Did Science Just Reinvent Espresso?
ฝัง
- เผยแพร่เมื่อ 1 ต.ค. 2024
- For a free trial of Squarespace, and a 10% discount on your first purchase go here: www.squarespac...
Read the paper: geni.us/reinve...
Music:
"Ad Infinitum" by Oh The City
Links:
Patreon: / jameshoffmann
My Book: geni.us/atlasof...
Limited Edition Merch: www.tenshundre...
Instagram: / jimseven
Twitter: / jimseven
My glasses: bit.ly/boldlondon
My hair product of choice: geni.us/forthe...
My coffee kit (studio): kit.co/jimseve...
My video kit: kit.co/jimseve...
Neewer Products I Use:
geni.us/neewer...
geni.us/neewer...
geni.us/neewer...
Hi James. I wanted to say thank you for making such a well-balanced review of our work, and also for helping to propogate the ongoing conversation about this. I'll be following the comments here with interest.
Lots of appreciation from this physics grad and coffee drinker! I really appreciate these sorts of wholesome, multidisciplinary papers: the balance between theory and experiment was absolutely on point here. It was a great and educational read.
Evidence based shot design, thank you for bringing some knowledge to an art. It can only be elevated. And thanks James, your words are like the interpretation of an article by a senior physician in medicine, filled with the pearls that make pure evidence relatable and creating real world usability.
Hi James, thanks for your paper. I enjoyed reading it. I can't help but wonder if the turning point in EY% is the emergent property that, in truth, defines a proper espresso. Could this property (micro-aggregate formation of fines) that allows for a creamy, yet well extracted shot, without crossing the over-extraction line, be the reason espresso migrates away from filter coffee? I wonder if the conversation can be two-pronged: (1) as James Hoffman suggested, how can we control the flow of water into the puck to be more thorough and cover a greater surface area and (2) how can we harness the emergent property to make excellent espresso more reproducible. Can this be achieved by layering coffee grounds such that the stratification artificially reproduces this effect? Would love to hear your thoughts.
@@fanothecardgame9996 No worries. Glad you enjoyed. The honest answer to your question about whether the turning point defines an espresso is that I am not sure. That being said, I do think it is a more useful definition than the presently accepted one by SCA, which, I think, is rather outdated and even misleading. An interesting point about stratification. With rather minor modifications, my code can actually simulate what we should expect from layered pucks. It'd be interesting to see what comes out of that. I need a willing grad student to try these things out.
I wonder if there's an ideal puck depth for extraction, and inversely an ideal puck cross sectional area (or even ideal shape) for a given measure.
I imagine they'll be a crossover much like there is with pressure where a deep puck could have better extraction (every ml of water travels through the most mass of coffee possible) but one that's too deep couldn't be overcome by the pressure and just jams.
This could lead to different (thinner & taller) basket / portafilter designs that use less coffee but obtain even greater extraction, and could be made to fit on existing group heads etc.
Would be an interesting experiment.
love the scientific term:"tasty point"
I think there is a sugar level in foods that manufacturers call the "bliss point".
I would argue to swap these names. "Bliss point" works much better for me when we're talking coffee than "tasty point".
Another captivating video. Thanks so much James.
It's nice we have a tasty point. Maybe scientists are not too far from a tasty line or zone. Who knows?
Jedidiah Solomon I thought it was tasting point
Best scientific term since “The Goldilocks Zone”.
@@stuffnuns My favorite scientific term is "Sausage instability".
I frequented Tailored in Eugene when I lived in the states. I would sometimes chat with Chris Hendon because we were there at the same time so often. These shots were really good. When the barista told me it was a 15 second shot, I glared. I tried it. It was good. He pointed at Chris. Chris looked pleased. He gave me the elevator pitch of the science. I got this kind of shot nearly everyday for the summer before I headed to Europe. I was still baffled by the explanation and still am, a bit.. The flavor is really good. I did taste a difference between the old, long pulls and the new pulls. Some flavors were lost, but other flavors appeared. I don't know if they were adjusting the bars yet, as this was two years ago and they were just trying it out for the first time. I bet they've improved the kinks. It's definitely worth having a go.
Chris also convinced me it was ok to freeze coffee if I wanted to keep it a while. He said the beans would preserve just fine. He turned out to be right.
At the end of the video, James reminds me of my overly excited high school physics teacher who himself was so excited about physics and desperately tried to transfer his excitement to the students.
haha! I will take this as a compliment!
