憑高酹酒，此興悠哉

沃富奔(Wolfburn)的創廠人Andrew Thompson於11/17/2016訪台，豪邁舉辦了一場餐酒會，席間我向Andrew提出一些技術問題，但無法盡興討論，所以會後寫了封信給他，相隔幾日後回信，洋洋灑灑寫了數千言解釋製程中種種技術問題，不僅讓我茅塞頓開，也解除了我許多一直難解的疑惑。我將回信翻譯如下，也附上原文，希望對於技術有興趣的酒友們好好研究。

問題1. 
你提到溫和蒸餾的重點在於讓液體高過於蒸氣管即可，那麼與蒸餾器的容量比較，相當於多少比例？在業界的平均標準又是多少？我於"Whisky: Technology, Production and Marketing"這本書上看到大概是2/3，正確嗎？

回覆1. 
蒸餾器的容量並非數值化(digital)的定值，製作者或訂製人必須與其他設備共同考慮，包括發酵槽、糖化槽的容量。以Wolfburn為例，採用所謂的「同批次平衡系統」(1 balances system)，也就是1噸的糖化槽=>發酵槽=>酒汁蒸餾器=>烈酒蒸餾器=>新酒暫存槽(Intermediate spirit receiver, ISR)。並非所有的酒廠都有ISR，部分酒廠製作的新酒直接送到SWRV (Spirit warehouse receiver vessel，或稱之SRWV)而後稀釋入桶，不過ISR的優點是，萬一某個批次的蒸餾出現問題，譬如酒心切點切錯，那麼只會浪費一個批次的酒，而不致白耗一整個星期的工作(如果SWRV容量可裝滿一個星期的製作量)，同時ISR也較容易計算每個批次的新酒產量。

如果沒有上述平衡系統，那麼必須在某個階段將容量相加，或分開處理，譬如Kilchoman酒廠1糖化槽=>發酵槽=>2酒汁蒸餾器，所以每批次的發酵必須蒸餾2次，可採行：
       1. 減緩糖化以便讓蒸餾跟上，或
       2. 裝置發酵汁暫存槽(wash charger)來放置發酵完成的wash，以清空發酵槽來容納下一批的糖化汁。

Kilchoman的實際做法是1，因為他們沒有發酵汁暫存槽，而Pulteney擁有一個巨大的暫存槽，所以他們無須等待。

考慮人工以及能源消耗，「同批次平衡系統」可說是最簡潔優美的方式，不平衡的設備可能因建廠時資金不足，只能採購二手設備，或後續產能擴張所致。

回到蒸餾器容量問題。Wolfburn的糖化槽是1噸，但我們實際放入1.1噸的磨碎麥芽，造成較厚的麥芽糊床以及較大的滲透壓力，因此濾出較多固態顆粒的糖化汁。差別雖小，但上帝就在這微小的細節裡，不過必須記住，糖化槽的規格建議底部篩板的最大壓力為1.6噸/平方米，超過可能將篩板完全堵塞，必須施加背水壓來沖洗篩板，以致得到完全非預期的糖化汁，或甚至破壞篩板，導致大量固態顆粒流入冷卻用的熱交換板(heat exchanger, HEX)而造成堵塞，在冷得要死的天氣裡去清理HEX是件要命的工作，所以謹記，任何一件小錯誤都可能衍生後續的大麻煩。

下表是Wolfburn一個星期的製作情形：

 
在1噸的糖化槽放入1.1噸的麥芽，可製作出約5000公升的糖化汁和發酵汁，再放入5500公升的酒汁蒸餾器。不過同樣的製程，糖化汁上下浮動可達200公升，比重也有差異，很難去控制所有的數值一致，因為大麥或麥芽總是不同，導致磨碎的麥芽(grist)比例不同，後續再努力仍無法得到完全相同的糖化汁。但記住，糖化汁的量也不算太要緊，要緊的是接下來如何蒸餾，以及得到多少的新酒，複雜嗎？感到疑惑嗎？我才剛開始暖身而已。

