Discussing Brew Methods

This post is a follow up to the poll I posted a little while ago. I was wondering if my own thinking about brew methods mirrored others, or whether they were different. I admit it was something of a flawed poll, but I wanted to use it as a stepping stone to this larger post. (This is a long blog post.)

My thinking in this revolves around the fact that we often talk about how a certain brewer could highlight certain aspects of a coffee’s taste and quality. I want to explore this idea in a little more depth, because this is such a fuzzy idea that I don’t think it is particularly helpful.

I’m going to avoid talking about specific manufacturers of equipment, primarily because I have a conflict of interest in this area.

I should also point out that a great deal of this post can be written off as anecdotal, and I’m more than happy to discuss contradictory experiences or opinions.

What do you want to highlight?

I figure the best place to start is by taking a slightly more objective look at the potential characteristics of a coffee that we might be highlighting. Most scoresheets used for cupping exist to assess green coffee but there are sheets used for other purposes. There are certain common attributes that are assessed:

- Sweetness
- Acidity
- Mouthfeel/Body
- Balance
- Aroma
- Flavour
- Finish

When brewing a coffee we should be thinking about how the process impacts each of these attributes and presents.  How do we influence each of these factors with our equipment and technique?  This is a more difficult question to answer, because it is all so interconnected.

Most of the above attributes are primarily linked to extraction.  The coffee itself is obviously the determining factor of how much sweetness is available, or how much positive acidity is available for us to capture in the cup but the quality and quantity of each of these in the brew is tied to extraction.

Extraction remains a sticky topic.  I believe that uniformity of extraction is important, and also that whatever people consider their desirable range of extraction is tied heavily to the grinder and its particle size distribution.  Regardless of what your desired level of extraction is, there are certain factors that have a strong influence when we are brewing.  Two of these are deeply interconnected:

- Grind Size
- Contact Time

The first two have the most obvious connection – the grind size determines surface area of coffee exposed and the rate at which solubles extract, and with all brew methods we must be careful to balance our grind size and our contact time.  You can brew a french press at a number of different grind sizes and have good results as long as the contact time is appropriate.

The moment you read that another thought might pop into your head – which is that surely brew time has an effect on brew temperature.  If I grind very coarsely, requiring a longer steep in my press pot, then surely the average temperature of the brew will be much lower?

Temperature is still something that we don’t really understand in brewing coffee.  We’re happy to talk about espresso brewing temperatures, yet references to the brewing temperature of drip coffees is remarkably absent from published recipes or discussion.  Temperature has a brute force effect on extraction – it supplies energy required for a soluble to enter a solution, so generally the more heat you have the more extraction you have.

Along with overall extraction we have the extra effect of thresholds of temperature required for certain compounds to be extracted.  The best example of this are the negative, bitter qualities we extract from even light roasted coffees when the brew temperature is very high, up towards boiling point.  Brewing at lower temperatures – 80C/175F for example – allows a reasonable mechanical extraction of the coffee, resulting in reasonable body, simplistic sweetness but overall a cup lacking complexity and character.  This experiment also highlights the unusual effect of temperature on acidity.  Brewing at temperatures like these doesn’t result in the sour cups we might come to expect, instead the acidity is notably absent – requiring more temperature to be fully extracted.

Back to Brewers

Temperature often seems to be a separate factor from brew method – more a function of technique and recipe rather than the brewer itself.  This is mostly true – your brewing temperature is determined by the temperature of water you start with, and the thermal mass you achieve in your brewing liquor/slurry.  A pourover brew with the bed kept very, very low with a slower pour will have a lower brewing temp than a faster pour and a higher cone – presuming water starting at the same temperature.  This is often exaggerated further by a slow pour from a pouring kettle which is losing heat as well.

This isn’t to presume an ideal technique, just to highlight that the brewer doesn’t really control the temperature.  Except for one:  the syphon.  I find it odd that the temperature profile of this brewer isn’t discussed more.  It is rare to brew coffee in an environment where the temperature is held stable across a period of 1 to 4 minutes, as we usually do not add any energy beyond the brewing water to the brew.  Many people dislike the taste, or are simply disappointed by, the taste of coffee from a syphon.  Some might argue that this is because the person brewing lacks the particular ridiculous ninja stirring skills, but more likely the temperature profile of the brew is what is making it stand apart. 1  In terms of combining infusion and pressure driven percolation it is remarkably similar to an aeropress – something which rarely is discussed.

