Posted Thu May 17, 2012, 5:43am Subject: Re: Coffee Extraction Discussion, Questions for the membership:
GlennV Said:
The results are impressive and the story compelling, however I'm puzzled that temperature isn't having an effect here. You said earlier that
in which case, assuming the specific heat capacity of ground coffee is 0.4 that of water (as in Scott Rao's book), the initial average slurry temperature of your 18% brew would have been over 3C lower than your 6% one. For example, if your water temperature was 95C and the dry grounds temperature 20C, and assuming no heat transfer into the plastic ...
Before you get to the stage where you can't face grinding another bean, might I suggest trying a protocol where you add the first 75g of water to the 6% brew at the same temperature as the water added in the 18% brew, so the hit temperature is the same, and then the rest of the 225g at a lower temperature in an attempt to equalize the initial slurry temp? Nevertheless, impressive stuff.
The apparent linearity of these results could be a coincidence.
If higher brew ratios (lower temperature profiles) result in lower effective extraction AND the grounds liquid is actually something less than full strength coffee then you could get a result that looks like the grounds liquid is full strength when it isn't.
For example, if the brew ratios are 6%, 12%, and 18% and the strengths are 1.3%, 2.6%, and 3.9% it looks like the grounds liquid strength equals the cup strength. That would imply an effective extraction (all the liquid) of 21.7% for each of the three brews. They might taste "okay".
But what if the strength of the grounds were actually only 50% of the cup strength? Then the effective extractions would be 20.8%, 19.8%, and 19.0%. These might all taste okay too.
The problem is that we don't know if our lone taster can reliably resolve differences of this magnitude. If he can only say "underextracted", "okay", "overextracted" with certainty it leaves a fair amount of wiggle room.
I'm not sure that your presciption for adjusting the temperature makes sense. I've spent a little time recording Aeropress temperature profiles and the concept of a "hit" temperature as a way to guarantee equal extraction profiles seems like a reach to me.
I was also surprised with the temperature, as I said before I expected a decrease in extraction specifically because of the change in brew temperature.
I don't plan on running the whole experiment again with an insulated press pot or anything, but here's what I have done to try and figure out why temp isn't playing as big a part as I expected:
-I used an insulated press pot (hand made from cozy cup material), pre-heated with boiling water, 208°F strike temperature, fine grind and 12% brew ratio. 8:00 contact time, and the brew slurry was still 179°F at press time. I did this because I was discussing with Vince F. why I couldn't achieve a "technical" overextraction even though it tasted like it was. I expected strength to be in the 4%+ range, but was really confused when I measured 3.00-3.04%. The result was really silty, and it's difficult to get a good measure of "produced" coffee when it's got so much dregs, but my best estimation was around 18.7% based on "yield", but the grounds were really wet, and I actually spilled some of it in the process of trying to get the grounds out of the bottom of the press.. It was also very difficult to let it filter through a basket. This brew was NASTY astringent and bitter (produced from Guatemala Antigua) - the worst I've ever had from a press pot. Also, this was done before I started thinking about the strength in an immersion brew - all I knew is I had a nasty overextracted coffee that didn't "measure" that way.
-I checked a med grind (D+3 in my nomenclature) ~6% brew ratio with pressing it at 30 seconds (AeroPress, Kenya AA Peaberry). This was surprisingly strong, as I expected it to be fairly weak - but compared to a second cup later that day at the full 3 minutes, the strengths were something like (going by memory) 1.22% at 30 seconds, 1.27% at 3 minutes.
-At 10 minutes for a cold brew, I noted the strength (just a few drops sample) within 0.20% of the final strength 24 hr later (something like a 15% brew ratio). Obviously, strike temperature is fairly meaningless for cold brew.