讓我們來談談管線。我們有糖化槽、發酵槽和酒汁、烈酒蒸餾器，槽與槽之間必須以管線連接，而如果槽間距離遠，那麼用泵浦抽送時必定有一定量的液體殘留在管線內。想用泵浦打入空氣來清空管線，可能會燒壞泵浦，如果只是抽乾槽體，那麼還是有殘留液體，怎麼辦？還是得靠壓縮空氣迫使液體流出，但必須使用另一支空氣管，這再度證明了容量僅僅是類比(analogue)而非數值(digital)。

所以回覆你的問題，蒸餾器的容量是很彈性的，具有調整空間，而你從表中數字可以看到，蒸餾者每天都需要調整。蒸餾器上標示的容量只不過是個數字，卻常常引發問題，我們的蒸餾器在製作商Richard Forsyths的報價單上是5300公升，在製作圖上是5500公升，而在委交文件中是5400公升，根本就不是個定值。

那麼哪一種蒸餾方是最好？這不是個好不好，而是你想要、你希望怎麼做的問題。溫和的蒸餾方式可以做出某種風格的新酒，強力的蒸餾方式則做出另一種，其間差別在於：

       1. 理想上，「同批次平衡系統」可以讓蒸餾者從容的以糖化所需時間來做蒸餾，假若蒸餾花8小時而糖化只需6小時，一天中可能浪費2小時來等待，所以我們的做法是讓這兩者花費時間相同，都是6小時，一天便可做2次蒸餾。我們使用板式熱交換器(HEX)，讓進入酒汁蒸餾器的發酵汁通過HEX來預先加熱，其流程如同我們官網上所示，而透過預熱可以節省時間。

       2. 強力蒸餾可能導致巨量波動(surging)－我們開始進入技術層次了。如果發酵汁充滿細微顆粒而較為混濁，如果蒸餾器裝得比較滿或是瘦長高聳，為了讓底部的熱傳遞到頂部，必須提高蒸餾管輸入的蒸氣溫度，而這種高溫將讓與蒸氣管接觸的液體立即蒸發，造成不均勻的傳熱，導致低度酒流進保險箱時，並非穩定流的狀態，而是一波一波的噴發出來。雖然噴出的低度酒同樣可以收集，但如果遲遲無法穩定則蒸餾將十分難以控制，也可能損毀保險箱內的收集器。就算是烈酒蒸餾也可能發生這種狀況，如果沒清理蒸氣管的話，所以我們每星期都必須清理一次。

問題2 
你同樣提到糖化的重要，我發現對於蘇格蘭慣用的糖化槽有多種說法，包括 "lauter tun", "full-lauter"以及"semi-lauter"等等，我知道"lautering"是啤酒業界對「過濾」的說法，但不清楚威士忌產業又是如何定義，能幫忙解釋嗎？

回覆2
在蘇格蘭我們將「耙」(rake)稱為"lauter"，如果是"semi-lauter"，那麼耙上的葉片是固定的，也就是說無法改變其平面角度，如果能變更，那麼就是"full-lauter"。耙(lauter)的作用並非一般人想像的攪拌功能，而是提起麥芽糊底來減輕篩板的壓力，藉由螺旋效應讓工人能控制滲流量，並產生糖化汁往下滲流的通道。耙移動得非常緩慢，但如果要增加滲流量則可加快速度，我們的糖化槽大概是每小時轉1.5圈，我自己一般控制在每小時1圈。如果是"full-lauter"，更能藉由調整葉片角度來改變提升時的阻力，進而影響滲流量，有些蒸餾者喜歡這種設備，有些不喜歡，我個人屬於後者，因為一旦有可活動的零件就有可能出錯。因為同樣的理由，我不喜歡在發酵槽上安裝掃除泡沫的switcher，但也拒絕使用消泡劑，所以我們有異常巨大的發酵槽。

