Biggest Scientific Study on Espresso Extraction - Surprising Results
-
Abe Carmeli
- Team HB
- Posts: 845
- Joined: 19 years ago
Today, the scientific journal Matter, published what may be called the biggest scientific paper on espresso extraction. The bottom line is this: Use 15 grams of coffee and time your shot to be 7-15 seconds. Adjust the grind to reach the optimal time for a particular coffee within the 7-15 sec. window, but keep it at 15 grams. This leads to a coarser grind. There is a ton more in this study. Click on the link below for the full article.
https://www.cell.com/matter/fulltext/S2 ... 19)30410-2
https://www.cell.com/matter/fulltext/S2 ... 19)30410-2
Abe Carmeli
-
caeffe
- Posts: 467
- Joined: 17 years ago
Whoa!!??
Grind coarser and go for a shorter time? Mainly in the name of saving coffee mass and time.
I haven't read the paper in detail as it hurts my head due to all the scientific equations but bottom line is rather than 20gin 40g out in 30 secs it says 15g in and xg out (I haven't found the ratio yet) in 7-15 secs.
All of this is counter to what we've been told and what has been practiced - as even admitted to by the paper's authors.
Would definitely like to know what the august body here thinks about this study. I'll have to read and understand this a bit more.
Grind coarser and go for a shorter time? Mainly in the name of saving coffee mass and time.
I haven't read the paper in detail as it hurts my head due to all the scientific equations but bottom line is rather than 20gin 40g out in 30 secs it says 15g in and xg out (I haven't found the ratio yet) in 7-15 secs.
All of this is counter to what we've been told and what has been practiced - as even admitted to by the paper's authors.
Would definitely like to know what the august body here thinks about this study. I'll have to read and understand this a bit more.
LMWDP #162
-
John_Doe
- Posts: 67
- Joined: 4 years ago
Opinions vary and I don't value scientific opinions much at all! Maybe those conducting the study can't handle more intense/involved espresso.
-
ojt
- Posts: 846
- Joined: 6 years ago
Well unsurprisingly consistency goes up with such a coarse grind. I have a few doubts about their brewing methods, including distribution which becomes more important with finer grind.
So why not suggest to use pressurized baskets and be done with it?
So why not suggest to use pressurized baskets and be done with it?
Osku
-
Somedays
- Posts: 3
- Joined: 4 years ago
The section of the article "Blending Shots" pretty much sums this up as a way to save the amount coffee being used and not as way to make better tasting espresso...just my take on it.
This part mostly "It is clear that espresso made at 22% EY in the partially clogged regime tastes more "complex" than a fast 22% EY obtained using the optimization routine presented in Figure 6"
Why not make 14 - 15g complex regular double shot even if it took an extra 10 - 15 seconds.
This part mostly "It is clear that espresso made at 22% EY in the partially clogged regime tastes more "complex" than a fast 22% EY obtained using the optimization routine presented in Figure 6"
Why not make 14 - 15g complex regular double shot even if it took an extra 10 - 15 seconds.
-
another_jim
- Team HB
- Posts: 13960
- Joined: 19 years ago
Hi Abe; nice way to return -- with a sciency espresso troll!
So where do they go wrong?
The article starts with standard flow equations and their linearizations. This is good and bad news, The bad news is that the linearizations do not actually account for the nitty-gritty of percolation, where solubles are both extracted and reabsorbed in the coffee bed, making equilibrium flow linearizations unreliable. The good news is that all the math is just sciency window dressing. After announcing that none of the parameters can be estimated, they move on to the empirical part, which makes no use of the flow equations.
Here they get the usual result that finer grind and lower doses lead to higher extractions. Then they make the dreadful discovery (cue in the scary music): fine ground shots can choke; and when they do, the extraction yields go down. Their recommendation, avoid choking at all costs by using 15 gram doses, and making 15 second, 30 gram shots.
So why are these shots not a complete disaster? My guess is because they are using 18 gram VST or Strada baskeets throughout. Do you know how fine you have to grind to get a 15 second, 30 gram shot out of an 18 gram VST dosed at 15 grams? When I tried it with a 68mm conical a few years back, I couldn't do it all. That poor EK was probably screeching in protest.
Another thing before you go dusting your baskets with a thin layer of Turkish grind - there's this brand new thing called pressure profiling and variable preinfusion. You really don't have to crank up the horror music and shake the cameras every time a shot chokes. You can preinfuse instead.
So where do they go wrong?
The article starts with standard flow equations and their linearizations. This is good and bad news, The bad news is that the linearizations do not actually account for the nitty-gritty of percolation, where solubles are both extracted and reabsorbed in the coffee bed, making equilibrium flow linearizations unreliable. The good news is that all the math is just sciency window dressing. After announcing that none of the parameters can be estimated, they move on to the empirical part, which makes no use of the flow equations.
Here they get the usual result that finer grind and lower doses lead to higher extractions. Then they make the dreadful discovery (cue in the scary music): fine ground shots can choke; and when they do, the extraction yields go down. Their recommendation, avoid choking at all costs by using 15 gram doses, and making 15 second, 30 gram shots.