Speaking of stirring we can discuss agitation.  This is a topic that we are a little squeamish about as an industry, because we’ll almost instantly wheel out the words “inconsistent” or “unrepeatable”.  Usually this is in relation to the manual movement of the coffee with a tool – such as stirring the bloom of a pourover, or stirring the slurry of a syphon.  I’m skeptical of magical stirring techniques in any brew, and it feels like we should also be looking at the other agitation going on – when the water hits the coffee.  How we pour, where we pour, from how high we pour – all this and more will affect the movement of the coffee in the bed and more movement will result in more extraction.

Once again – this isn’t really a function of our chosen brew method – more our chosen technique.

So far there really hasn’t been much about the brewing process that is determined by the brewing device, rather than the brewing human.

Filtration

As we look to other attributes of the coffee an obvious place to see the impact of the brewer is the influence filtration medium.   One important aspect of coffee is how it feels to drink – the body, the mouthfeel, the texture.

This is determined by several things divided into two categories:  Dissolved things and Undissolved things.

The quantity of dissolved things determines our strength, the more solubles the stronger the drink will be and the heavier the body and mouthfeel will be.  This is linked the extraction (technique again) and recipe – not the brewer itself.  The coffee determines the composition of the solubles, and certain coffees contain solubles that increase our perception of mouthfeel.  All brew methods produce a cup where the dissolved solids play a fundamental role.

Undissolved solids are interesting too.  The undissolved we most often talk about are the lipids (oils) in the coffee, and also the tiny pieces of coffee that can be suspended in the brew.  These both have strong influences on the mouthfeel of the resulting coffee, not always in a beneficial way.

Paper filters are considered the most thorough way to remove the undissolved solids, resulting in a much clearer brew with an attribute we would often describe as clean.  Cloth filters allow through more of the undissolved materials, noticeably the lipid fraction but also some of the fines too.  However, there is usually so little fines overall that we usually experience a fuller mouthfeel – something most enjoy and I think cloth would be a more popular filtration method if it weren’t for the annoyances of maintaining the cloth and preventing it imparting unpleasant flavours.  This leaves metal filtration which simply removes the largest pieces of suspended material (i.e. the bulk of the brewing grounds) and results in a cloudier, muddier brew that can still be incredibly enjoyable even if that final mouthfeel is probably worth leaving in the cup.

So far, with the various factors discussed, it appears that the brewer has relatively little influence on cup quality compared to technique and few brewers demand certain techniques – instead allowing infinite varieties of recipes, methods of agitation, or brew temperatures to be used.  There is, however, one more distinguishing factor that should be discussed.

Infusion Vs Percolation

This is probably the biggest distinction between brewers, and the most important.  We have brewers that are solely infusion – like a french press – where extraction occurs as coffee infuses into the water.  We have percolation brewers where water flows through a bed of coffee, washing out soluble materials, and we have hybrid brewers where both occur at different stages – such as a syphon where the water a coffee steep, before the heat is removed and the water is dragged through the bed of coffee against the paper or cloth filter.

When it comes to assessing techniques and extraction I think it is fair to say that using infusion it is easier to achieve relatively uniform extraction, compared to percolation.  With percolation something is driving the movement of the water – it might be gravity, or applied pressure – but this means that paths of least resistance are preferable for the brewing water.  We must adapt techniques to allow for this, and help prevent too much uneven brewing.

Uneven brewing is a problem because it is so hard to measure – outside of taste.  I firmly believe that measuring overall extraction is incredibly useful and has many practical applications.  The critics of measurement have often cited our inability to measure evenness as a reason to not bother measuring at all.  This makes absolutely sense to me, but perhaps not hugely on topic. (Clarification:  AndyS points out that this is a horribly phrased, confusing sentence.  My explanation in the comments follows.)

The fact that extraction measurement can’t quantify evenness is not a reason to abandon and ignore a very useful tool that does exactly what it promises: to measure how much of the total coffee was extracted.  Pointing out what it can’t do as a flaw creates something of a straw man argument to me. Same thing as pointing out that it is misused and therefore a bad tool. (which I see online a lot too)

Creating a brew recipe

Have I argued that any brew method should be able to brew any coffee as desired, or have I argued the opposite?  To answer the question I thought I’d look at how I might make a choice about how I would want to present different coffees, and how I’d choose a brew method or recipe.

