Thursday, 28 June 2012

AeroPress, Hario Mini Mill Grinder, and Gold Cup Extraction

At the time of writing this, the most-read post on here is one I wrote quite early in this self-education process. It covered my early attempts to find the best grind setting with my Hario Mini Mill when brewing with the AeroPress.  I think it's a really important combination of brewing equipment, since the AeroPress is touted as such a simple way to get a clean cup without an expensive pouring kettle... and also because low cost hand grinders are possibly the best way to get more people drinking fresh coffee in their own home.
When I wrote my previous post confounding factors got in the way, such as:
  1. my inexperience with the AeroPress
  2. being new to the grinder
  3. having only a basic understanding of brewing (chemistry etc)
  4. and the biggest problem - having no frame of reference other than repetition, trial and error to guide my tasting towards an end result that seemed perhaps acceptable.
So I'm now a little older, and (perhaps) a little wiser in the ways of coffee, and for some time I've had a nagging desire to go back to the AeroPress/hand grinder combo and try again.  Then the other day someone asked my to give them some advice regarding using the AeroPress and a Porlex, which seemed like too good an opportunity to ignore.  I'm also planning to have the AeroPress as a brewing method in my forthcoming cafe, and to help customers learn how to use their own brewer to make tasty coffees back in their own home, with a hand grinder.  So an updated blog post is probably warranted.

VST LAB refractometer

The great thing is that I have a refractometer now, which is a massive help with overcoming problem #4 from above.  I can experiment until I find the grind setting, temperature, steep time, degree of agitation etc etc that gets me into the Gold Cup zone, and from there start refining based on taste.

My starting technique was:
14g coffee
Mini Mill setting of 5 clicks back from the 'fully closed' position
230g water at 80 degrees C
Fill, agitate gently at 30, 60 and 90 secs, then plunge slowly.

My first three attempts... changing grind setting from 5 to 6 and back to 5 again, agitating more, raising water temperature a little... were all under-extracted, coming in at a concentration strength (TDS) of 1.03%, 0.97% and 1.10% (I'm aiming for 1.30% TDS, giving an extraction yield of 19%). This was really surprising as the grind setting is really as fine as you would ever want it for the AeroPress and any finer would just clog the filter.

Perhaps the grind quality from the Hario Mini Mill isn't helping the extraction process take place effectively.

So anyway, I needed to raise the TDS without changing the brewing ratio.

A degree of success
I raised the water temperature to 90C (1 min 20 from boil with the kettle lid opened).
Half fill, swirl.
Fill to 230g.
At 30s sink crust. At 1 min stir left, right, left, right. 1m30s repeat. 2min lid on and plunge. 2m30s done.

The TDS of this brew came in at 1.25%, with an extraction yield of 18.26%. Statistically a good extraction, within the Gold Cup range.
It tasted ok... rounded, balanced, inoffensive and quite enjoyable. But very little in terms of aroma and flavour. I expect more from this fantastic coffee... Nicaragua Finca San Jose... so something is still not right and again I suspect that this grinder is part of the problem.

Perhaps if I grind coarser and increase steep time and agitation further...? But this isn't the direction I particularly want to take as I don't want to turn the AeroPress into primarily an immersion system, which is what would happen if the grinds are so course that there is very little resistance when plunging. Nonetheless, let's try it...

Interlude: Spot the 'deliberate' mistake...
At 7:30am today I tried to make a coffee. In my bleary, tired state this happened. I suspect it's something that most people have done at some point... at least I hope it isn't just me!!

Reaching the limit
Keeping the grind the same (5 on the Mini Mill) I increased temperature to around 95C, just a few moments off the boil.  I extended the steep time to 3:30 minutes, with a further 30 seconds for plunging (total brew time 4 mins).
Well the taste was better. More rounded, more complex, more acidity. Still a little lacking in depth of aroma though.  Damn! TDS = 1.26%, giving an extraction yield of 18.41%.  So close to my target, but I'm amazed that almost doubling the steep time only increased the concentration of the coffee by one hundredth of a percent!
Something else is at play here. It must be.  And when that happens then the first place I always look is Water. I'm using bottled water - Highland Spring. What could there be about this water that might inhibit the extraction process?
Having read the SCAA Water Quality Handbook I have no real issues with any of these figures except one: "Bicarbonate 150mg/L". This is a measure of Total Alkilinity. The SCAA Standards for this are 40mg/L (ideal), with an acceptable range of 10-100mg/L.