So why are these shots not a complete disaster? My guess is because they are using 18 gram VST or Strada baskeets throughout. Do you know how fine you have to grind to get a 15 second, 30 gram shot out of an 18 gram VST dosed at 15 grams? When I tried it with a 68mm conical a few years back, I couldn't do it all. That poor EK was probably screeching in protest.
Another thing before you go dusting your baskets with a thin layer of Turkish grind - there's this brand new thing called pressure profiling and variable preinfusion. You really don't have to crank up the horror music and shake the cameras every time a shot chokes. You can preinfuse instead.
Jim Schulman
-
Peppersass
- Supporter ❤
- Posts: 3694
- Joined: 15 years ago
There are many flaws in this "study". Among them:
1. The use of only one so-called "espresso roast" coffee without reporting the roast level and completely disregarding the role played by roast level, how the coffee was processed, the varietal, etc. They got a balanced cup of the particular coffee they selected by adjusting the grind so a 15g dose would produce 40g in 7-15 seconds. I very much doubt that would work with a light-roasted specialty coffee because coarse grind will never produce a balanced extraction for such a roast. It would be very sour and underextracted.
2. The use of one grinder, one espresso machine and one basket. There was no attempt to compare results with other equipment. To the extent there was inconsistency in the 20g --> 40g shots, one might wonder if the burrs of the EK-43 they used were aligned, a known issue with that grinder, or how gentle (or not) the preinfusion is on the selected espresso machine.
3. Similarly, there was no attempt to determine if the inconsistency was due to poor distribution technique. Maybe the 20g dose had headroom issues in the 20g basket.
4. Prioritization of time and cost of coffee (lack of waste) over flavor. They acknowledge that the 15g --> 40g shot will be diluted. That's an understatement. To most of us, it would be watery with very thin mouthfeel. Perhaps OK for milk drinks, but not acceptable for straight espresso. The "blended shot" alternative essentially concedes this. Why bother? Why not address the consistency issues with the 20g --> 40g ratio?
I understand that consistency is a problem for shops that use sub-par equipment and don't train their baristas well. So this sounds like advice for those cafes, which probably sell mostly milk drinks anyway. But why go to a place like that instead of Starbucks? Perhaps because they save enough time and money with this technique to sell their drinks cheaper? A cafe with well-trained baristas using top-quality coffee and gear will consistently produce great espresso shots with higher brew ratios and will get my business.
1. The use of only one so-called "espresso roast" coffee without reporting the roast level and completely disregarding the role played by roast level, how the coffee was processed, the varietal, etc. They got a balanced cup of the particular coffee they selected by adjusting the grind so a 15g dose would produce 40g in 7-15 seconds. I very much doubt that would work with a light-roasted specialty coffee because coarse grind will never produce a balanced extraction for such a roast. It would be very sour and underextracted.
2. The use of one grinder, one espresso machine and one basket. There was no attempt to compare results with other equipment. To the extent there was inconsistency in the 20g --> 40g shots, one might wonder if the burrs of the EK-43 they used were aligned, a known issue with that grinder, or how gentle (or not) the preinfusion is on the selected espresso machine.
3. Similarly, there was no attempt to determine if the inconsistency was due to poor distribution technique. Maybe the 20g dose had headroom issues in the 20g basket.
4. Prioritization of time and cost of coffee (lack of waste) over flavor. They acknowledge that the 15g --> 40g shot will be diluted. That's an understatement. To most of us, it would be watery with very thin mouthfeel. Perhaps OK for milk drinks, but not acceptable for straight espresso. The "blended shot" alternative essentially concedes this. Why bother? Why not address the consistency issues with the 20g --> 40g ratio?
I understand that consistency is a problem for shops that use sub-par equipment and don't train their baristas well. So this sounds like advice for those cafes, which probably sell mostly milk drinks anyway. But why go to a place like that instead of Starbucks? Perhaps because they save enough time and money with this technique to sell their drinks cheaper? A cafe with well-trained baristas using top-quality coffee and gear will consistently produce great espresso shots with higher brew ratios and will get my business.
-
BuzzedLightyear
- Posts: 76
- Joined: 18 years ago
Also reported on CNN:
https://www.cnn.com/2020/01/22/us/how-t ... index.html
Love to hear everyone's opinion if 15 grams and coarser grinds is the better route to go
https://www.cnn.com/2020/01/22/us/how-t ... index.html
Love to hear everyone's opinion if 15 grams and coarser grinds is the better route to go
-
MPantani
- Posts: 45
- Joined: 4 years ago
@Peppersass,
It may not be to sell drinks at lower prices. I'm not in the industry but my field is tech entrepreneurship. I marvel at the coffee business because it is the only industry I can think of where the best and the mediocre sell for the same price. Maybe the goal is greater profit? The study authors also had a stated goal of reducing coffee consumption because of global warming.
It may not be to sell drinks at lower prices. I'm not in the industry but my field is tech entrepreneurship. I marvel at the coffee business because it is the only industry I can think of where the best and the mediocre sell for the same price. Maybe the goal is greater profit? The study authors also had a stated goal of reducing coffee consumption because of global warming.
-
thesharpener
- Posts: 167
- Joined: 8 years ago
I think the answer is no.BuzzedLightyear wrote:Love to hear everyone's opinion if 15 grams and coarser grinds is the better route to go
Pete - LMWDP #572
