his runs, but I'd suspect it would taste pretty good. (At least thats my reading after sitting quietly on a plane for too long and trying to untangle that chapter. If anyone reads it differently, I'd be really interested to hear.)
So here are some other techniques you can do when using a fractionating still to make whiskey. They can be combined or used individually. First, lets ask the question "what is wrong with a pot still or an alembic still for making whiskey?" Not much really. All good scotches are made with a pot/alembic still. But there are several things that can be improved upon with variable reflux fractionating column.
If your wash has too much ethyl acetate for your taste (a common case in home made whiskey), with a pot still you have to make a late switch from foreshots/heads to collected whiskey. You then have some orphan spirits that are smelly and need to be thrown out or get additional treatment. Carbon won't make it taste better.
With a fractionating column, you can start the run on complete reflux, let it stabilize, then set it for a high reflux ratio. The foreshots (really early heads) - methyl acohol, ethyl acetate and acetone - can be removed without much loss of ethyl alcohol. When the distillation is through the this stage adjust the still to no reflux and it will then act like a pot still for the next stage of the run. If you decide you want a bit of ethyl acetate (some people like it) you can blend a touch back in later. This way you have complete control, while not losing much alcohol with what you discard. You can also split your heads into several different stages which will have different tastes which you can blend at your leisure.
The second technique that can used to improve things over a standard pot still is around effective use of tails. I hate to throw tails out - they have most of the flavour of the whiskey. You can collect them up and distill them on their own, but if you are like me and use corn one batch, rye the next, barley the next, maybe some grappa etc, you end up with a very mixed collection. What I'd prefer do is to add the tails to the parent spirit in a way that enhances it.
So here is the recipe: run your fractionating column as a pot still until it reaches the endpoint - 60% or whatever you like. Then you have made a normal pot still whiskey. Then switch collection containers and adjust things to push up the reflux ratio. The alcohol content of the distillate will go up. This can be repeated several times until the still is on a very high reflux ratio and the flow is very slow. Then turn the still off.
The taste the contents of those late collection vessels will be terrible. But don't despair and don't throw them out! The reason they taste terrible is because they are out of balance, not because they are bad. A bit like eating a bird's eye chilli - on its own it is inedible, but in a curry with all those other things it enhances the flavour. It can be scary adding really vile stuff to your nice clean hearts, but I have been totally amazed at the transformation.
So what you have to do is find out whether the tails enhance or detract from your whiskey. So take some of the main cut, and measure a proportional content from the tails cuts. This can be done with a small syringe - 1ml or 5ml or whatever. e.g. suppose you have 3l of main cut, and 300ml of tails for your test you will need 25ml of main cut and 2.5ml of tails. Mix them up and taste it. If you don't like it, throw it out or add it to the main cut. The tiny bit of tails will get lost. Or maybe make a shot with 1.25ml and one with 2.5ml and compare. How much tails you add depends on what you plan to do with your whiskey and how you like it. The whiskey will soften with age, so if you want it to be ballsy when it is 10, like many great whiskeys, it will be very intense when it is new. Adding tails is a risk or an investment depending on your point of view.
Another thing you can do with your fractionating column is to concentrate your whiskey. Suppose you have a whiskey recipe you like, but you want to make one that is just more intense. So make your normal wash, distill it with a fractionating column but with no reflux i.e. pot still mode. A quarter way through the run, set the reflux ratio to infinite and let the still equilibrate. Switch containers. Then set the still for a reflux ratio of 5 to 10. Then take a volume equal to (say) 1/2 of the entire batch. This will be neutral, or very close depending on the length of the column. Then switch containers again and open the needle valve right up so the still goes back to being a pot still. Complete the run as normal.
By blending the first and third containers you will have a cogener profile the same as your normal whiskey, but with half the ethyl alcohol. Maybe you don't like it. Then blend back the neutral spirits.
So the idea that a fractionating column strips all the flavour out of the whiskey is misleading - nothing is stopping you from adding it back in.
And I'm sure there are lots of other things you can do. A fractionating column with a variable reflux ratio is a very flexible device.
How critical are heads and tails cuts?
A lot of books (including Smiley) talk about the "right" place to make the cuts, but I don't believe it is that simple. Some people like light whiskeys, some like whiskeys that cause your chest hairs to drop out. In scotches there is Glenlivet and Laphroaig. Light whiskeys end the tails cut early, big ones go late. There is no "right" point. So how does one decide? Smiley talks about putting all the whiskey in one container straight off and redistilling if its wrong. What a waste of time! Distil into separate jars. Do your tasting experiments the next day (or next week) with a set of measuring cylinders & syringes. Find your favourite blend (making up single shots) then blend the parents according to the ratio you like. Or make up different bottles and see how they age. Or age the late tails separate from the rest and blend in 5 years time. Or whatever ... so many choices, so little room for dogma.
