Deliciously overextracted

Beginner and pro baristas share tips and tricks for making espresso.
User avatar
AssafL
Posts: 2588
Joined: 14 years ago

#1: Post by AssafL »

In his last post Scott Rao praises a system called ground control that can achieve 26% extractions. He hypothesizes that with a somewhat more unimodal grinder like the EK one can get to 28% extraction.

http://scottrao.com/blog/highlights-fro ... o-seattle/

He praises it due to the economic yield - which is obvious, but also for being a better brew - which is to me surprising.

Does over extraction only exist in traditional methods measured against brew control charts? Or does it also exist for new grinders and new extractions processes? If so at what yield does over extraction start?
Scraping away (slowly) at the tyranny of biases and dogma.

User avatar
Almico
Posts: 3612
Joined: 10 years ago

#2: Post by Almico »

I think the key word is 'over'. Like over done, over board, over exposed..., the very definition is that it is not desirable. This comment was telling:
Scott Rao wrote:Some of the brews were lovely: soft and sweet with little hint of overextraction. However, after tasting about 12 batches, it was clear that the brewer was inconsistent... often as much as 10% of the grounds would end up high and dry on the filter wall, the water delivery into the coffee bed was poorly designed for even extraction, and the agitation of the slurry often left the coffee bed quite lumpy during the drawdowns, causing channeling.
This thing sounds like a percolator, but with new water being introduced instead of recirculated coffee water.

I'm not convinced that higher extraction equates to a better flavor experience. Coffee has over 1000 chemical compounds including formaldehyde and arsenic. The good stuff extracts relatively easily, the bad stuff no so much. But if you try hard enough, you can get coffee to taste like turpentine.

I found his closing comment interesting:
One of the most surprising and ironic things about the Expo each year is how difficult it can be to find delicious coffee. I'm not alone in that opinion- at least 20 people echoed that sentiment to me personally this weekend. I tasted plenty of burnt, underdeveloped, and baked coffees on the show floor. Some of them may have been 88+ point coffees, but their roasting and extraction left much to be desired. None of us are perfect, myself included, but we certainly have a long way to go as an industry.
I love that he was 'forced' to use a Chemex.

User avatar
AssafL (original poster)
Posts: 2588
Joined: 14 years ago

#3: Post by AssafL (original poster) »

Over extraction I have no problem with. BCC would lead us to believe that EY at 26% should be very over extracted. Not just a "hint".

i also loved that the best coffee at the show was his clients. Perhaps a slight bias?
Scraping away (slowly) at the tyranny of biases and dogma.

User avatar
another_jim
Team HB
Posts: 13871
Joined: 19 years ago

#4: Post by another_jim »

The bad over extraction is due to hydrolysis, the breakdown of long chain molecules. For instance, when coffee is converted into instant coffee, the extraction rates (using high pressure, high heat, and long time periods) is up to 50%. Drink a little instant coffeee, and you'll know what bad over extraction tastes like. It's very hard to over extract when brewing, you'd need to boil the brew or steep/percolate it for long periods (a tradtional percolator can actually manage such over extraction). On an espresso machine it would take a fairly overheated and choked shot.

Barring that, higher extractions add body and mild caramels, a good thing unless the coffee is already soft tasting.

What percentage is bad over extraction? That's a very good question. The definition of extraction can be based on dissolved, colloidally suspended, or all solids in the coffee; and the literature varies on this. The older literature, which evaporates the coffee to get TDS and extraction cites around 25%; measures based on conductivity or refractive index cite around 22%. I've argued for physics based, evaporation measures long enough and won't go over it again; my point here is that the extraction numbers are largely undefined. In any case, the proper way to detect bad overextraction would be chemical, to test for hydrolysis or the unique compounds created by it. None of the self appointed authorities on coffee extraction is remotely qualified to do this. So draw your own conclusions.
Jim Schulman

User avatar
AssafL (original poster)
Posts: 2588
Joined: 14 years ago

#5: Post by AssafL (original poster) »

Thank you for the succinct answer.

So since it would seem to be yet another abuse of a refractometer to have a target yield perhaps one of the Near IR spectrometer's coming out could be programmed for such an analysis? (I do have one, but it is woefully inaccurate; can't accurately differentiate mundane 60 and 70% chocolate...; to think I purchased it on kickstarter in the hopes that it could analyze ice cream bases...).
Scraping away (slowly) at the tyranny of biases and dogma.

User avatar
another_jim
Team HB
Posts: 13871
Joined: 19 years ago

#6: Post by another_jim »

AssafL wrote:perhaps one of the Near IR spectrometer's coming out could be programmed for such an analysis?
You would have to describe how they work.