Did you have Mr. Niznick (sp?) too? He made Bill Nye look mellow.
aeugler Did ist work? ;)
The best teachers! I’m guessing you remember particular lessons and experiments?
@@ehtikhet no.
I can speak anecdotally, but I’ve been a frequent flyer at that cafe in Eugene Oregon (Tailored Coffee Roasters), and I can speak that their espresso has always been the best I’ve ever had. I’ve spoken to the baristas and they’ve always said that their espresso was pulled faster and different. Thanks to this paper I understand what they mean now. I’ve also met Dr. Hendon once - super super smart.
So excited for this video James!!!
Hey! I live there! Will have to check them out!
I miss that place a lot :(
I really hoped this was about Centrifugally Separated Turkish Coffee.
Me too
boy do i want to see this
Isn't half the goal of Turkish coffee to have that mouth feel?
funny thing I actually build a system in my university years to do exactly that. The result was not as satisfying as the original thing
@@mildlydazed9608 Probably, but if you get clean ibrik (filtered) you might found out that its about best coffee you ever had. Except you will need to do it yourself.
I usually do cappuccinos but I've had really good results with shots that ran super long due to a way too fine grind, and good results with a way too fast shot (too course).
It bugs me that some people just throw out a shot of espresso because they think it didn't turn out good, especially if they never even tried it! If it's not good and you're not anti-dairy, add some milk and/or sugar and it's still salvageable :p People worked very hard to grow and prep that coffee. :O
“This is not a recipe to remove the barista from the process”
*Perger has left the chat*
TEST SETUP
Essp. Machine: GS3 AV
Pressure: 6 bar
Grinder: Mahlkonig E65S
Basket Size: 14g
Refractometer: ATAGO
Brewing Temperature: 93.7 ºC
Coffee Beans & Roast Date: RED BRICK Espresso Blend (20/11/2019)
Brewing Water: 3rd Wave (Espresso Profile) 167 PPM (TDS)
Scales: acaia
RESULTS
(1)
Water/Coffee Ratio: 2.14
Grinder Setting Range: 0.55 to 1.1
Peak Extraction %: 20.63 at 24.5 sec with 0.8 Grind Setting
(2)
Water/Coffee Ratio: 2.84
Grinder Setting Range: 0.8 to 2.05
Peak Extraction %: 22.66 at 19 sec with 1.7 Grind Setting
Best Tasting Recipe: 2.05 Grind Setting, 14g Dose, 39.9 Yield, at 16.5 sec, with 7.5% TDS (Strength), and 21.37% Extraction Yield
NOTE: I have had a home subscription with SQUARE MILE for a while and so far have consumed around 8kg of the RED BRICK Espresso Blend. RED BRICK never tasted this good.
To James Hoffmann,
We cannot thank you enough for your inspiring dedication and generous contributions to our lives!
I love how this paper will push people into the "less sexy" part of the scientific method, the replication studies part. That is the area where the nailing down of the facts and methods are made.
WE've been pulling 15 second shots at our cafe for the past 4-5 month; 16g input/30g output/15 second brew time... 5-6bars of pressure. Excellent results!
19.5g => 24g out is one of my faves no research paper necessary
@@zeropuckprep 🤔
@@zeropuckprep ratio is really just down to taste. you just preffer ristretto
Great to hear your thoughts on this James. As a natural skeptic, but capable of being able to try this at home I spent some time making some shots within their parameters. Similarly to your view, I did find the shots of 15g in and 40g out in 15 seconds to be surprising pleasant. I expected clear under extraction flavors, but they were sweet and had very high clarity. But I didn't find them capable of surviving an onslaught of milk. Like you mentioned, these things may vary a lot based on the coffee, and likely the equipment used (I used a Linea Mini and a Mazzer Mini).
It's interesting to hear your take aways, and I after hearing your thoughts I do wish I could go back in to my video and maybe talk a bit more on the espresso side and forgo a the bit on environmental and economic impacts (yet are still relevant, but tend to be more polarizing). I think the topic of even flow like you mentioned could definitely be something fleshed out a bit more and plan on digging into that a bit both on and off camera.
Thanks again for your sharing your insight, hope to see you at Expo in Portland. If so first piece of Tiramisu is on me.
In theory if you had a super wide basket with very thin layer of finely ground beans would it make a good even shot? Or if you had a compartment that filled up with the full amount of water and had a tamping method that would push it down all at once into the group head..