If I love a coffee because it has a plump, sweet fruit quality then I might choose to brew it using cloth filtration at a slightly higher recipe than usual.  Considering I take 60g/l as my own preferred starting point, I might instead brew at 65g/l to further emphasise the body.  When talking about the coffee I want to highlight a few simple reasons why someone would enjoy it, and then use my prep to make those points evident.  Plush, rich, heavy, jammy mouthfeel is a big promise, but I feel that if fulfilled a great experience would be had.

Conversely, if I loved the light, delicate tea-like quality of a floral coffee then I might drop back to 55g/l, and make sure it was paper filtered and brewed with a relatively hot brewing liquor.  Sometimes reducing concentration can help with clarity.  I’m not saying all coffees like this would be brewed this way, but to instead talk a bit about how we might think about our recipes and analyse our brewing choices.

I don’t think I’ve really come to any conclusion here, nor did I really mean to.  I just wanted the opportunity to write up a few thoughts on brewing and try and give them some structure.  I’m open to input on this, open to disagreement and probably being wrong about a whole heap of things.  On the one hand I feel like I’ve missed out a great deal but on the other hand this post is probably long enough!  I look forward to your feedback!

  1. My personal preference for coffees to brew in syphons would be very light roasted, dense coffees, that are often difficult to extract  ↩︎

20 Comments

  1. My preference for brewing pourover is a significantly higher brew ratio 80g/L, using either cloth or v.fine double ss mesh.  Pour is reasonably quick, with bed of coffee kept high.  Total brewing time around 2:30-2:45.

  2. Great post. Thanks for the thoughts.

    Just another question…what is your thought about the much touted sentiment, that cupping yields the most interesting and revealing extraction of all methods? I.e. this is the method most revered for assessing a coffee, why?

    Because it is done consistently and repeatably by all roasters, or because it is actually the the best extraction method for revealing the inherent qualities of the coffee and roast.

  3. Cupping is really about treating each bowl/coffee as similarly as possible.  By removing as many variables as possible (ideally anyway) you have a great understanding of the quality and (most importantly) the uniformity of the coffee you are extracting.

    Cupping bowls are the easiest way to brew numerous consistent brews – filtration being impractical and wasteful on a table where you might have 4 bowls of 10 different coffees.

    I don’t think it is the only way we should be tasting coffee – though I think it is entirely suited to its purpose of assessment of the green coffee or of the roasting process.

  4. James, you say, “The critics of measurement have often cited our inability to measure
    evenness as a reason to not bother measuring at all.  This makes
    absolutely sense to me, but perhaps not hugely on topic.”

    Would you mind clarifying what you mean in the second sentence? I don’t understand. Thanks,

  5. Reading it back it makes less sense than intended. The fact that extraction measurement can’t quantify evenness is not a reason to abandon and ignore a very useful tool that does exactly what it promises: to measure how much of the total coffee was extracted.

    Pointing out what it can’t do as a flaw creates something of a straw man argument to me.

    Same thing as pointing out that it is misused and therefore a bad tool. (which I see online a lot too)

  6. Reading it back it makes less sense than intended. The fact that extraction measurement can’t quantify evenness is not a reason to abandon and ignore a very useful tool that does exactly what it promises: to measure how much of the total coffee was extracted.
    Pointing out what it can’t do as a flaw creates something of a straw man argument to me.
    Same thing as pointing out that it is misused and therefore a bad tool. (which I see online a lot too)

  7. I find the pouring of the water of great importance. When I use my own kettle I can get great blooms, but when the water is just chucked in it always tastes worse. 

    I read somewhere you’d looked into synthetic cloth filters, any thoughts?

  8. Hi. This post has created so many questions about my brewing techniques. When we talk about even brewing, I started thinking about my pour-over technique and how temperature stable it is. Using my roast profiling software, I placed thermocouples in the bed of the coffee in the v60 and in the tip of my kettle. I was surprised by the fact that it took almost 2 minutes of pouring (including a 35 second bloom) for the solution of grounds and water to reach 200F. Isn’t that a little slow? That doesn’t seem even to me. If our technique determines our brew more than the pairing of coffee to method, maybe we should be getting to know our brewers better by compensating for what the brewer is demanding of the barista through variables such as ambient temperatures of coffee and brewers for example.

