Background:
We run a one-grouphead Synesso espresso machine (purchased 2015), which allows for three pressure settings to be set (low, medium, high - e.g. 3bar, 5bar, 9bar), and for four timed stages to be configured (stage1 is the duration of low pressure, followed by stage2 which is the duration of medium pressure, followed by stage3 which is the duration of high pressure, followed by stage4 which is another duration of medium pressure). We allow baristas to mess with the durations of stages 1-4, temperature of the grouphead boiler, and grinder grind-size, at their ease.
This training note was provided to an intermediate level barista who has used to machine for about a year, without really discussing its usage in great detail. The barista was a visual design student, and it occurred to me after writing the note that espresso calibration is really a design problem that needs to be resolved (!) on a sometimes-hourly basis. So what we're discussing below actually applies to searching for viable design solutions in a multivariable space. How do you do this efficiently?
I'm looking at your team logs now... last 36 hours or so.
You mentioned different preinfusions (stages 1/2)... I looked and saw the range was:
(a)
S1: 8.0s
S2: 0.0s
(b)
S1: 5.5s
S2: 2.5s
[S1-S3: total 26-28seconds]
So, I will point out that (a) and (b) are actually pretty close together. (S1 is around 3.0bar, and S2 is around 5.0 bar... so... taking 3 seconds of S1 and moving it to S2 is literally adding 2.0bar of pressure for 3.0seconds.)
Now cooking [at 1bar for 10seconds] and cooking [at 10bar for 1 second] is not exactly the same thing. But if we pretend for discussion's sake that it's a useful metric... then the difference between (a) and (b) is less than adding 1second of cooking time at 9bar. Hehe.
Sure, fiddling with such small differences will alter flavour slightly... (and this is just my intuition based on experience so far) but it's probably equivalent to just knocking the temperature around by 0.5C.
And knocking the temperature around is actually wayyyyyy easier than messing with the pressure profile. E.g. if the difference between (a) and (b) is like 0.5C difference, and we call it "one small step", then it is actually much simpler to jump "4-5 small steps" by just bumping the temperature up by 2.0C...
.... why would you want to jump 4-5 small steps instead of just one?
Well the advice I have is, if you look at the log, everyone's been calibrating within a very small range of the universe. One step left, one step right, etc. No, stop that, if you're quite experienced, taking a greater leap (of faith? whatever you want to call it)... sends you scouting in unknown territory!
Once you get used to taking larger jumps, you may find that experience will help you calibrate faster because you can cover a greater space of variables more quickly.
... bringing it back to the Wahana at hand:
So far it seems (to me):
- sweetness is good
- acidity is good
- no foul smells
- no massive bitterness
- not much body either...
The "chocolate" note is quite light, and not very heavy. Do we want more? Maybe try big jumps like:
- 1-2 notches finer
- 1.5 - 3.0 degrees hotter
- 15-30% increase in brew ratio
- e.g. +10% is [ (going from 24g out to 26.4g out) / (over 15g in) ]
- e.g. +20% is [ (going from 24g out to 28.8g out) / (over 15g in) ]
- e.g. +30% is [ (going from 24g out to 31.2g out) / (over 15g in) ]
Bearing in mind that:
- Messing with the grindsize will always make the biggest difference to overall flavour, and irreversible (cannot undo).
- Messing with the temperature is reversible, and quite precise thanks to our digital temperature settings.
- Messing with the brew ratio is less precise and less reversible than messing with the temperature
So generally, my preference/advice with this particular machine is:
- Set stage 1 to exactly what it takes to wet the bottom of the puck.
- Set stage 2 to zero; except when stage 1 is taking too long, then convert X seconds from S1 to S2.
- If you're looking for small, precise changes to flavour, mess with the temperature.
- If you're looking for large, grand changes, mess with the grindsize.
🤷🏻♂
May the odds be in your favour.
PS. Reading the logs and demanding details from others may help. Only if you like words hehe. Not everyone does, I know. But I encourage it.