-I've also tried "stepping" the water to the brew with effective 9.1% brew ratio med grind (D+2) over 8 minutes on 4/16 with EoC coffee. I heated each step in the microwave to boiling in between introduction of the brew water over about 6 minutes and agitated for the final 2 minute, but found the strength only made it to about 1.92%. This tasted fairly decent (a complete shock), I would not have called it overextracted, but was trending toward it (though knowing how I brewed it may have skewed my assessment, but my wife did call it "a bit 'off'" - somewhat oily/bitter) - and it had me checking the settings and math over and over again. 23.5g coffee, 225.9g coffee produced at 1.92%. Brew water was 258.2g. (I happened to have this brew handy in an email because I've been corresponding with a friend on my findings).
-I simmered fine ground coffee on the stove for 10 minutes. I used med-fine grind (D-1, slightly coarser than the settings for the experiment) 25g of coffee and started with boiling 500g of water. (I actually had to do this experiment twice - turns out that if you just dump dry coffee into boiling water, you end up with a HUGELY foaming mess!) When I transferred to a pyrex mixing cup, the slurry was around 415g at the time of filtering, but it took a half hour to filter the stuff, so I ended up with 324.1g of coffee produced, with the wet grounds 72.1g - so with all the loss due to evaporation, at the end I had to call the coffee somewhere around 6.7% brew ratio. VSTcr was 1.51% - for a yield calculated extraction of 19.6% - after ALL of that. Obviously, this stuff tasted like boiled tar combined with old shoelaces laced with a hint of coffee. (LOL, maybe I exaggerate a bit, but not much...)
-Checking some of the recommended AeroPress brews, they may not be as underextracted as we think. It depends on the grind, but at fine grind the strengths for the huge brew ratios (10 second steep n stir, 25 second press, 175°F strike temp, approximately 25%-30% brew ratio if you use two scoops and fill to the (2) on the cylinder)) are still in the 4.5% range. The yield calculated extraction says 12% - but it doesn't taste like its THAT low. (~27g coffee, ~100g brew water, ~70g produced, give or take)
I'm not sure what will happen when I change the strike temperature, but I suspect that grind is probably going to be the primary control variable (which supports the advice that a good grinder is an excellent thing to obtain for brewing good coffee). Contact time will probably end up the secondary control variable, and my guess at this time is that the contact time and grind are interactive (contact time is probably more important when the extraction rate is slow, i.e. coarser grind, vs. fine grind - especially if immersion brews have some sort of effective "extraction saturation"). I'm beginning to suspect that the temperature dissipation is less important than the strike temperature itself, and the strike temperature is probably more important for finer grind than coarser grind.
Now, with a drip brewer (or basically any percolation method) - temperature, time, and grind are all intertwined. Finer grind increases the contact time and extraction rate. Coarser grind decreases the contact time AND extraction rate. So, grind is even MORE important, especially if percolation methods increase propensity for overextraction.
------------------------------------------ ----------------------------------------- Le café doit être noir comme le diable, chaud comme l'enfer, pur comme un ange, et doux comme l'amour.
"There is no right answer with coffee. There is only the elixir in your cup at the moment you partake."
"...I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind;..." - Lord Kelvin RECIPES thread => http://www.coffeegeek.com/forums/coffee/machines/585708
jpender Senior Member Joined: 11 Jul 2011 Posts: 428 Location: California Expertise: I like coffee
Grinder: Kyocera CM-50 Vac Pot: S/S Moka Pot Drip: Aeropress
Posted Thu May 17, 2012, 10:17am Subject: Re: Coffee Extraction Discussion, Questions for the membership:
Your initial results are suggestive and it's a hypothesis I'd like to believe. But so far the proof isn't there.
You have no way to directly measure the strength of the grounds liquid. The effective extraction is being measured by one guy's nose and taste buds. These are presumably not done blind so inadvertant bias could play a role.
You boiled coffee slurry in a microwave on and off for 8 minutes and it tasted okay? Really??
One way to try and tease out a temperature effect: Brew three samples identically at, say, 24% brew ratio. Then dilute them with water at the slurry temperature to 6%, 12%, and 18% just before seperation from the grounds. If the strengths are still proportional to the brew ratios then the temperature variation isn't an issue. This would also help to exclude any possible effect concentration might have on extraction.
This might not work either, though. Maybe you'd end up with grounds liquid that is stronger than what's in the cup.