    For a  closer look into our analysis you can check the results on our blog through our twitter @drumroaster:twitter

    Thanks!

  9. Love this, and hope the debate runs. Considering the majority of voters believe there is a direct match between certain coffees and brew methods, I think perhaps there’s a danger that your last few paragraphs might be interpreted as agreeing with them.  But if I’m correct that’s not your assertion…(or is it?)  You’ve picked the ‘plump, sweet fruit’ characteristics as an example, and decided on a brew device to highlight it, but the same coffee may also have other characteristics, and on a different day you may wish to accentuate those instead, via a different brew method. Just cos it’s sweet and full of body, that doesn’t mean it isn’t also bright, complex and perhaps nutty, benefiting from a different brewing ratio in a chemex & paper filter. “What do you want to highlight?” is absolutely the question and the answer (always IMHO).

  10. There is certainly a difference between 200ºF slurry temperature and 200ºF brew water. In a pour-over application, you will always have a positive-slope (increasing) slurry temperature, even though the temperature of the brew water may be negative slope (decreasing over time). This is due to the heat-capacity of the coffee. The brew water will always be hotter than the slurry, so the more water you add, the more the heat-sink is diminished. 

    When we talk about “slurry temperature,” I think we really ought to be talking about median temperature. Besides – who says 200ºF slurry temperature is ideal anyway? Under which conditions can it even be possible? or tasty?

  11. Interesting points, James. 

    “So far, with the various factors discussed, it appears that the brewer has relatively little influence on cup quality compared to technique and few brewers demand certain techniques – instead allowing infinite varieties of recipes, methods of agitation, or brew temperatures to be used. ”

    In my experience, brew equipment does have influence on the cup, but only so far as the constraints they provide for the barista, i.e., the choices they force us to make. For example – a Chemex has a  fairly restricted flow compared to something like the V60. Consequently, I generally choose to grind quite coarse for the Chemex, while I grind quite fine for the V60. Naturally, the extraction times are different. While the finished product is different because of the technique I used in each case, each device provides a different set of mechanical restraints that forces me do different things. In my experience, I can get a Chemex-like cup from a V60, or a V-60 like cup from the Chemex, but I’ll only get close.

    Siphons raise an interesting problem with peoples’ approach to temperature. I’ve had lots of awful siphons served to me. More bad ones than good ones. The biggest flaw in all of them: they taste like they were brewed to damn hot. I was absolutely boggled when I saw a barista describe to a roaster how to make a siphon: “you want to keep that flame up so the slurry reaches 205ºF.” …!!!! 205ºF!! That’s almost the slurry temperature you get when you dump coffee into boiling water! Let me reiterate: !!!!!

    As I mentioned in my response to Drumroaster above, I think we really ought to be talking about median temperature, rather than maximum slurry temperature. In a pourover environment, you might max out at 200ºF, but you are starting off at 190ºF or so when the water is introduced, with the slurry temperature rising constantly. Your median temperature is thus 195ºF, and in my experience, this is a good temperature to brew a siphon at, owing to the flat temperature profile (you don’t have to compensate for lost heat). I’ve become somewhat allergic to the 200ºF prescriptive temp. I have never seen any argument for it based on taste. Not that it’s a bad starting point, but… 200ºF, whether brew water temperature or slurry temperature, means different things in different contexts: shorter extractions will suffer less heat loss and will thus have a higher median temperature, longer extractions will lose more heat and therefor have a lower median temperature. Rather than stick thermal probes in our coffee and shoot for some prescriptive number, we should make coffee and TASTE IT. Make a coffee with “cooler” water. Make coffee with really hot water. Understand what temperature deficient and temperature excessive extractions taste like.  How do these relative faults shift when we change our brew time? Also important (maybe most important): how does it taste as it cools??