Thats it from me for now. I'm going to make a final admission lest someone catches me out - most of what I have written here I learnt making grappa and rum not whiskey. I have got my pot still whiskey recipe down and am about to fire up my old girl (Serena), and have been brooding on this for quite a while, so I think its right. And hey, its a forum for thinking and brainstorming!
It would be good if someday Ian updates his book, with some ideas from this forum, this thread and the comments I'm hoping for from some of the other masters. His book could then be truly excellent. (A friend, at his first day of a Med degree was told "half of what we are going to teach you is wrong. Unfortunately we don't know which half.")
I'm going to start with two simple points and move on to the most interesting and important chapter.
Leaving your still unattended
Ian mentions that you can do this under certain circumstances. This he probably took from Nixon's book which also discusses this. I, and others in these pages, think this is incredibly stupid. Suppose the (cooling) water stops due to municipal repairs? How will you know if you are out shopping? Where will that flammable alcohol vapour go? Where does your home insurance sit relative to an unattended still?
Is the flavour in the heads or the tails?
At different places he makes different statements. Both contribute, the most important is the tails (I'd much prefer a whiskey without heads to one without tails.) See http://homedistiller.org/forum/viewtopi ... y#p7278767
The smell of the heads is usually dominated by ethyl acetate. It has a nail polish smell. Many people confuse this with acetone, which is similar. Nail polish can contain both acetone and ethyl acetate. Ethyl acetate is produced when certain bacteria combine oxygen and alcohol to produce acetic acid, which then reacts with the ethyl alcohol to become ethyl acetate (so ferment with an airlock and distil straight after fermentation eases to minimise ethyl acetate.)
Ethyl acetate can be detected in many bourbons (e.g. Makers Mark), whiskeys & rums. It is the primary smell component of sugar wash spirits that are purified with carbon - as carbon does not remove it, though most other things get taken out. However a good fractionating column will separate it, even though its boiling point differs from the ethyl alcohol + water azeotrope by just 1 degree C.
Tails have an earthy taste. Although they can be unpleasant, this is often because they are separated from the overall spirit and are out of balance. When drinking a good whiskey (say Laphraoig), I wonder how bitter was the spirit when the distiller ended the tails cut. Ian claims that for more full-bodied spirits the distilleries will end the tails cut quite late, and that makes sense to me.
Using a fractionating column to make whiskey
This gets a whole chapter of this book, and he only describes one way of doing it. Smiley (correctly) observes that you can use a batch fractionating column to make good whiskey, but also tries to stick within the legal definition of whiskey - a spirit collected with an alcohol content not exceeding 80%. But the commercial distilleries use continuous reflux stills or pot (/alembic) stills so this mix of still type and law doesn't make sense to me. Many people making whiskey at home are breaking a much bigger law than this one!
And there are some other very interesting things you can do with a variable reflux fractionating still. A fractionating still sorts liquids by boiling point. In an ideal fractionating still, the temperature would rise in steps and the different components of the wash would come off in boiling point order: acetone 56, methyl alcohol 64.7, ethyl acetate 77.1, ethyl alcohol 78.4, propyl alcohol 97. In practice, compounds azeotrope and because the still is not perfect, get smeared together. But in a column with an infinite number of plates and with an infinite reflux ratio, you would get a stepped temperature and corresponding compounds in the output. And by changing the collection vessel each time the temperature steps, you would sort the components of your wash. With a pot still, the distinction of heads/heart/tails and remaining low wines is a sorting into boiling point categories, but with the components smeared across the collection vessels.
A fractionating column with adjustable variable reflux ratio is a truly strange beast. With the reflux set to 0 (no recirculation) and assuming it is well insulated it is like a pot still. With a little bit of reflux it is more like an alembic still. And with high reflux it is only itself. And you can change this through the run.
So what Ian does is to run his still at a reflux ratio of 2 or so (though he is not fussy about that) but then starts collecting at 80% alcohol. The same wash in a pot still might start out below 80% so it would go into the final whiskey. In a pot still in which there is no fractionation, low wines below 55% will not produce a distillate exceeding 80%. But with a fractionating columnn the product will start with a much higher alcohol concentration, so a lot of alcohol is discarded. And then, as he has compressed his distillation - though while still trying to use the accepted percentages for the tails cuts - he has a tight window as he works out where the end point is. He worries about this a lot, and even has the boiling point changing while filling a hydrometer! So I bet he does not get much whiskey out of his runs, but I'd suspect it would taste pretty good. (At least thats my reading after sitting quietly on a plane for too long and trying to untangle that chapter. If anyone reads it differently, I'd be really interested to hear.)
So here are some other techniques you can do when using a fractionating still to make whiskey. They can be combined or used individually. First, lets ask the question "what is wrong with a pot still or an alembic still for making whiskey?" Not much really. All good scotches are made with a pot/alembic still. But there are several things that can be improved upon with variable reflux fractionating column.