Glenfarclas使用的semi-lauter糖化槽

混合磨碎的麥芽和熱水須使用strike，也就是攪拌器，這個程序稱為mashing-in，須控制熱水和麥芽注入速率，以便得到最適當的水粉比。這個程序比想像中困難，因為麥芽顆粒尺寸不一，每天進行磨麥處理都可能改變。有人會說為什麼不減緩混合速度就好？建議不錯，但是忘了水溫。熱水來自熱水槽(hot liquor tank, HLT)，內部有加熱蒸氣管，但不能降溫，因為降溫得添加冷水，如此一來則槽內水量將超過下一批次的需求量。假設HLT的水溫為65℃，並不代表進入糖化槽的混合液也是65℃，因為得通過輸送管而喪失溫度。現在你知道這有多困難了，但這只是表面而已，糖化是動態操作，需要絕大的技巧才能得到穩定一致且有效率的結果，有人說這是整個蒸餾廠最困難也最需要技術的流程，同儕會藉此來評斷蒸餾者的能力。你的酒精產出率夠不夠、品質好不好，如果以上皆是，那麼你擁有一位厲害的mash man，至於效率差的糖化並不代表新酒風味有問題，但產出率一定不好。

在蘇格蘭威士忌產業，如果你宣稱你的酒精產出率為370或380 LPA/噸，我不認為你會待多久，甚至根本不會被雇用。我參訪過世界各地許多蒸餾廠，但很奇怪的，許多酒廠並不在意低產出，我無法理解這一點。你的mash man不僅代表了你的糖化汁品質，也代表了新酒品質的一致性，同時也是原料與能源使用的效率象徵。如果新酒品質相同，但每噸麥芽的產率有著410 LPA和390 LPA的差異，那麼每一批次的糖化便有著60瓶酒的差異。
 

原文如下：

Dave,

Apologies for only getting back to you now. I have had a truly epic travel schedule with Wolfburn. I am now sat on what I hope is my last flight of many for the time being and have answers to your questions. Before I do though I must thank you for coming to see me and learn about Wolfburn. It is a great pleasure meeting enthusiasts like yourself with a thirst for not only whisky but knowledge. 

1. A stills capacity is not a digital figure. It can not be measured as a finite number. When the still is made it is made to cope with whatever the wash back puts into it and the wash back is filled from whatever comes out of the mash tun. So in the case of Wolfburn we have what is called a ‘1te balances system’ - so this means we have a 1 tonne mash tun and that one mash = one filling of a wash back = one filling of the wash still = one filling of the spirit still = one filling of the ISR (intermediate spirit receiver). Now a few notes before I carry on:

Not all people have an ISR and instead send spirit straight from the spirit still to the SWRV (spirit warehouse receiver vessel NB: some people call it s SRWV but they are the same vessel). The advantage of an ISR is if you have a problem or a mistake is made that sends something other than your spirit cut onwards you would only spoil one distillation periods worth of spirit and not a whole weeks if you had a full SWRV. It is also much easier to calculate your spirit produced if you have an ISR. 
If you do not have a balances system then you are dividing up, or adding together, various liquids at certain stages. Such as:
At Kilchoman one mash  = one wash back = two distillations - so you have to distill twice for every mash
So you either slow down your mashing and allow the stills to catch up OR
Have a wash charger - a wash charger is a tank that is used to store wash so as the wash back can be emptied and the next mash got on with. 
Kilchoman have to wait as they do not have wash charger and Pulteney have a huge wash charger tank so do not have to wait. 
So a balanced system is the most efficient in terms of man hours worked and fuel expended to get the same volume of spirit at the end. Unbalanced systems usually exist because of either expansion of production or lack of funds at the initial build phase. 

