Is Brewing an Art or a Science?

This is a question I think I've danced around nicely in the subtitle of this blog. Is brewing an art or a science? This is a common question and one without a good answer (I may muse about it for a while, but I promise I won't come up with a real answer, either). I recently watched a preview for Tim Webb's Beer Amongst the Belgians, which is a series about which I am very excited. In it, Tim Webb interviews Jean Van Roy, head brewer of the Cantillon brewery. Cantillon makes some profoundly amazing beers (all lambic/geuze), so I hold the greatest respect for Van Roy. In the interview, he states that brewing is 100% art. He sleeps with his beer, brews by feeling. He thinks of himself as an artist, and I, having had several of his products, am inclined to agree. But that quote makes me wonder what he would think of my own approach to brewing. I tend to think about it in a scientific way, though I'd like to think I'm not cold and mathematical about it. There is a world of difference between Cantillon and the large, industrial brewing operations that dominate the American market. They are the other extreme. Their mission is to replicate their product time after time as efficiently as possible; their brewers are engineers, their breweries are factories. My own approach is far more artistic by comparison. Do the brewers in these massive operations feel any kind of emotional connection with their beer? I doubt it. Now watch the interview with Jean Van Roy, and tell me he isn't emotionally attached to his work. It almost brings a tear to my eye. I think I fall in between. I believe there is a happy middle ground between the pure art and the pure science. The science of brewing is very useful. Understanding the chemistry of the beer and the biology of the yeast can help hone processes and design a very tasty brew. It's particularly handy for troubleshooting: if something goes wrong, a scientific approach is necessary to find and correct an error. But having said that, there are some aspects where pure science won't do and the brewer must be an artist (or, more appropriately I suppose, a craftsman). There is a conceptual difference in selecting hops for, say, a hefeweizen because their subtle floral aromas complement the dusty malt, and selecting based sheerly on cohumulone and alpha acid content. While the chemical makeup shouldn't be ignored, the abstract role in the overall beer is more important. I could say the same for most other aspects of the brewing process. So I guess my ultimate thought is that the brewer should play artist first and develop a beer as an idea, or to use an appropriate metaphor, paint a picture of it. Then the scientist should kick in to make hone the process to bring that beer into reality. These are just my own thoughts, and I'm sure everyone is going to approach this in his/her own way.

Thanks for humoring me.
Cheers,
--joe

Revisiting Corsendonk Abbey Pale

Somehow I got it into my head that I need to have a Corsendonk Abbey Pale Ale. I haven't had this beer in many years. It is a Belgian tripel, and due to its extremely wide distribution, in my head it is an entry-level beer into the sophisticated world of Belgian ales. But I decided that I'm being unfair, especially seeing as how I barely remember the beer at all given the length of time since I tried it last. So it's about time that I give it another go. I may as well share my thoughts here.

I started with a 750 mL corked and caged bottle. My very first observation is the Champagne-esque pop of the cork when I open it. I generally try not to do this with beer, but I wasn't able to prevent it. It's a well-carbonated brew. Pouring into my glass I'm actually a little surprised that it doesn't foam over, given the pressure that had built up in the bottle. Pretty impressive. There is plenty of effervescence, though. Vigorous bubbles maintain the white head for a very long time. The beer itself is a clear golden yellow. It's an attractive beer, so we're off to a pretty good start. I get my nose to it quickly to get a strong whiff of the aroma. The energetic carbonation helps disperse the aroma nicely. The first scent that registers is a phenolic yeast character common to Belgian ales of this sort. In this case the phenols come across as spicy, reminding me of coriander, fresh peppercorns, and lesser amounts of rose hips and sage. After I take in a few whiffs, the spiciness begins to take a backseat to the obvious Pilsner malts, marked by a subtle doughy malt sweetness. Also mixed in there I detect notes of lemon, subtle light honey, and just a bit of toasted bread.

Enough sniffing, it's time to take a sip. I notice the texture before I'm able to pick out specific flavors. That effervescence makes it seem to foam up in my mouth, giving a beer that may otherwise seem thin (it's fermented down to a fairly dry state) quite a full impression. Much like the aroma, the flavor is dominated by the phenolic notes. However, it tastes somewhat less balanced than it smells. Those phenolic notes outweigh the malt more than I care for. Not to say the malt isn't there; there is a slight honey and toast melange in the background. The beer does have some nice fruity esters. The lemon I noticed in the aroma is joined by some faint pineapple, orange, cantaloupe, and mango. Very faint, but present. It finishes on the dry side, as should be expected for a highly attenuated beer such as this. Dusty yeasty notes, a little bit of a wooden tone, and something like saltine crackers bring the sip to a close. Despite the dry finish, the aftertaste returns to the little bit of maltiness, but the spicy phenols stick around longer.