Except for inference, by taking a press, then squeezing to try and get the last bit of drippings and checking the strength. So far, this is a coin toss within +/-0.02% on the VST of the produced coffee.
Regardless, it can't be "yield" based extraction for immersion brews, otherwise your resulting extraction would be based on whether you took one or three cups out of your press pot and used that for a basis.
So far for me, it explains why I can't brew a high brew ratio in a Press Pot that achieves a yield-based extraction even close to what makes sense.
jpender Said:
The effective extraction is being measured by one guy's nose and taste buds.
That's why I post what I've done - so others can repeat if they feel so inclined. Don't just trust my taste buds - the hypothesis I've proposed has a physical limit that anyone can test.
If someone could do an 18% brew ratio press pot and achieve 5.2% strength - I'd love to see the "recipe" so I could repeat it. I wouldn't even care what it tasted like - if what I'm saying has merit, this should be pretty much physically impossible, requiring over 30% extraction (immersion method) to achieve. What I'm suggesting has passed what I "taste" into a physical description which is open to disproving.
So that's the challenge:
Use a press pot (the pretty much common immersion method that most people can do)- even heating the water, insulation, etc., and a steep time <=10 minutes, and attempt to produce a brew at 18% brew ratio that achieves 5.2% strength per a VST coffee refractometer.
Then, post the results (grams of coffee used, grind level, steep time, grams of brew water used, and VST data).
(BTW - it's all VST's fault, you know. The first thing I was trying to do was produce taste-boundary samples, and when I couldn't, THAT'S where the trouble began... )
jpender Said:
These are presumably not done blind so inadvertant bias could play a role.
"The experiment" was not blind but side-by-side paired comparisons of the randomized runs, semi-blind where possible (i.e. I have two identical cups, and I put a sticker on the bottom of one after pulling and preparing the samples for the refractometer). But yes, inadvertant bias can play a role. The temperature of one of the cups was always cooler than the other, too.
Most of the time I had taste-sampled it before I had the strength number - and even so, I had "expected" strengths (remember, the original reason for the experiment was to validate the brew control charts) so there were several tastings where I said "AH HA! HERE is an overextracted sample!" but the strength comes in a full percent less than expected. I'll tell ya, you'll really question your palate when an instrument is telling you something different from what you're experiencing.
Other samples were not blind to me, but every sample I give to my wife is completely blind. She has a surprisingly good palate, and has completely identified fruitiness, bitterness, over and underextraction, sour notes, spice, chocolate, etc. If a coffee tastes a certain way, she ain't shy in telling me - and we have probably 85%+ agreement or more. Neither of us are professional tasters, tho.
jpender Said:
You boiled coffee slurry in a microwave on and off for 8 minutes and it tasted okay? Really??
I think you misunderstood. I set up the dry coffee in the Aeropress. Then I boiled the brew water, and added 30g of the water, stirred. While stirring, I re-microwaved the WATER to a roiling boil. Then the next minute, I added another 30-40 grams, and stirred. While stirring, I re-microwaved the WATER to a boil, and added some MORE just out of microwave boiling water to the brew. Repeat until I got to around 260g of water into the AP, which took around 6 minutes.
I was trying to simulate less heat loss than the AeroPress normally does - by adding very hot water in steps over the brew time. I'm afraid to microwave my AP in the microwave, and I doubt it would fit anyway.
jpender Said:
One way to try and tease out a temperature effect: Brew three samples identically at, say, 24% brew ratio. Then dilute them with water at the slurry temperature to 6%, 12%, and 18% just before seperation from the grounds. If the strengths are still proportional to the brew ratios then the temperature variation isn't an issue. This would also help to exclude any possible effect concentration might have on extraction.
This might not work either, though. Maybe you'd end up with grounds liquid that is stronger than what's in the cup.
Proposed would be actually 4 samples, set to same grind. I don't think it matters whether it's fine or medium. I'd probably avoid extremely coarse grind.
I think this could be done in either a Press Pot or an AeroPress.
