Introduction

This article describes a simple method for adjusting the "mano variables," grind setting, coffee dose, and shot time, to achieve close to the best possible taste from any espresso blend or SO. It does not give instructions on how to adjust temperature or pressure in response to taste, nor what basket to select; instead it assumes these remain fixed. The method will not find the optimal mano variables for every coffee; but in my experience, it will get close to optimal for most.

It is addressed to intermediate baristas with enough experience to pull non-channeling shots, to stop them at a correct and consistent blonding point, to adjust their grinders and dose consistently, and to keep the pressure and temperature profiles of their machines fairly constant and within normal espresso bounds. Experienced baristas may be interested in the way this method classifies taste and mano variables, but it won't improve their skills in dialing in coffees.



The Mano Variables:

Grind setting, dose, and shot time are the actual physical adjustments a barista makes, but these do not translate well to taste changes. Instead, we need to reconceptualize the three physical adjustments into three cognitive variables that relate better to the taste (in math speak, it's doing a canonical correlation or factor analysis by rotating the axes of the variable space):

• Up/Normal/Down Dose: the dose of coffee required to either normally, over or underfill the basket, along with the grind setting required to get an intermediate shot, i.e. a double roughly 30 seconds long, 1.5 ounces in volume, 1 ounce in weight, stopped at the normal blonding point. The exact weight and grind setting needed to accomplish this will depend on type of basket, group, and grinder being used.
  
  
• Normale/Ristretto: the smaller grind and/or dose changes required to move from the settings above to either a ristretto shot, a double roughly 35 seconds long, 1 ounce in volume, 1/2 ounce in weight, stopped at the normal blonding point; or to a classic normale, a double roughly 25 seconds long, 2 ounces in volume, 1.5 ounces in weight, stopped at the normal blonding point.
  
  
• Pale/Dark stopped: Stopping shots adjusted as above earlier or later than normal, when the flow is correspondingly darker or lighter colored.

The Taste Variables:

Exact flavors, for instance, cherry or chocolate, are an aspect of the coffee itself, and cannot be altered. Neither can the specifics of the mouthfeel, the sparkle and astringency, the fine texture of the crema. However, the way these flavors and textures balance in the shot can be altered and fine tuned. As with the Mano variables, they have to be classified in a way that makes them amenable to control.

• Acids, Caramels, Bitters:
  • Acids: Fruit acids and fruit sugars, along with floral and herbal aromas.
    
  • Caramels: Chocolate, caramel, malt, and nut, along with cake baking in the oven aromas.
    
  • Bitters: Spice, wood, sherry, and citrus peel, along with smoke, tobacco and tar aromas.
  
  
• Foamy versus Oily Mouthfeel: When ideally balanced, the crema and liquid portions of the shot are almost folded together and fairly stable over the minute or so in which the espresso is consumed. Usually, if the crema is unbalanced, either in a thin oily layer, or with a pile of large bubbles, the drink will not be as pleasant feeling on the tongue, nor will it last as long.

The Relation of Mano to Taste

• Dose and Taste Loudness: Up dosed shots emphasize both extremes, the acids and the bitters, at the expense of the middle caramel flavors. Down dosed shots do the opposite, emphasizing the caramels at the expense of the bitters and sours.
  
  
  
• Shot Volume and Acid/Bitter balance: Classic normales, that is faster flowing, larger volume shots, emphasize the acidic flavors. Ristrettos, that is slower flowing, lower volume shots, emphasize the bitter flavors.
  
  
  
• Flow Color and Mouthfeel: Allowing the shot to go longer and paler creates a lighter texture, cutting it shorter and darker creates a heavier texture. Unfortunately, ristrettos are inherently heavier textured than normales, so the timing differences need to be exaggerated in order to compensate (e.g. the ristretto may need to run to platinum blonde and almost a gush in order to have a really fine texture)

In my experience, dose is more powerful than shot volume, and shot volume is more powerful than flow color. So I find making the adjustments in that order to be easiest. However, I don't know if this is just my habit or true for everyone.



Hope this has helped.

malachi wrote:I think you might find ... less consistency with washed coffees

I don't think washed coffees have as much buffering in the crema and heavier suspended particles; and getting the flow and stopping point right is more difficult -- three or four seconds more or less usually doesn't make much difference on a natural; and it can be huge on a washed coffee.