So we get back to the stills volume. Our mash tun is 1 tonne, but to further refine the drain and create a clearer wort we put 1.1 tonnes into the mash tun. This creates a higher bed pressure and so filters out more solids. The difference is tiny but as we know it is this attention to detail that gets you what you want. So remember 1.6 tonnes per m2 of floor plate is your maximum recommended bed pressure. You can go lower but not really much higher. If you go higher you run the risk of blocking the plates completely which then required a back flush and then you will have a very different drain (and therefore different style wort, wash and spirit further down the line) than you wish for OR damaging your plates and them bending to the point that too much of the fine particulate gets through the seams and blocks your worts cooler heat exchanger (HEX). Stripping and cleaning a HEX in the cold is an awful job and is a disaster in terms of your drain time; remember anything that stops one operation has a trickle down affect and affects everything further down the line. 

So I have a 1 tonnes system but put 1.1 tonnes through it. I also have a 5,500 litre still but only put 5,000 litres in it. Here is a snapshot of a weeks mashing:

So you can see a mash can fluctuate by as much as 200 litres further changing the volume that actually goes into the still. You can’t force more water through to get this as a uniform figure. It does’t work like that. The barely is always different, so the grist is always a struggle to get consistent and no matter how hard you try it will always be a slightly different drain each time. And ultimately remember it does not really matter how much liquid you get through your mash tun as it is what comes off it that counts. What you distill from this liquid is the volume that counts and that is measured in the alcohol you get off each run. Complicated? Confused yet? I haven’t even started. 

Lets talk about pipework. The vessels we have in a distillery are the mash tun, washbacks and stills. But if your vessels are far apart then you can have a considerable volume of liquid left in your pipes each time you pump from one vessel to another. If you run the pumps dry you will damage them and blow a pump, if you stop them at just the right moment you will still have liquid in the pipes. You can an air purge them but now require air lines which can leak and suffer if your pump runs at a high pressure. So another example of your volumes being moved abut being analogue not digital. 

Getting back to your original question - the still is made with room for manoeuvre. Flexibility built in to allow the manager freedom of manoeuvre when he needs it and as you can see he needs it every day. The figure that is usually written on the stills somewhere showing volume is purely a customs requirement to be displayed and probably why people get drawn to the volume question. The man who made my stills, Richard Forsyths called it a 5,300 litre still on his quote, a 5,500 litre still on his technical drawing and a 5,400 litre still on his commissioning documentation. It’s just not a clearly definable number. 

So what is best? It is not a question of best but more a equation of what you prefer and what you want. A slow gentle distillation is an option to create a certain style of spirit, a harder distillation is the same an option to create a certain style of spirit. There are issues with both:

A gentle distillation may not be quick enough for you to get through the still what you need to get through. Ideally in a balanced system your mash takes around the same time as your distillation to allow for a smooth work process. If you a gentle distillation means you need 8hrs to get your stills through but only 6 for your mash then at some point you lose two hours a day with all that entails. For us we have it just rift at 6hrs for each meaning a double distillation and mash in one day takes 12 hours, not 16 as would be the case in the example given. Mine is 6/6 because I use a effluent heat scavenger HEX charge system. Ask me if this is an unknown to you. Not many people do this. It is a means of preheating the wash as it goes into the wash still using a HEX and the effluent form the last still run. For a further explanation of this check out my website with a diagram showing the flow of this. By preheating it I save time. 
A hard distillation could cause surging (now we are really getting technical……). If you have a very cloudy wash you have particulate in the liquid you put into your stills. If you have a very full still, or narrow and tall, and need to get heat all the way up it you have to have a high temperature in those coils. This high temperature though can cause the particulate to burn onto the steam pipe. If this happens you get unequal heat transfer and then you get surging. Surging is when the low wines are flowing into the safe they do not come out steadily but in spurts that can be very powerful. It makes no real difference to what you are collecting but is a very hard still to control as it never settles down. It can be so powerful it smashes your collection jars. This can even happen with a clear wash if you do not clean your coils. We clean our coils once a week with wire wool. They look exactly the same before and after the clean which shows you the tiny amount of particulate on the could that can radically affect the distillation. 