If your wash has too much ethyl acetate for your taste (a common case in home made whiskey), with a pot still you have to make a late switch from foreshots/heads to collected whiskey. You then have some orphan spirits that are smelly and need to be thrown out or get additional treatment. Carbon won't make it taste better.
With a fractionating column, you can start the run on complete reflux, let it stabilize, then set it for a high reflux ratio. The foreshots (really early heads) - methyl acohol, ethyl acetate and acetone - can be removed without much loss of ethyl alcohol. When the distillation is through the this stage adjust the still to no reflux and it will then act like a pot still for the next stage of the run. If you decide you want a bit of ethyl acetate (some people like it) you can blend a touch back in later. This way you have complete control, while not losing much alcohol with what you discard. You can also split your heads into several different stages which will have different tastes which you can blend at your leisure.
The second technique that can used to improve things over a standard pot still is around effective use of tails. I hate to throw tails out - they have most of the flavour of the whiskey. You can collect them up and distill them on their own, but if you are like me and use corn one batch, rye the next, barley the next, maybe some grappa etc, you end up with a very mixed collection. What I'd prefer do is to add the tails to the parent spirit in a way that enhances it.
So here is the recipe: run your fractionating column as a pot still until it reaches the endpoint - 60% or whatever you like. Then you have made a normal pot still whiskey. Then switch collection containers and adjust things to push up the reflux ratio. The alcohol content of the distillate will go up. This can be repeated several times until the still is on a very high reflux ratio and the flow is very slow. Then turn the still off.
The taste the contents of those late collection vessels will be terrible. But don't despair and don't throw them out! The reason they taste terrible is because they are out of balance, not because they are bad. A bit like eating a bird's eye chilli - on its own it is inedible, but in a curry with all those other things it enhances the flavour. It can be scary adding really vile stuff to your nice clean hearts, but I have been totally amazed at the transformation.
So what you have to do is find out whether the tails enhance or detract from your whiskey. So take some of the main cut, and measure a proportional content from the tails cuts. This can be done with a small syringe - 1ml or 5ml or whatever. e.g. suppose you have 3l of main cut, and 300ml of tails for your test you will need 25ml of main cut and 2.5ml of tails. Mix them up and taste it. If you don't like it, throw it out or add it to the main cut. The tiny bit of tails will get lost. Or maybe make a shot with 1.25ml and one with 2.5ml and compare. How much tails you add depends on what you plan to do with your whiskey and how you like it. The whiskey will soften with age, so if you want it to be ballsy when it is 10, like many great whiskeys, it will be very intense when it is new. Adding tails is a risk or an investment depending on your point of view.
Another thing you can do with your fractionating column is to concentrate your whiskey. Suppose you have a whiskey recipe you like, but you want to make one that is just more intense. So make your normal wash, distill it with a fractionating column but with no reflux i.e. pot still mode. A quarter way through the run, set the reflux ratio to infinite and let the still equilibrate. Switch containers. Then set the still for a reflux ratio of 5 to 10. Then take a volume equal to (say) 1/2 of the entire batch. This will be neutral, or very close depending on the length of the column. Then switch containers again and open the needle valve right up so the still goes back to being a pot still. Complete the run as normal.
By blending the first and third containers you will have a cogener profile the same as your normal whiskey, but with half the ethyl alcohol. Maybe you don't like it. Then blend back the neutral spirits.
So the idea that a fractionating column strips all the flavour out of the whiskey is misleading - nothing is stopping you from adding it back in.
And I'm sure there are lots of other things you can do. A fractionating column with a variable reflux ratio is a very flexible device.
How critical are heads and tails cuts?
A lot of books (including Smiley) talk about the "right" place to make the cuts, but I don't believe it is that simple. Some people like light whiskeys, some like whiskeys that cause your chest hairs to drop out. In scotches there is Glenlivet and Laphroaig. Light whiskeys end the tails cut early, big ones go late. There is no "right" point. So how does one decide? Smiley talks about putting all the whiskey in one container straight off and redistilling if its wrong. What a waste of time! Distil into separate jars. Do your tasting experiments the next day (or next week) with a set of measuring cylinders & syringes. Find your favourite blend (making up single shots) then blend the parents according to the ratio you like. Or make up different bottles and see how they age. Or age the late tails separate from the rest and blend in 5 years time. Or whatever ... so many choices, so little room for dogma.
Thats it from me for now. I'm going to make a final admission lest someone catches me out - most of what I have written here I learnt making grappa and rum not whiskey. I have got my pot still whiskey recipe down and am about to fire up my old girl (Serena), and have been brooding on this for quite a while, so I think its right. And hey, its a forum for thinking and brainstorming!