Why does this matter? Because the higher levels of alkilinity in the water does two things. First the alkilinity neutralizes the acidity inherent within the coffee, reducing flavours and making it taste 'flat'. Secondly (according to the Handbook) it makes the coffee particles expand more, which slows down the movement of water in the brewing process, increasing the steep time. (This second statement is open to interpretation though. Personally I take it to mean that more time is needed to achieve a target extraction, but I could be wrong.)

So I do believe the Highland Spring water, despite being better for coffee brewing than most bottled waters, is limiting how well I can brew with the AeroPress... or indeed with any brew method.
In a bid to check this I changed to my Mahlkoenig Guatemala grinder and made a brew with the AeroPress, attempting to use a similar size grind. I ended up using a slightly coarser grind, which will have reduced the TDS of the final brew, but the results seemed to confirm my thoughts regarding the water. Using the same method as above, with a total brew time of 4 mins.  TDS came in at 1.18%, giving an extraction yield of 17.23% ... outside the Gold Cup range. It actually tasted better than the Mini Mill brews that WERE in the Gold Cup range, but clearly there is something hindering extraction other than the grinder, grind size, water temperature, and other typical brew parameters... water.

**Update** a few days later I found a water source with an identical TDS and a calcium hardness level of 61mg/L - statistically very good from a hardness persepctive. It has a pH of around 8mg/L... a little too high... and a sodium content of around 11mg/L, almost double the Highland Spring, but close to the SCAA target.  I haven't yet established the alkilinity levels. Anyway, this different water, with the an almost identical TDS but differences in OTHER water characteristics, had a profoundly different effect upon the extraction process. The result was a BIG difference to both the extraction stats and the tastes and flavours.  TDS went up to 1.37%, and extraction yield to 20.04%.  The flavours were more pronounced, which is great, and it was ok to drink. Unfortunately it now tasted slightly over-extracted to me. There were hints of (perhaps) acids that didn't enhance the flavour.  I had now left it to brew for too long!

So I guess here's one way to get a fairly good Gold Cup extraction from the AeroPress using a Mini Mill. I'm sure there are others, and if you find a good approach please let me know :)

Extraction Yield 18.41 %
TDS 1.26 %
Dose Weight 14.0 g
Brew Water Weight 230 g

Coffee Details
Brew Water Start Temp 95°C
Brew Time I think this is a critical factor defined by your water chemistry. See comments below.
Roaster Name
Roast Style
Grinder Hario Mini Mill
Grind Setting 5 Clicks

14g coffee
Minimill setting 5. Quite Fine.
230g Water 95C, just off boil.
Inverted AeroPress.
Insert grinds. Half fill with water. Stir well to ensure pre-infusion. Fill to 230g (to the rim).
Stir left, right, left, right at 30secs, 1min, 2min, 3min. Fit the filter. NB - Steep Time could range anywhere from 1 minute to up to 4 minutes.  I suggest a 2 minute brew time and then a 30 second plunge, and on subsequent brews either increase or decrease the time if it tastes too weak or too strong.  

At the 2012 Nordic Barista Cup event Vince Fedele of VST Inc, who designed and manufacture the VST Coffee Refractometer, made in important announcement regarding coffee brewing using Gold Cup standards. The announcement directly affects brewing with the AeroPress. When you use infusion/immersion brewing methods (as I am using the AeroPress above) and then filter it, part of the coffee liquor is retained in the grinds. This skews how the resulting extraction yield is presented. So what we may have previously thought was an 18.5% extraction is actually over 20%! To compensate for this it is necessary to increase the dose of coffee used in the brewing process. So looking back at the figures in my summary, the theory goes that I should actually have used a dose of 16g rather than 14g if I wanted to achieve an extraction of around 18.5%. That 2g will have a huge impact upon flavour, and it might help explain why throughout all the above brews, none of them tasted any better than merely 'good'.

1 comment:

  1. fantastic article Mike, thanks so much! I'll look forward to trying to achieve better results tomorrow, using this as my guide :)