    James, as far as brew recipes go, this is my experience: finer grind, compensatory quicker extraction often yields a brighter, more aromatic cup. The same coffee, ground coarser with a consequently longer extraction, will bring out more body and sweetness, and often, more savory qualities. This is very general, but I think we see a shift in the balance between percolation and infusion. An analog in roasting would be a shift in balance between convective and conductive heat transfer (though these aren’t necessarily time dependent). I suppose, but only experientially, that the shorter (more percolation) extractions favor extraction of lighter (by molecular weight) compounds (like aromatics, acids), and longer extractions (more infusion) favor extraction heavier compounds (sugars, oligosaccharides, etc.). It’s like you are shifting the modality of a bell curve. So, while brew ratio is certainly a powerful tool for interpreting a coffee, I think extraction technique can’t be discounted as well.

  12. A few quick thoughts:

    I think it’s important to note that those “common attributes” of taste, while differentiated on cupping forms and such, do not exist independently from each other. That said, more and more, I’m finding myself focused on sweetness as the primary focus in my brewing, as well as in roasting. Brew to maximize sweetness, while keeping negative taste-attributes as minimal as possible. Roast to maximize sweetness, while not introducing “roasty/toasty” flavors. An absolute oversimplification to be sure, but it seems to be the most helpful and relevant foci.

    Why sweetness? Because acidity is easy to extract, sweetness is more challenging. Short extraction times produce sour cups not because they produce comparatively high acidity, but because the acidity is not supported by sweetness, which takes longer to extract. It’s more challenging because in the effort to extract more sweetness, we begin to encounter the unpleasant flavor constituents.

    Regarding infusion vs. percolation: I’ve been known to declare that “infusion” (a.k.a. full-immersion”) is effectively a MYTH, and that it does not exist. The only brewing method that really comes close to exhibiting the brew-dynamic that “infusion” purports to be is cupping. It’s sort of a long story as to why I believe this, so I’ll leave it at that for now.

    Regarding agitation, “How we pour, where we pour, from how high we pour” is indeed important and definitely more relevant than looking at it as stirring-as-agitation. However, if evenness in brewing is a fundamental value (as I believe it to be), then the first step is to analyze the brew for the agitation which occurs without active participation from the person. In other words, you need to look at the flow of water through the bed due to gravity and fluid-dyanamics. Then your goal is identified: to pour the water to come as close to possible to that flow.

    The element which nobody seems to want to discuss (anymore) is geometry of the coffee bed. There is geometry and brew-dynamics which corresponds with evenness in brewing, and geometry and brew-dynamics which are inherently uneven.

    Finally, the way grind-size effects brewing is something I’ve been giving special attention to lately. I’ll throw this idea into the mix: in filter brewing, you should always grind as coarse as you can get away with. If your choice is grind finer or dose more coffee in order to achieve the brew time you want, dosing more coffee will almost always yield better results.

  13. First off, Kudos James! Great post.  
    Nos, can we not out rule that each coffee is roasted differently, even if it is the same coffee?  Therefore it will brew differently?  Like a barista pulling a shot of “X Roaster Blend” at “X Roaster Cafe” vs another wholesale account  where there is barista pulling that same blend at their cafe.  The shots will be different.  So, lets that 5 roasters all roasted the same Single Origin Sidamo Moredocofe, and sent it out to their pertaining cafes, wouldn’t that same coffee taste different from roaster to roaster and brew method to brew method despite it being the same coffee? Also, to add to the siphon discussion.  I too have been unhappy with too hot coffee from siphon.  I have to wait 15 minutes to sip.  So I too began to lower the temperature and elongated the time of brew.  I found it to be a little less sour and more enjoyable.  But at that point i was almost mimicking the clarity or a pour over. So, then I thought Siphon is a big show.  Can we actually come to a consensus on how Siphon should be brewed ideally in regards to temperature?  Don’t  get me wrong, it’s the coolest way to see coffee brewed.  Not too long ago, I started wondering if we are over looking the processing of coffee?  Natural, Honey, Washed, etc.Do certain process shine better under different brew methods?  I’ll leave this question simple as is. Thoughts anyone?

  14. So, after reading this, I spent my sunday brewing the same coffee (Suko Quto washed Sidamo Grade 2) using the same parameters as far as possible (60g/l, 88°C, 1:30min steep chosen rather randomly in an attempt to not disadvantage any method, same grind which was a bit of a guess, too) in a French Press, an AeroPress and a syphon.

    While I do follow your theory about how – metal-, paper- and cloth-filtering aside – the coffee shouldn’t taste too different, I simply couldn’t put it into practice on this first run. The three cups tasted vastly different, the differences being what you’d expect from the different methods: sturdy, tasty, dirty French Press; crisp, acidic, clear syphon and obviously awesome all around Aeropress (okay, I may like the Aeropress a bit too much to be objective here).