TrlstanC wrote:If you could give an example of what you do the next time you dial in an SO (or even just a tricky blend) it would be interesting to see what changes you make, and how many tries it takes you until you think you've got it down.

I'm currently running a blend of 50% Bourbon Antigua washed, 25% Typica Oriente pulp natural, and 25% Yellow Bourbon Brazil pulp natural. I'm looking for cherry, toasted almond, smoky dark caramels, not too sweet (think Florentine cookie). At a medium dose (15.5 grams on my setup) and shot time, I had one tasty shot and one where the cherry was missing, and the taste too opaque and dull.

Tomorrow I'll correct it by the book,

• dosing higher (up to 16.5 grams) to get away from the dullness,
• and pulling faster (by setting the grinder where I normally would do a 17 gram regular flow shot) to bring out the cherry.

If that doesn't get the target profile more reliably, I'll be scratching my head.

Peppersass wrote:Is it possible to do a similar guide for temperature variations, or are they too coffee-specific? For example, I've seen references to increasing temperature to bring out the sweetness of two big blends, Vivace Dolce and Counter Culture Aficionado. Is this typical of big blends or very specific to those two particular blends?

The conventional wisdom since Schomer started writing about it is that higher brew temperatures counteract acidity and work best for light roasted blends, whereas lower brew temperatures counteract bitterness and work best for dark roast blends. This is for the 90C to 95C range. I've heard that very low brew temperatures of around 85C can be used to get very sweet shots from ultra-light roasts like Terroir's North Italian (at least from a few years back). I am not sure how across the board this rule of thumb is.

I know excessive brew pressure is marked by low flow, and an oily bitterness (everything tastes like a bad ristretto), while too low brew pressure tastes weak and watery. What I don't know is what brew pressure variations do when they remain in the acceptable zone. The current leading edge is to investigate if a declining brew pressure creates more taste clarity, as one gets in many lever machines.

This is a bit heretical; but personally I'm hoping it'll be possible to design machines with fixed, factory set temperature and pressure profiles, with which one can get all the taste possibilities out of any given blend or SO by just changing grind, dose and shot time. That way, it wouldn't take a ton up to dial in.

another_jim wrote: ... I'm looking for cherry, toasted almond, smoky dark caramels, not too sweet (think Florentine cookie). At a medium dose (15.5 grams on my setup) and shot time, I had one tasty shot and one where the cherry was missing, and the taste too opaque and dull.

Tomorrow I'll correct it by the book,

• dosing higher (up to 16.5 grams) to get away from the dullness,
• and pulling faster (by setting the grinder where I normally would do a 17 gram regular flow shot) to bring out the cherry.

If that doesn't get the target profile more reliably, I'll be scratching my head.

The 16.5 gram normale shot worked out as expected in terms of flow, giving me two ounces in 25 seconds. The dullness was gone too, and there was plenty of cherry and almond, but there was also too much lemon. So back to the regular double and see what that is like.

But do I add 1/2 a gram to the dose or choke back on the grind? I wasn't sure. This year's iteration of Black Cat uses Guats and Brazils (my blend is a shameless knockoff), so I went over to the Coffee Studio to compare notes. They pull Black Cat, and do it a little mellower and brighter than Intelly itself, a style I slightly prefer. After waiting in line and hearing 3 cortado orders ahead of me (and nary a big milk drink in sight --I was floored), I ordered and got my espresso. It was a less acidic and a little gentler than what I had just pulled, but still with a vivid and clear taste -- better than my shot. So I decided no further updose, but tighter on the grind. The results was a nicely balanced and clear shot, a solid 3.5 to 4 in a barista competition. The lemon still emerged in the long aftertaste; and I may try going down to 16 grams from 16.5. If that turns out to be too dull again, I know I'll need to tweak the blend.

An addendum: as Chris has noted, I'm more experienced with African than Central coffees. What I'm finding is that in comparison, they seem more sensitive to flow changes, and I needed to get the timing more precise than usual. On the other hand, updosing increased the volume, but didn't send the coffee over the edge as happens with East Africans. So it could be that with these coffees, the shot time/volume is the first thing to get right, and tweaking the dose comes second.