Would like yo hear both your thoughts and James on my two ideas
Why don't you make a video about this new recipe and taste impressions?
cristi m I did in fact make a video about this, it’s the most recent one on my channel.
David f theoretically those sound like they would crate even extraction. The tricky part would be having to complete design an espresso machine like one of these from the ground up.
@@Sprometheus yes you would need to remake every single tool and machine used for espresso making if you make a new much weider size basket but if it works it will be worth the investment. Not saying that every coffee shot needs to switch over but any new place that opens or shops that are planning on upgrading any way can switch over. And the idea if a machine with a compartment that fills up with water and pushes it down mechanically right into the puc will only need a change in the actual espresso machine which companies do anyway all the time when coming out with a new series.
Only question is well these methods actually work and if they will it is worth investing in
The first person that popped into my mind when I saw an article about this was you.
*read paper....waits for James to release his thoughts*
*video pops up next day*
I literally hit him up on twitter before reading the paper hahahahahahaha
James is the peer reviewer that carries weight!
I tried this at my cafe during a quiet period.
I found very similar results to yours, the texture definitely suffered a little, but the method still produced a nice cup. I found it hard to grow accustom to however... Especially when you compare it to what I'm normally tasting.
It had a lot of similarities to a V60 in terms of the clarity flavour in my opinion.
My recipe:
16g in
38g out
@ 15s
Equipment:
EK 43
Black Eagle
PUQ Press 2
Coffee:
Pablo & Rusty's
'Trailblazer' (seasonal blend)
Fuck yeah P&R make incredibly great coffee.
Is there a chance Nestle already knows this? I think I recall there being some questions a while back about how Nespresso pods are able to use so little coffee and get such high extraction? Some people even guessed they might be mixing in instant coffee, but that was proven not to be the case.
Thats an excellent point...I think I measured it at something like 6.3 grams. damn tasty too!
Nestle's market cap is 317 Billion dollars. They might know a thing or two about extraction.
Heard an interview with the author of the paper last week and capsule systems were brought up. He said that capsules were extracted at very high pressure (15-16bar) so that approach was quite different to the recommendations from the paper
as i understand it, capsule systems incorporate a restriction, like pressurized basket machines, and similarly the pressure seen by the puck is pump pressure minus restriction pressure. the reason for the high pressure of these systems is that high pressure plus high restriction means that the grind has little effect on the flow rate. it also means the pressure is too high to use a non-pressurized basket without difficulty, unless there is also flow metering ... in which case the pressure applied to a coarse puck is way less than the 15 bar rating, which is merely the maximum pressure the pump can apply as flow decreases
They taste like dirty water. Wash your mouth out!
My taste buds are dull. I probably couldn't tell a delicious espresso from strong pour-over. I'm just here for the smooth voice and the knowledge that nerdy people can be lovely and beautiful. So for that, I thank you.
when you brush your teeth try brushing your tongue too
The more you know the less you know. I kinda have the same oppinion about my taste buds, but then i remember:
- once pulled a shot so good i was smiling like an idiot, could taste the forest fruit notes so clearly, it wasn't even described in the coffee description, some local blend based on brasil;
- once had an aeropress at a coffee shop, actually had more at that shop but this one was something else, again very clear fruity notes, peach, nicaragua i think;
- at another shop they were testing some speciallity/expensive coffee for a future contest and they gave me to taste it in spoon, i recognized the lemon but i wasn't sure about my taste buds, and then the guy there told me first note is lemon :o
Anyway, i don't think even good baristas can pull 100% consistent shots. I can recognize good shots from bad ones, but rarely taste some notes.
The same, chemo is unkind to taste buds
@@bakhitalketbi9460 hey, it's great againt bad breath
@@bakhitalketbi9460 That's a good idea. I could also focus more on what I'm drinking.
I've got a low-end home setup consisting of a Gaggia Classic and a Sage BCG600SIL. Having watched this video, I checked the brew head pressure and found it to be 10 bar. I typically use 17 grams coffee for a 20 seconds 40 grams pour and to be honest I have struggled with pulling a good shot, but lowering the pressure to 6 bar and grinding slightly coarser to achieve a 15 second for 40 grams yields a much better shot. I pretty sure it would not be up to the standards that many who watch James' videos would be satisfied with but it's a considerable improvement for me. I suspect at 10 bar I was getting puck compaction and channelling, as I noticed the pour speed increase over the length of the pour. Maybe this has some value to us low-end brewers.