I decided that this beer needed a cheese pairing. Fortunately I had some Asiago cheese on hand. This turned out to be a very successful pair. Asiago cheese is fairly subtle, so doesn't overpower the beer, which lacks any very strong flavors. The earthy and musty tones of the cheese go well with the spicy phenols of the beer. And Asiago isn't creamy at all and gives an impression of dryness, also in harmony with the beer.

So, all in all, Corsendonk Abbey Pale Ale is a pretty good beer. I'm definitely glad I tried it again. It might even be worth drinking once in a while. It would definitely be a good beer for pairing with food (as in meals, not just cheese); Thai, maybe? I may have to do that one day. For now, though, I'm gonna sit back and enjoy the rest of this bottle.

Cheers,

--joe

Experimental Mash Math

Warning: this is gonna get kinda technical. If you don't like math, you should probably stop reading. This is more of a mental exercise than a useful procedure...

Anyway, I've been tinkering with an idea for a mashing technique a little bit different than my usual process. It has very limited applicability, but might come in handy from time to time. I started thinking about it when planning a multi-grain ale I'd like to brew soon, including barley, wheat, rye, and oats. With these other grains involved, it would be beneficial to include a rest in the mash to break down β-glucans before the main saccharification rest. Otherwise the mash can get very gummy and hard to work with; there could also be an unattractive haze in the finished product. The enzymes that degrade β-glucans are active at a relatively low temperature, between 98 and 113 °F. So I would like to hold the mash at this temperature for 30 minutes or so before raising it to 150°F or so for the amylase enzyme party. While this is not strictly necessary, I take pride in my craft, so a little bit of perfectionism is to be expected. It is also common to include a rest to degrade excess proteins when using different grains. In this case, however, I'm going to let the protein rest slide. Between the highly-modified base malt and the interesting mouthfeel I'm hoping to get in the finished beer, I think the protein level will be acceptable.

In short, I need to mash in at 105°F, rest, then raise the temperature to 150°F. Most homebrewers do this simply by adding a calculated amount of boiling water to raise the temperature. Some still use the old German practice of decoction (boiling a portion of the whole mash then mixing it back). I don't care for either of these. The water additions dilute the mash and usually result in a water-to-grain ratio much higher than I prefer. Decoction does more harm than good; boiling the grains extracts tannins and a host of other things (including the β-glucan I'm trying to reduce), and moving around portions of the mash can be a pain. Well, I thought, what if I raise the temperature by draining a portion of the wort from the mash (liquid only, no grains included), boil that, and add it back? Besides eliminating my complaints about the other two techniques, it will also help precipitate undesirable proteins and should cause a slight Maillard reaction (i.e., caramelization) that could add a subtle complexity to the brew. The one downside to the process is that quite a bit of amylase enzyme will be deactivated by the boiling. Make sure that the total diastic power of the mash is enough to convert the starches even with a loss of a large percentage of the enzyme. My base malt will be of high diastic power, so this should work.

So then. Assuming that this process is actually a good idea (still up for debate), we come to the problem of figuring out how much wort to drain and boil to achieve the desired temperature change. There's a fairly common equation used to calculate boiling water additions (in quarts, pounds, and Fahrenheit):

W_a = [(T₂ – T₁)(.2G + W_m)] / (T_w – T₂)

where W_a is the boiling water to be added, T₁ is the current temperature of the mash, T₂ is the desired final temperature, G is the amount of grain, W_m is the amount of water in the mash, and T_w is the temperature of the addition (boiling). In the process I've described, though, water isn't added, it's recycled from the mash. We can adjust the equation above by substituting (W_m – W_a) for W_m, since the volume W_a will have been removed from the mash. Solve again for W_a, and we get:

W_a = [(T₂ – T₁)(.2G + W_m)] / (T_w – T₁)

Fair enough, but this equation assumes either that the removed wort has the same thermodynamic properties as water or that it is somehow possible to drain pure water from the mash, devoid of grain constituents. Of course, neither of these is the case. I decided to re-work the equation to make it more-closely reflect the thermodynamic reality of my experiment. It will be interesting and useful to know how much difference these adjustments make in the result.