    Bottom line: give it a shot, though… while I couldn’t reproduce the same taste across the different methods, it certainly helped me to nail down the brew-method-specific effects on taste and texture a bit to give me more information to decide in the future which device to put a coffee into if I want to achieve something specific.

    Maybe someone more skilled in brewing coffee could get them to taste more similarly, though. I’d be happy if anyone has any pointers on how to re-do this experiment.

  15. I’n by no means “skilled in brewing” but here’s a pointer… I ran an experiment over the course of a few months, aiming to see if beans could be matched to brew methods (based on their inherent characteristics). I fairly quickly realised that unless you build in some form of brewing control to ensure you’re comparing like with like then the comparison is meaningless. Using a TDS meter (even with its accuracy limitations) allows you to adopt the SCAA/SCAE brewing control chart, which enables you to reject any brews that are not comparible due to significantly different TDS and extraction yield. It doesn’t automatically give you the skill to hit a consistent TDS and extraction, but it gives you a greater ability to learn to do so more often. Hope that is helpful in some small way :)

  16. Uneven brewing is a problem because it is so hard to measure – outside of taste.

    It seems to me that many of 3W baristas today spend too much time trying to impress others with scientific terms and supposed scientific methods – when all our guests want is something great to drink.  To my mind, taste is the ultimate measure.

    In many respects, I liken our craft to that of cooking. There’s definitely a science behind what happens in the pan, and one can take things to the extreme with precise temperature baths and “molecular” techniques, but at the end of the service, we’re just trying to provide something tasty and delicious.

    At Spro, we work with many different brew methods on a daily basis, giving us a unique perspective on how these methods (pourover, vac pot, press, eva, clever, aeropress, chemex, cold tower, espresso, viet) affect the flavors of individual coffees.  We also have the interesting journey of seeing how different filters affect each method as well.  One thing I have come to conclusion on is that I don’t much care for metal filters for Chemex, Vac Pot,  and the like.  Of course, I prefer the more traditional approach to the methods (meaning original paper or cloth filters).

    Returning to the subject at hand, much of what we do is as instinctual and “artistic” (or craft-based) as a cook. One develops a feel for the craft and the brew and can make minor (or even major) adjustments to the brew mid-cycle as conditions warrant.  To my mind, there’s more to what we do that narrowing it down to exact science.  And while perhaps it’s good to understand that science, I don’t let it overshadow what we do for a living.

  17. “It seems to me that many of 3W baristas today spend too much time trying
    to impress others with scientific terms and supposed scientific methods
    – when all our guests want is something great to drink.  To my mind,
    taste is the ultimate measure.”

    I’ve yet to meet a barista who valued “impressing others with scientific terms and supposed scientific methods” over taste. To the contrary, everyone I’ve interacted with understands that science and careful measurement are simply a means to an end, great tasting coffee.

    On rare occasions I HAVE met coffee people who are insecure about their ability to do simple math and/or make simple measurements with accuracy. Sometimes these people try and overcompensate by railing against the science types and bragging about their own golden taste buds.

    But no matter; whatever one’s preconceptions, consistency, care, open-mindedness and respect for the coffee will win out over all our coffee delusions.

  18. Thanks for this thought-provoking post.

    I have been thinking lately about how/if temperature plays a significant role in extraction after the first 1/3 or 1/2 of the total time/water.  I have no doubts that the water temp is crucial in the early stages of brewing, but does temperature loss matter as much nearing the end of the brewing process, or does it serve more simply as a  balancing out of the ratio in the cup?

    I’ve been playing around with tasting the extraction in thirds of a pour-over brew. I’m starting to think that the last 1/3 may be unnecessary extraction and too risky in contributing unwanted bitters and negative attributes to the cup. This has made me want to try two things. The first is to consider brewing 2/3 of the appropriate water volume per dose, and adding fresh water to the final batch. The second is dosing higher and shortening my time just a bit. (My usual go to is the fairly standard 60g/L, up and down from there to taste).

    Anyway, just some thoughts I’ve had recently as I’ve been considering stripping down everything I’ve “known” about brewing and looking at starting at zero again. Thanks again.

Submit a comment