I have a Classic(w PID controller installed) as well with a Baratza sette 270. I lowered the pressure from 15 bar to 9 bar for 17g dose, then to 6 bar cause I switched to a 15g dose/basket and needed a much slower flow rate for the smaller dose. I aim for a shot between 15s to 24s. My grind is close to the finest setting on the Sette to the point where distribution/even flow are a sensitive situation and I could only get consistency with a distribution tool(using a naked portafilter). I find these faster shots are less sour and plump enough, but I'm still looking for the sweet spot with natural sweetness/flavor. I'm close but not there yet. I'm still searching to find this cross point where the grind size is the finest to slow down my fast shots but doesn't begin to mess with even distribution.
A couple days back, I was in a very small Maine cafe (way off-season), in a very small town, and the owner already knew about this study and was excited about learning more. There is hope for humanity.
I'm a PhD student so I have very modest equipment. My background is in computational physics, applied math and control systems engineering, so this paper is well within my wheelhouse. I taught myself to make Espresso a couple of years ago by standard doctrine. The timing of this video and paper, for me, is delightfully coincidental-as I've recently been discovering some of this on my own-just by toying around morning to morning. One morning I forgot to adjust my grind and noticed that I got a much nicer americano than expected on a looser grind. I started thinking about the tradeoff between channelling and surface area and postulated this sweet spot. I can't wait to read the paper. Thanks so much!
You are the only TH-camr that gets me pumped about uniformly shredding burnt berry pips. And it isn't unpleasant at all! Thank you for the video and your pragmatic approach to the paper and news.
Darn it James, you've interrupted my watching your older videos with another quality upload. How will I decide which to watch first!
Impressive gerund use, Tsunimo!
Literally happened to me with the aeropress video, just start watching this channel a week ago
I’ve been a barista for 13 years and this is the most thought inducing conversation for a very long time.
Happy to see that i am not too crazy... La Pavoni Europiccola with localy roasted Ethiopia Sidamo beans
14g in, 36 out (coarser grind)
10 sec 1.5 bar, 5sec at 4bars and around 10-12 seconds at 6-7bars.
I use and like this recipe since 1.5 years now and was thinking that my coffee taste was a bit odd...
Thanks for the great content!
You and sprometheus?! This must be an important espresso topic.
Perhaps a takeaway from the experiment is that manufacturers should make a wider and shallower portafilters/baskets/groupheads so that you can put enough flow through the puck with 18g-20g of coffee in 15 seconds to produce 1:2 extraction. Hello 70mm basket?
Not read the paper, but logically I'm not sure why you'd want to do this... A thinner puck would mean each ml of water travels through less coffee leading to less extraction?
I feel a narrower setup would lead to higher extraction especially with working with 6 bars as it would dramatically reduce the chance of channeling, however they'll be a crossover where there's too much work to get through the puck.
If only someone was working on that grinds distribution and puck longevity problem...
grinds distribution in the puck? get the grains frinded down to perfect cubes and sort them out and you'll get the most compact order.
don't hammet the puck on a solid surface to keep it from breaking, or get one where the crystall allignment in the material is ordered to withstand forces implied in hit direction
Matt Perger really looking forward to seeing your new gizmo. 🙌
Matt Perger eyyyyyyyyy
It is amazing how much science and engineering go into food and drink preparation. I believe the same things that drew me to chemical engineering draw me to watching your videos (as well as a bunch of food videos). I appreciate the places in life where science and art coalesce. Thank you for another very interesting video!👍🏽
I was waiting for this. I saw the articles, and was thinking "Alright, I hope James sees this and tries it out."
I don't see where the innovation is? Matt Perger propagated for years that you extract higher the finer you grind but there is a turning point once the puck clogs. The thing with the 15g shot in 15s only extracts higher because you actually need FINE grind not coarse grind to be able to hit 15s with such a low dose. Especially when you use an EK43 like they did, you still have enough contact area for a 15s shot. That lower doses and finer grinds extract higher is known since Jim Schulman published his paper "Some aspects of espresso extraction" in 2007. But even Jim doesn't do that anymore because with long +20s preinfusions you can use finer grinds and extract a lot higher even with standard doses. I for example repeatably pull 8.5 % TDS 25+ % EY shots with no bitterness or astringency. It's not magic anymore. I think the study better told you about the benefits of long preinfusion instead of advocating to mix two espressos at two different grind settings (how do you do that in a cafe and still save costs???).