So, to begin, let's take a step backward. The equation above derives from a basic statement of the heat energy of a substance:

Q=mCT

where Q is the heat energy, m is mass, C is specific heat (that is, the energy required to raise the temp of 1kg of the substance by 1°C), and T is temperature. Also relevant is the fact that the total heat energy of a system or mixture is the sum of the heat energies of its parts. So, in our case (treating the water remaining in the mash and the wort removed as separate parts),

Q_total = Q_grain + Q_water + Q_wort

then, by substituting from the equation above:

m_totalC_totalT_total = m_grainC_grainT_grain + m_waterC_waterT_water + m_wortC_wortT_wort

The total mass of the system is obviously the sum of the masses of the parts. So we'll substitute that in. Likewise, the specific heat is the sum of the individual specific heats. The total temperature is the desired final temperature, so let's just rename it T₂ for the sake of simplicity. T_water and T_grain will be equal in this scenario, so let's rename them both as T₁. The wort removed will be boiled, so conceivably I could go ahead and substitute 100°C for T_wort. However, I'm going to leave it as T_boil just to allow some flexibility. So that gives us this equation to start tinkering with:

(m_grain + m_water + m_wort)(C_grain + C_water + C_wort)T₂ = (m_grainC_grainT₁) + (m_waterC_waterT₁) + (m_wortC_wortT_boil)

Now the fun begins. I'm not going to retype the equation for every step in this process, but there is a lot of tweaking to do from here. First of all, note that since 1L of water is 1kg, mass and volume can be interchanged. This is not the case for the wort, however. Since we ultimately want to figure out the volume of wort to draw off, we must consider the specific gravity of the wort (which can be read with a quick measurement). The usual specific gravity reading is conveniently expressed as a ratio of the wort's density to that of water. We can, therefore, substitute (V_wort × SG) for m_wort.

Next, we can also express the specific heats as ratios of that to water; C_water becomes 1 and reduces away; 0.4 is the typical value for C_grain. C_wort will again be dependent upon the specific gravity, so I'll leave it as a variable for now and address it later.

Again the amount of water in the mash will be however much is left after wort is removed. So for m_water we should insert (V_water – V_wort), where V_water represents the total amount of water in the system. Similarly, remember that the wort will contain starches from the grains as well, so some mass will have to be subtracted from it. The amount is again based on the specific gravity of the wort. To compensate for this transfer of material, we can replace m_grain with the term [m_grain – V_wort(SG – 1)]. Then it is just a matter of the tedious process of solving for V_wort. I'll spare you that. Take my word for it that the resulting equation is:

V_wort = [(T₂ – T₁)(V_water + 0.4m_grain)] / {(T₂ – T₁)[1 – 0.4(1 – SG)] + C_wortSG(T_boil – T₂)}

As I said a moment ago, I left C_wort as a variable. It is easier to figure its value first, then insert it into the equation, lest the equation become more cumbersome than it already is. The calculation involves the strange conversion to °Plato (which describes density in terms of percent sugar in solution) and takes into account the different specific heats of the water and the starches dissolved in it. Furthermore, remember that we used ratios relative to the specific heat of water, so that must also be taken into account. So when all of this is combined, we arrive at:

C_wort = {4.1868 – [7.577(SG – 1)] / [1 + .8796(SG – 1)]} / 4.1868

Plug in the specific gravity, then plug the result into the other equation.

It is important to note also that both the input and output of the equation are in metric units. If you want to use quarts and pounds (which I myself usually do), it's much simpler to do the unit conversion first, then convert the output back again, than it is to try to factor conversions into the equation. Also note that I did not account for the thermal mass of the mash tun; from experience, its contribution to heat loss is negligible.

Now let's make a hypothetical mash and see what we get. Let's say I'm going to mash 10 lb. (4.54 kg) of grain in 15 qt. (14.2 L) of water. The first rest will be at 105°F (40.6°C), then I'll use my unusual method to raise it to 150°F (65.6°C). For this example I'm going to assume the wort SG to be 1.050 for ease of calculation; in practice I would have to take a sample of the wort, cool it quickly, and take a hydrometer reading. For the sake of comparison, using the altered boiling water addition formula, I come up with 7.15 qt. To draw out and boil. My new fancier equation advises me to remove 6.965 L, or 7.36 qt. For boiling. The difference is 0.21 qt, or a little less than 7 fl. oz. That's only about 3% variation, which doesn't seem too major. But, out of curiosity, let's use my more accurate equation to see what temperature I would have arrived at had I used the 7.15 qt. (gotta remember to convert to 6.77 L!) suggested by the simpler equation. Turns out that would have landed me at 64.9°C, or about 148.8°F. Still just a small difference, but it only takes a few degrees of temperature difference to change the character of the finished beer. Also, the difference will increase with increased gravity of the wort, which will vary based on a number of factors. Remember that the 1.050 I used for my example was merely a convenient approximation. Your mileage may vary.

So, long story short, in the name of precision, I intend to try this process and this calculation. I'm not convinced that it will make a huge difference, but I'll settle for the satisfaction of successfully negotiating all of this math and physics. Thanks for bearing with me.

I promise lighter fare for my next few posts.

Cheers
--joe