I am sure I have now brought up more questions on that subject and I am happy to answer them. 

2. In scotch a lauter is what we can the rake in the tun. The rake is the arm that turns around as we mash. If it is a semi lauter the blades on the rake are set i.e they do not move in a horizontal plane to change their angle. If it is a full lauter it means you can change the angle of these blades. What does the lauter do? It lifts the bed to decrease bed pressure, by a corkscrew effect, to allow the mash man to control his drain and at the same time create channels for the worts to run down further influencing the drain. It is not for mixing as a lot of people presume. It moves very slowly. If we are trying to increase the drain it will move more quite quickly for us at 1.5 full rotations per hour. Normally though we rotate at just 1 rotation an hour. So if you have a full lauter you can further influence your drain by adjusting the blades angle as eel for more or less liftl. Some people like them, some people don’t. I do not like them as they have moving parts and can go wrong. The same reason I don’t like switchers but refuse to use anti foam so have extra large washbacks. 

The mixing is done at the strike. The strike is when you add to the mash tun the hot water and the grist. They have to come in at exactly the same time and finish coming in at exactly the same time and at your preferred rate of entry. If you have water left over (remember for your 1st water it is the 3rd water from the previous mash) you can’t just throw it in as it will sit on the top as a puddle and not drain and if you do not add it then you have lost some of the sugar that is already in your water as it is your 3rd AND 1st water so has sugar remnant in it. So they need to come in at a very specific rate to achieve as best a mix as possible. This is harder than you think and changes every day as of the grist which changes remember as of the barley grain size varies through nature. Many people ask why you do not just then do the mix very slowly to get a great mix. Good idea but they forget one thing. The water temperature. Your water temperature is controlled at the hot liquor tank (HLT). The HLT is where you can add heat as it has a steam coil but you can’t cool it down as that would mean adding cold water and then if it is your 3rd and 1st water you will then have too much water for the next mash - disaster. So if your HLT is perfect at say 65 degrees that does not mean it will be 65 deg when it goes into the mash tun. Your strike temp will always be lower than your HLT temp as the HLT is connected to the tun by a pipe. That pipe may be cold, it may be hot, it will certainly lose temperature if you pass the water through it very slowly. You could ask well why not insulate the pipe? As then you can’t use the pipe to cool the water if the still man needs a tiny adjustment. So the mash man has to try and achieve not only getting all the water in but also at the right temperature. Now this is very, very hard and only scratches the surface of the intricacies of the mash. Mashing by it’s nature is a dynamic constantly changing process that requires great skill to get right in a consistent and efficient manner. It is arguably the most difficult/skilful job in the distillery and where a distillery is judged by its peers. What is your alc yield per tonne and does your whisky taste good? If the answer to both of those is yes then your mash man is good. An inefficient mash though does not mean a drop in taste quality - just poor production volume.  

I have visited many of the new world distilleries and this is the area that is very different to Scotland in quite a few of them. If you were declaring yields of 370 or 380 lpa per tonne in Scotland I do not think you would stay employed very long,  certainly would not employ you. Abroad though it seems people are quite happy with lower yields in quite a few cases. I do not know why this is. The quality of your mash man not only means the quality of your worts, and therefore the quality and consistency of your spirit but also how much of it you make for the same cost of raw materials and energy. If your spirit is exactly the same but one guy is able to make you 410 lpa per tonne and the other only 390 lpa per tonnes then one guy is losing you 60 bottles every mash. It is by no means everyone as Kavalan for example are very very efficient as is Yamazaki. More the smaller guys in my experience. 

So going full circle a full lauter is another aid to getting that drain right. There is no best way to do it, some people prefer one way and others another way but whichever way you are doing it you must be good at it. 

I hope I have answered your question in some way and not just confused you. I expect though from the high level of your questions both at the forum and in your mail my answers will give you something to cogitate upon. 

Feel free to ask for clarification on any of the points above an thank you for being interested in Wolfburn. 

Andrew