If in an AeroPress:
Start out with a 12:60 brew (24%). Obtain a strength. Brew with 12:60 (24%), just before pressing, top off to 67g of water at same temp as the brew, probably around 175°F (18%), quickly stir and press, then obtain a strength. Brew with 12g:60, just before pressing, top off to 100g of water (12%), quickly stir and press, and obtain a strength. Brew with 12g:60, just before pressing, top off to 200g of water (6%), quickly stir and press, and obtain a strength.
If the dilution is pure enough, and the original brew achieved a strength of ~4.00%, then, the other three strengths should simply end up as diluted versions of the original strength (3.6%, 2.4%, 1.2% respectively). Departure from this relationship implies a brew ratio effect. That seem right?
------------------------------------------ ----------------------------------------- Le café doit être noir comme le diable, chaud comme l'enfer, pur comme un ange, et doux comme l'amour.
"There is no right answer with coffee. There is only the elixir in your cup at the moment you partake."
"...I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind;..." - Lord Kelvin RECIPES thread => http://www.coffeegeek.com/forums/coffee/machines/585708
Posted Thu May 17, 2012, 1:53pm Subject: Re: Coffee Extraction Discussion, Questions for the membership:
Thanks in public for the email that reminded me that in simmering fine-ground coffee on the stove - I basically made coarser-than-normal unsweetened Turkish coffee.
LOL
Apparently, very few people have ever measured the TDS of Turkish coffee.
------------------------------------------ ----------------------------------------- Le café doit être noir comme le diable, chaud comme l'enfer, pur comme un ange, et doux comme l'amour.
"There is no right answer with coffee. There is only the elixir in your cup at the moment you partake."
"...I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind;..." - Lord Kelvin RECIPES thread => http://www.coffeegeek.com/forums/coffee/machines/585708
jpender Senior Member Joined: 11 Jul 2011 Posts: 428 Location: California Expertise: I like coffee
Grinder: Kyocera CM-50 Vac Pot: S/S Moka Pot Drip: Aeropress
Posted Thu May 17, 2012, 3:24pm Subject: Re: Coffee Extraction Discussion, Questions for the membership:
Netphilosopher Said:
Except for inference, by taking a press, then squeezing to try and get the last bit of drippings and checking the strength. So far, this is a coin toss within +/-0.02% on the VST of the produced coffee.
Regardless, it can't be "yield" based extraction for immersion brews, otherwise your resulting extraction would be based on whether you took one or three cups out of your press pot and used that for a basis.
One alternative hypothesis is that there is grounds water that is bound in some fashion that does not contain the same concentration of dissolved solids. If that were true then squeezing the grounds wouldn't reveal it. You could infer it by measuring the effective extraction but the only tool you have for that is your taste.
Suppose I give you a cup of drip brewed coffee at a standard strength and ask you to estimate the extraction yield. How accurately can you reliably do it? Within 1%? Within 2%? Within 4%?
Netphilosopher Said:
So that's the challenge:
Use a press pot (the pretty much common immersion method that most people can do)- even heating the water, insulation, etc., and a steep time <=10 minutes, and attempt to produce a brew at 18% brew ratio that achieves 5.2% strength per a VST coffee refractometer.
Oven dehydration has fallen out of favor with you??
Netphilosopher Said:
That's a good idea, if I understand it correctly.
Proposed would be actually 4 samples, set to same grind. I don't think it matters whether it's fine or medium. I'd probably avoid extremely coarse grind.
I think this could be done in either a Press Pot or an AeroPress.
If in an AeroPress:
Start out with a 12:60 brew (24%). Obtain a strength. Brew with 12:60 (24%), just before pressing, top off to 67g of water at same temp as the brew, probably around 175°F (18%), quickly stir and press, then obtain a strength. Brew with 12g:60, just before pressing, top off to 100g of water (12%), quickly stir and press, and obtain a strength. Brew with 12g:60, just before pressing, top off to 200g of water (6%), quickly stir and press, and obtain a strength.
If the dilution is pure enough, and the original brew achieved a strength of ~4.00%, then, the other three strengths should simply end up as diluted versions of the original strength (3.6%, 2.4%, 1.2% respectively). Departure from this relationship implies a brew ratio effect. That seem right?