My take is that high extraction is a very poor correlate to the presence of hydrolyzed compounds in the cup. For instance: pass high pressure steam through a puck for 2 seconds, and the coffee bed will be have a very low extraction, while the resulting cup will taste awful due to the hydrolyzed compounds in it. The ideal test would simply be a paper strip impregnated with a reagent that changes color if the offensive compounds are present; like a pH or hardness strip. That would be simple and idiot proof.

Coffee professionals can do many things to make coffee better; but hard science and chemical engineering are not among them. So I think a lot of the stuff being said about extraction is "alternative science." That might win elections; but it won't make your coffee taste better.
Jim Schulman

User avatar
AssafL (original poster)
Posts: 2588
Joined: 14 years ago

#7: Post by AssafL (original poster) »

Well, the device I have is called SCIO from an Israeli company called Consumer Physics. It shines a bright light at whatever it is you are trying to analyze, measures the reflected light and plots a low resolution graph of light absorption (or reflection) vs. wavelength.

In theory, high resolution spectroscopy (big lab machines) can identify the chemical makeup (even then, assuming you more or less know what you are looking for).

It does have an SDK (at a cost) but I don't know what the "behind the scenes" resolution is. So I don't know if it is possible to write an analytical app that just detects the unwanted hydrolyzed chemistry. It also has a "teach it yourself" function, so perhaps one can get the chemistry of overextracted coffee by teaching it what a good pull is and juxtapositioning it with overextracted stuff (set boiler temp to 98C and choke the flow). In the former (analytical) model one would get a numerical value associated with the concentration of these components. In the latter model who knows where the threshold will lie between normal extraction and over extraction...

(BTW - I purchased it as I wanted it to do Ice Cream base analysis - identify % fat, % water and % solids. Needless to say it doesn't do anything close to that...).
Scraping away (slowly) at the tyranny of biases and dogma.

User avatar
another_jim
Team HB
Posts: 13871
Joined: 19 years ago

#8: Post by another_jim »

It sounds like the right idea, a mass spectrometer variant; but probably requires a much more expensive execution.
Jim Schulman

vogacoffee
Posts: 2
Joined: 8 years ago

#9: Post by vogacoffee »

These are some great points. Tom Chips was able to do a comprehensive review of Ground Control, where he discusses his experiences in tasting coffee brewed on our system: Voga Coffee- Ground Control .

The interesting thing about over-extraction, as another_jim mentions, is that really what you're concerned about is the breakdown of certain compounds, as well as the extraction of certain bitter compounds that tend to come out later in the brewing process (their extraction curves are significantly offset, as compared to those for most flavor-imparting compounds). By carefully controlling the length of time that any given volume of water is in contact with the grounds, Ground Control is able to simultaneously reduce the amount of undesirable bitterness in the cup, while increasing the flavor definition and nuance. It's the unique method, involving a series of mini-brews (substantially drying the grounds between each mini-brew), that allows us to achieve our results.

Our co-founder Jason Sarley discusses the science further, in this video:
eli
co-founder and ceo, voga coffee, makers of ground control

User avatar
AssafL (original poster)
Posts: 2588
Joined: 14 years ago

#10: Post by AssafL (original poster) »

Kudos on a very thoughtful and interesting system.

Too bad I no longer live in the Bay Area and can experience the system firsthand. Hopefully the numbers make it so that even in frugal Israel the machine will start popping up in local coffee shops.

One point I think Jim talked about (and I agree with generally - not just for coffee) is the lack of scientific substantiation for claims. I would extend that and state that for many coffee trials and tribulations - it is not just measurements - but even if those exists it is a lack of a base hypothesis to disprove.

I am not claiming this here (far too early to tell), but one could as easily interpret Scott Rao's comments as the bleeding edge of percolation - all the way to "borderline overextraction" as an "Americano" of the batch brewer world (as many of the espresso fans will admit - many an undrinkable trial demitasses can ends up as "better than Nescafé" Americanos - by mere dilution!).

Chemical analysis could easily prove this right or wrong (or help set control targets aimed at fixing process issues - lowering temp? Agitation time? Etc). Or one can take the road most travelled by and hope marketing hype delivers against the promise.

Many will say this endless state of innovation is good for the industry - very hard to argue with - except to say the too much breeds cynicism. Cynicism is what the high end audio world is learning to cope with right now (and it isn't pretty).

In any case - I do wish you much success and hope your system proves itself and ends up popping up in coffee places around the world. I will be one of the first to try one out when it does (or next time I am in the Bay Area).
Scraping away (slowly) at the tyranny of biases and dogma.

Post Reply