Imagine two baristas pulling morning shots to "dial in"
their equipment. After a few pulls, the espresso is still wanting. The
conversation might go like this:
Barista 1: Whoa, what was that? It's harsh man,
the grind is off.
Barista 2: Maybe you're right, it was a slow pour. Let me
loosen it up.
...one minute later...
Barista 1: Better. But a little sour. Did you futz with the
Barista 2: Me? No, I checked it this morning.
Barista 1: Well, it's still a little sour. I wasn't
looking closely, did you dose the same amount? That pour looked too fast
Barista 2: Hey, I dosed like always. You want to show me how
it's done smart guy?
All baristas, if pressed, will admit that sometimes they are guessing
what went wrong with an extraction. So following the adage, "wise
baristas learn from their mistakes," this section will focus on
intentionally bad extractions. Please pause for a moment to reflect on
the many wonderful beans that were sacrificed to bring you this learning
Loss of adhesion
leading to channeling
Our study begins with one of the more common extraction flaws, as
shown in the image to the left.
To readily produce this shot, I increased the brew pressure to 10 bar
and intentionally canted the tamp with one side of the puck visibly
higher than the other by about three millimeters. The tiger-striping is
gorgeous on the left-hand side and the beautiful coloring held there
until the end, but trouble began very early on the right-hand side. The
photo was taken approximately seventeen seconds into the pull.
The channeling on the right side is likely due to broken adhesion
between the walls of the basket and the puck. In addition to canting,
hard or excessive tapping to knock grinds off the side of the basket
during tamping can cause this type of channeling. It is more difficult
to detect this kind of channeling than catastrophic failures since the
stream would look passable if viewed from a normal pour spout. Without
the bottomless portafilter, the scant evidence of a problem would be a
thin blond thread in an otherwise medium to dark stream. You may also
see twirling "barber pole" striping in such pours. The shot
itself was drinkable by average cafe standards; it did however suffer
from lower body and reduced sweetness.
The Blond Gusher
Blond gusher = pours too
The next example is regrettably the most common mistake you'll
encounter in commercial establishments: The infamous fifteen second
To produce this blond gusher, I reduced the pressure to 8.5 bar and
moved the grinder adjustment two clicks coarser. The tamp was not
intentionally flawed and you can see that overall the extraction is
even. The photo was taken at approximately twelve seconds into the pour.
The cone has already formed and is growing at this stage; a correct pour
would still require several more seconds until the cone reached this
size. In addition to a grind setting that is too coarse, extremely high
temperatures will accelerate blonding, as will happen if the barista
forgets to perform a cooling flush before the extraction.
High speed pours are invariably thin-bodied, sour, and weak. If
you're served a shot like this at a cafe, you may be tempted to
blame low temperature for the sourness, but quick pulls are more often
the real cause. It's very difficult to distinguish a low brew
temperature espresso from a blond gusher by taste alone.
(A Brief Intermission for the Test of Thirds)
A good way to roughly categorize the effects an extraction fault
introduces is to do a proper extraction divided into three parts, what I
call the "test of thirds." Improper extractions exhibit a
stronger tendency in one of these three parts, which throws the rest of
the shot's qualities out of balance. Therefore recognizing the
taste characteristics of a divided espresso will give you important
clues about the stage in which the extraction went astray.
The first third of a pour is the most pungent and gives the espresso
its "punch" in the same way that the higher percentage of
cocoa defines the character of dark chocolate. Depending on the initial
starting temperature, the first third can also tend towards sour
flavors. The second third of an espresso pour is the crowd pleaser. In
fact, some baristas intentionally let the first few seconds of the pour
fall into the driptray to emphasize the sweet and creamy nature of the
second third, but in doing so you'll potentially sacrifice the
character that makes the blend interesting. The last third is weaker,
lacks body and sweetness, and in some blends tends towards bitter
flavors as more caffeine is extracted. Some baristas are tempted to end
the pour well before the last third is complete and before the onset of
blonding, but skilled professionals often let it flow longer than
newcomers, knowing that the last portion can add a pleasing complexity
to the shot.
When everything goes wrong
Returning to the tour, please direct your attention to the
disastrously bad extraction shown on the left.
To produce this nightmarish pour, I returned the grind to the same
setting as the first in the series, but intentionally distributed the
grounds unevenly, tapped the portafilter against the countertop, and
tamped with scant attention to proper level. In the hopes of producing a
really good jet stream, I also raised the brew pressure to 11 bar. My
reward is the wonderful "dot dot dot" captured in the photo
as a fine mist sprayed out from a pinhole in the bottom of the puck.
If you look carefully near the right edge of the photo, you'll
see side channeling that establishes a new low point in extraction
mishaps. The leak spewed out with such force that the channeling stream
overran onto the side of the portafilter and missed the cup completely.
I was surprised the camera lens didn't get sprayed! The front of
the espresso machine and parts of driptray were covered with teeny droplets of
coffee. Perhaps safety glasses are wise for the first few uses of a
One bit of common advice offered to those reporting poor shots is to
reduce the brew pressure. But based on the bottomless portafilter
extraction analysis, I suspect in many cases those blaming excessive
extraction pressure for lackluster shots are actually seeing the
consequences of increased channeling due to the higher pressure. That
is, the prevailing advice offered to inexperienced baristas to improve
shot quality by lowering brew pressure may be more about reduced
channeling than a genuine comparison of a proper extraction at a lower
versus higher pressure.
Too Much Information
Beware of T.M.I.
One drawback of the bottomless portafilter is it can provide too much
information. The image to the right is an example of analysis taken a
step too far. Highlighted is a "dead spot" where no coffee
is coming out and a light reflection that tarnished the appearance of an
otherwise beautiful extraction, but the taste was as good as the double
shown at the beginning of this article. Focusing too much on the process
can blind you to the goal—enjoying better espresso!
The bottomless portafilter is a handy diagnostic aid, especially for
espresso machines that demand careful attention to technique. But there
are limitations to what it reveals since not all channeling is visible
from the bottom of the basket. Intuitively one might think of channeling
as a "hole" that runs vertically through the puck, but
uneven water pathways aren't always straight conduits from top to
bottom. Fissures can be transversal between layers in the puck; you can
produce impressive horizontal channeling by hard packing a half-dosed
portafilter, dosing the remainder, and then lightly tamping. The upper
layer of coffee will be less extracted than the lower layer below.
Visible evidence of what is occurring will be absent from the vantage
point a bottomless portafilter offers — the giveaway is extraction
coloring that begins dark, the tiger-striping ends early, and boom! A
sudden uniform blonding about fifteen seconds into the pour.
Now allow me to introduce professional barista Chris Tacy as
"guest author" to close out this how-to with his perspective
on what really matters to the pros.