Posted Fri May 18, 2012, 4:52am Subject: Re: Coffee Extraction Discussion, Questions for the membership:
And here's the quick chart for grind level vs. resulting strength.
I'll be moving to contact time next (using a D-setting of 4), but at 18:240 I'll be having a hard time getting 10 second infusion and the rest of the 240g of brew water into the AeroPress for the 30 second run. We'll see. At worst, I'll have to re-do it at like 15:200 or something like that.
(Click for larger image)
------------------------------------------ ----------------------------------------- Le café doit être noir comme le diable, chaud comme l'enfer, pur comme un ange, et doux comme l'amour.
"There is no right answer with coffee. There is only the elixir in your cup at the moment you partake."
"...I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind;..." - Lord Kelvin RECIPES thread => http://www.coffeegeek.com/forums/coffee/machines/585708
One alternative hypothesis is that there is grounds water that is bound in some fashion that does not contain the same concentration of dissolved solids. If that were true then squeezing the grounds wouldn't reveal it. You could infer it by measuring the effective extraction but the only tool you have for that is your taste.
Suppose I give you a cup of drip brewed coffee at a standard strength and ask you to estimate the extraction yield. How accurately can you reliably do it? Within 1%? Within 2%? Within 4%?
I think I could probably tell if it's pushing over 22%. I could definitely pick out under 18% with a drip - but I'm finding that fairly rare. Can I tell the difference between 19% and 20%? No, not with any reliability. That's why I'm pushing the boundaries so I can define unequivocally assessed over extraction.
When I can remove the subjective from the equation, then we're getting somewhere - based on the hypothesis, I've found a testable physical phenomenon that is completely independent of taste - at a particular brew ratio, using immersion, extract at much as you possibly can. Do anything so long as you don't change the method itself (the brew slurry remains in contact up until filtering).
jpender Said:
Oven dehydration has fallen out of favor with you??
I'm actually running a separate study on that, and the bottom line is that I can get darn close to the VST with home-oven dehydration. Of course, I have access to a centrifuge, too. The dehydration studies I've done have mostly contained a measurable amount of undissolved solids, so I'm seeing if there is a way to estimate the error.
jpender Said:
Yeah, something like that. How long do you suppose it takes for the grounds to equilibrate with the added water when you dilute to a lower brew ratio?
------------------------------------------ ----------------------------------------- Le café doit être noir comme le diable, chaud comme l'enfer, pur comme un ange, et doux comme l'amour.
"There is no right answer with coffee. There is only the elixir in your cup at the moment you partake."
"...I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind;..." - Lord Kelvin RECIPES thread => http://www.coffeegeek.com/forums/coffee/machines/585708
Posted Fri May 18, 2012, 7:58am Subject: Re: Coffee Extraction Discussion, Questions for the membership:
Regardless of taste, if it were physically possible to extract to 30%, the immersion hypothesis indicates a limitation in strength.
If we consider 30% extraction the rough physical limit for coffee extraction for any coffee, then we're not picking subjective evaluations of what is "ideal", or a "sweet spot", or "over extracted". There is a testable limit for which any brew will be below.
If there is enough data at maximum extraction, then this limit can be adjusted if we start seeing data points above the prescribed line.
The other interesting thing is that increasing the absorption term increases the curvature from straight (immersion hypothesis) to quite non-linear at very high brew ratios and percolation settings for absorption. So not only can we test the limit itself, but we can test the shape.
(Click for larger image)
------------------------------------------ ----------------------------------------- Le café doit être noir comme le diable, chaud comme l'enfer, pur comme un ange, et doux comme l'amour.
"There is no right answer with coffee. There is only the elixir in your cup at the moment you partake."
