Mano Lite: A Short Guide to Dialing in Espresso SOs and Blends - Page 2

Beginner and pro baristas share tips and tricks.
User avatar
Team HB

Postby another_jim » Apr 18, 2010, 2:47 pm

Oops, corrected the graphs, and trundling off for the Alzheimer's test.

I do list dark and light stopping shots as a Mano variable. I should have stressed that this is relative to ones normal practice of picking a blonding point. For instance, I've seen Heather Perry run shots to a very bleached blond if she doesn't like the look of the crema at a more conventional stopping point. Perhaps I should have said this more simply: one should stop by considering both the degree of blonding in the flow, and the texture of the crema in the cup. Doing this may extend the times of already slow ristretto shots to add some frothiness, and shorten the time of already fast normale shots to cut it.

The same is true for ristretto/normale (I don't think anyone pulls lungos). I was assuming everyone has a "standard shot" (mine being somewhere between a classic ristretto and normale), and if needed, they can go to slower or faster flow rates, less and more volume.

My instinct is to do the later timing decisions at the margins, so they it don't interact too strongly with previous dosing and flow decisions; Chris seems to be much more willing to take it further. But I did say these are intermediate level instructions; after a certain point, systems and rules need to be replaced by experience formed instinct, and instructions like these become moot.

This may be why such guides are absent among pros. Obviously, there are lots of "PBTC pros" who could use guidance; but it would be rude to address them to the professional community in general. I assume SCAA training courses have information like this.
Jim Schulman

User avatar

Postby AndyS » Apr 18, 2010, 7:05 pm

another_jim wrote:I do list dark and light stopping shots as a Mano variable.

Sorry, I didn't mean to say that you didn't list it as a Mano variable. I just meant to say that I find it to be a control variable for flavor balance as well as texture.
VST refractometer/filter basket beta tester, no financial interest in the company

User avatar
Team HB

Postby another_jim » Apr 18, 2010, 8:17 pm

I agree it's a gap in this system. I just hope it's minor.

Crema definitely affects both taste and mouthfeel. I was surprised by, and haven't digested the posts by Klaus Thommen and James Hoffman about crema taste. In my tries, the crema sometimes did taste dry and bitterish as they reported, but sometimes it tasted pleasing. My thinking is that this is not quite a question of flavor, nor of texture, but a series of poorly understood tactile effects similar to the clean cut effect of some astringency (spearmint) or the crisp feel of some acids (citrus). I'm suggesting that "dusty," "dry," "opaque," "dull" may be bad tactile effects like this rather than bad flavors. They certainly play a far greater role in espresso than in brewed coffee, and I think the flavor/texture of the crema is responsible.

Running shots extra long, or cutting them short, is a somewhat lame approach to controlling all this, but for now, I don't know anything that would be consistently better.
Jim Schulman


Postby CoffeeOwl » Apr 19, 2010, 5:32 pm

It is very nice work, thanks.

I have but one comment, I think you should make clear what you adjust to have the effect that you observe on taste. What I mean is: if you have your coffee dialed in for a normale, at a certain grind and dose to get a certain volume stoppable at certain color, then if you run another shot just longer, it will not correspond to the observation you make. I think it needs to be clearly mentioned otherwise there will be confusion. What needs to be changed for then making the changes to the variables you describe and get the result you mention is the grind. One changes the grind a tad and then for example changes the dose expecting to get the same volume and stop at the same color - then it will produce the result you mention.

And another point worth mentioning is that one can adjust the grind for a lower dose shot, which will yield less volume stopped at the same stop color as the higher dose&volume shot and will have all the taste characteristic of the previous shot, only it will be less to drink (less volume).

another_jim wrote: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.

I usually ignore the shot volume as variable and play with dose and stop color. Will check with the volume.
'a a ha sha sa ma!

LMWDP #199

User avatar
Team HB

Postby another_jim » Apr 19, 2010, 6:30 pm

CoffeeOwl wrote:I think you should make clear what you adjust to have the effect that you observe on taste.

I thought that was what I was doing.

I think it needs to be clearly mentioned otherwise there will be confusion. What needs to be changed for then making the changes to the variables you describe and get the result you mention is the grind.

This is an intermediate level instruction. It assumes that the reader can go from say, a 16 gram, 27 second shot, to a 14 gram 33 second shot without much difficulty or thought. If the reader doesn't know how to do this, they need to first refine their shot making skills.
Jim Schulman


Postby nbarth » Apr 22, 2010, 4:32 am

Thanks Jim for another excellent (and very practical) guide.

Your discussion of how machine parameters affect outcome parameters and changing your "mental coordinates" is very much how photographic exposure works, which may be instructive. Also of interest may be this essay by Bruce Tognazzini on how well-designed machines align these axes (search on "35 mm") - we can dream of (super-duper-auto) espresso machines where you push the "more fruit" button or "more chocolate" button!