"...I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind;..." - Lord Kelvin RECIPES thread => http://www.coffeegeek.com/forums/coffee/machines/585708
jpender Senior Member Joined: 11 Jul 2011 Posts: 428 Location: California Expertise: I like coffee
Grinder: Kyocera CM-50 Vac Pot: S/S Moka Pot Drip: Aeropress
Posted Fri May 18, 2012, 8:59am Subject: Re: Coffee Extraction Discussion, Questions for the membership:
Netphilosopher Said:
Regardless of taste, if it were physically possible to extract to 30%, the immersion hypothesis indicates a limitation in strength.
If we consider 30% extraction the rough physical limit for coffee extraction for any coffee, then we're not picking subjective evaluations of what is "ideal", or a "sweet spot", or "over extracted". There is a testable limit for which any brew will be below.
If there is enough data at maximum extraction, then this limit can be adjusted if we start seeing data points above the prescribed line.
The other interesting thing is that increasing the absorption term increases the curvature from straight (immersion hypothesis) to quite non-linear at very high brew ratios and percolation settings for absorption. So not only can we test the limit itself, but we can test the shape.
This seems like a reasonable approach, provided the machinations required to push extraction to its logical conclusion don't alter the nature of the spent grounds in some important way.
Posted Mon May 21, 2012, 6:45am Subject: Re: Coffee Extraction Discussion, Questions for the membership:
Okay... so strength for a given grind and brew ratio is not linear with respect to time.
This is a brew ratio and grind that gives me approximately 1.44-1.45% strength at 3 minutes.
Looking at a modified Rayleigh distribution function as a possible model for the strength vs. time (decent shape with only one input variable, easily differentiable for continuous extraction rate profile - IF it works).
I've got a different press pot 7.5% brew ratio at 8 minutes on same grind level that I did several weeks ago, similar/same strength (approx 1.45%), but I plan on filling out the 3 1/2 to 8 minutes over the next week with AeroPress brews. It does appear that for this grind level the extraction begins to slow down/stop at the longer time ranges.
Getting anything less than 40 seconds is a real challenge, even with Eight o Clock - the bloom is hard to deal with, even with a very quick pre-soak.
(Click for larger image)
------------------------------------------ ----------------------------------------- Le café doit être noir comme le diable, chaud comme l'enfer, pur comme un ange, et doux comme l'amour.
"There is no right answer with coffee. There is only the elixir in your cup at the moment you partake."
"...I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind;..." - Lord Kelvin RECIPES thread => http://www.coffeegeek.com/forums/coffee/machines/585708
Symbols: = New Posts since your last visit = No New Posts since last visit = Newest post
Forum Rules: No profanity, illegal acts or personal attacks will be tolerated in these discussion boards. No commercial posting of any nature will be tolerated; only private sales by private individuals, in the "Buy and Sell" forum. No cross posting allowed - do not post your topic to more than one forum, nor repost a topic to the same forum. Who Can Read The Forum? Anyone can read posts in these discussion boards. Who Can Post New Topics? Any registered CoffeeGeek member can post new topics. Who Can Post Replies? Any registered CoffeeGeek member can post replies. Can Photos be posted? Anyone can post photos in their new topics or replies. Who can change or delete posts? Any CoffeeGeek member can edit their own posts. Only moderators can delete posts. Probationary Period: If you are a new signup for CoffeeGeek, you cannot promote, endorse, criticise or otherwise post an unsolicited endorsement for any company, product or service in your first five postings.
![Netphilosopher: Grind Chart - 7-5pctBR180secondContact.jpg](javascript:ImagePopup("/images/57601/GrindChart-7-5pctBR180secondContact.jpg", 914, 665, "Netphilosopher: Grind Chart - 7-5pctBR180secondContact.jpg");)
![Netphilosopher: 30PCT Extraction Limit Plot.jpg](javascript:ImagePopup("/images/57604/30PCTExtractionLimitPlot.jpg", 800, 582, "Netphilosopher: 30PCT Extraction Limit Plot.jpg");)
![Netphilosopher: 7.5PctBrewRatio-ContactTime.jpg](javascript:ImagePopup("/images/57628/7.5PctBrewRatio-ContactTime.jpg", 914, 665, "Netphilosopher: 7.5PctBrewRatio-ContactTime.jpg");)
= No New Posts since last visit 