To summarize:
  • key parameters are dose, grind, and cut
  • ...with dose + grind (+ cut, a little) = length (shot volume)
  • (you don't include temperature and pressure, though temperature seems to have a wider acceptable range, while pressure is just right or wrong)
  • the flavors are acids, caramels, and bitters
You give mano -> taste as:
  • downdosing increases caramels, decreases acids and bitters (and conversely)
  • ristretto decreases acids, increases bitters
  • cut affects primarily crema, hence mouthfeel and flavor in a complex way
You don't give the opposite taste -> mano direction explicitly, but I think that's the real point, right? I.e., given a particular flavor, what should you adjust?

You illustrate it, and I'll explain as I understand, but perhaps you could give your own guide?

In terms of desired flavors:
  • if dull (lacking acid and bitter), increase dose
  • if acid-bitter (ow!), decrease dose
  • if too acidic, decrease dose or shorten length (grind finer)
  • if too bitter, decrease dose or increase length (grind coarser)
  • to bring out fruit, increase dose or increase length (grind coarser) - increases acids
  • to make sweeter (chocolate, caramel), decrease dose
  • to make bitterer, add Campari
  • (alternatively, increase dose or decrease length (grind finer))
In terms of which to do:
  • primarily modify dose, secondarily length
  • in general you'll want to modify both at once - you demonstrate increasing dose and increasing length (grind coarser), which both work to bring out fruit, but offset to avoid increasing bitters
  • sometimes the way forward is unclear ("la diritta via [è] smarrita"), since different parameters can have the same desired main effect, but different secondary effects - you demonstrate with ambiguity regarding reducing lemon but keeping cherry
  • ultimately, you are limited by the beans in what flavors exist and what proportions of flavors can be expressed (esp. due to different flavors that extract similarly).

This also provides a guide for understanding the "triple ristrettos" which are popular in Cascadia - the high dose is acidic, with the short length (fine grind) counteracting this, so it's not too acidic. However, the result is less sweet and more bitter than a typical Italian, which has a lower dose and is longer (coarser grind); hence the "more pronounced fruit (?), less balanced" of this sodden clime.

By way of my own example, I've currently got some "Haile Selassie" Sidamo, Ethiopian, Natural process beans (sourced by The Coffee Shrub, roasted by Sterling Coffee Roasters of Portland), which has strong blueberry notes. At the roaster's bar, the blueberry was pronounced and delightful, but I didn't get much caramel/sweetness. At home, with much lower dose (13 grams), I got the same blueberry, but much more chocolate and overall rounded and balanced flavor ("chocolate-covered blueberries").

I assume you're drawing on your seminal Some Aspects of Espresso Extraction (2007), where you identify two kinds of bitters - fast-extracting (Mallaird) and slow-extracting (dry distillates) (beyond that, fruit = fast-extracting, caramel = medium-extracting). Perhaps you could elaborate on the connection, specifically which bitters you're referring to in this guide? (I'm assuming mostly dry distillates, as you state that Mallaird completely extract regardless, but maybe you mean a bit of both?)

Thanks again Jim - hope I've understood what you're saying and not butchered it too much!

User avatar
Team HB

Postby another_jim » Apr 22, 2010, 6:01 pm

Nice summary!

I didn't distinguish the kinds of bitters here, but stick to the empirical relations. The paper has problems, and needs to go back to the drawing board. The effect of dose on extraction numbers change from grinder to grinder and machine to machine, while its effect on taste seems to remain the same. This means the classification I use, of tastes based on the degree of extraction, is flawed. There are several dozen different tongue receptors for protein fragments, proteins, long chain organic acids, etc. that classify as "bitter" or "savory." So the story in how they respond to extraction changes is probably more complicated than for acids, sugars and salts.
Jim Schulman

User avatar

Postby malachi » Apr 22, 2010, 8:42 pm

nbarth wrote:[*] (you don't include temperature and pressure, though temperature seems to have a wider acceptable range, while pressure is just right or wrong)

Actually.... pressure has a wide range as well and in addition both temp and pressure can either be static or variable.

And - of course - the primary variable (unmentioned in your summary) is the coffee (with its own set of variables including age, roast, cultivar, processing method, etc).
"Taste is the only morality." -- John Ruskin


Postby Aaron » Apr 23, 2010, 5:17 am

malachi wrote:Nice work.

My only feedback would be that I'd suggest doing a specific cut against naturals, pulped naturals and washed coffees. In looking at this and comparing against my experience I think you might find it's going to be most consistently helpful across all coffee with certain naturals and many pulped naturals, slightly less consistent with other naturals, and least consistent with washed coffees.

But I could be wrong.

Malachi - what do you mean by pulped naturals and washed coffees? Does this have something to do with the way the coffee bean is prepared before being roasted? How does one know the difference when buying a blend or a SO?
“The powers of a man's mind are proportionate to the quantity of coffee he drinks” - James McKintosh

User avatar

Postby malachi » Apr 23, 2010, 12:41 pm

they are different methods of processing green coffee post-harvest.
good roasters will always tell you what the processing method for a coffee is (washed, natural, pulped natural, semi washed, aqua pulped, whatever description they use for the method).
"Taste is the only morality." -- John Ruskin