The other day I did a very informal experiment where we dialled in our espresso as we normally do with our normal shop parameters, and then dialled in the same coffee with the parameters suggested in the research paper. Unfortunately we did not have a refractometer :(
Either way, it was very interesting to taste the differences in the shots. Along with what James said in the video, I noticed that brewing with these new parameters had more wiggle room than our normal 1:2 shots. A shot that wasn't perfectly dialled in was way more drinkable than a non dialled in shot using a 1:2 ratio. I also noticed less acidity and tartness compared to our 1:2 shots. Wow what a time to be a barista!!
Bitchin’, James.
I am downloading the paper.
I’m curious of implications of channeling and the cake for beer brewers.
I'm also a beer enthusiast. But I don't know what your comment means? Any quick explanation or link?
@@kellypg Not OP but I assume they're referring to sparging: brewwiki.com/index.php/Sparging
It's exactly the same issue in all grain brewing. But because we don't brew under pressure we can get around this with a rotating sprinkler which means tunnels are much less likely to form.
Thanks!
Kelly Grabill As a commercial brewer with 30 Barrel system, our “puck” is over 2000 lb and the “pull” takes over 3 hours at as close to 172F as possible without going over.
We usually have to stop halfway through and reinfuse the “puck” and stir the entire pile to destroy channels cut in the “filter cake”as we cal” it.
I was about to go to sleep but now I want to pull some shots
Chris Noonan aye, but I want to read this paper!
James, has anyone ever tried to build a device that uses somewhat high-frequency vibration of the cake during extraction? Wouldn't that help counteract channeling by closing the channels as they start to form?
It sounds amazing
What is your opinion James
Interesting idea. However, there is a good chance that this will cause hindrance to the flow and will alter the diffusion process. Due to these factors, the optimum frequency will change based on the characteristics of the coffee and the water used, so it may not be feasible to calculate the frequency every time. The mathematical model for this process will be quite complicated so I assume CFD simulations will be a much better option to study this.
By high frequency do you mean amplitude
They tested this for the coffee section of Modernist Cuisine. Not the success one might hope
@@jameshoffmann Bummer
I just love when people think of a theory about something, flesh out their thesis, and when testing it out, figure out something that is as ground breaking, but completely different than their original theory. This remind me a lot of Dave Arnold theorising that is the science of cocktail making, the shape and kind of ice used in your shaker didn't actually matter, and when he tested his thesis, he actually found out that yes, in fact, it really does matter, and the best way to use your ice in the process of shaking a cocktail was to use one big block of ice, for displacing and mixing your drink well, and a bunch of smaller ice shards around, to act on the percentage of dilution.
As always, I loved your video, thank you.
Reminds me I need to buy some Cocktail Cubes!
We've got to adapt our hypotheses based on new data. Our OG hypothesis was that you'd grind finer and get more out. 🤷🏼♀️ I guess that was not a good one in hindsight hehe
Next espresso extraction breakthrough: how to use lunar phase cycle to pull the best shot
I only brew when Mercury is in retrograde. And only on the fifth day. That makes sure that the color of the espresso is perfect and the flavor is slightly mirrored in the moon.
Best espresso ever.
My rising moon is Libra but my sun sign is Taurus so my best shots tend to be at 7 bars with a 9 grind setting.
Sipping coffee while reading comment section is good.
Hahaha! Much like biodynamic farming.
"you don't need to necessarily understand the math, you just need to trust that the people doing it understand the maths involved" is exactly how most scientists are using statistical methods. :D
Factual. I looked at the equation for Anova and I was like
I'm ok with not understanding this :)
I’ve been using faster shots between 16sec to 19secs at WBC and was very pleased with the results. The lack of mouthfeel can be mitigated by using a fresh roasted coffee that didn’t totally degassed.
Prof Hendon visited our café last year and did a lecture about this topic. We then tested immediately the fast shots. Surprisingly the coffee that tastes great were the washed. The Natural were better with slower shots.
I'd be curious to look at the history of pucks. As a mechanical engineer, I feel like there are better ways to get 'even' flow through the entire puck than just having water forced from the top to the bottom of the puck. Just a thought....
Over at Kruve that have proposed layering of the pick by grind size in the way a sediment filter would work. Fine grounds on the bottom, a middle size, and larger grinds at the top with tamping between each layer. Of course, keep out your grain of salt since Kruve makes equipment for screening coffee grounds.
Maybe a sort of inverted cone-shell with a conical insert to distribute water even inside. This should minimize channeling while still retaining a consistent thickness of grounds. The primary issue is being able consistently form the grounds to a uniform shape/density.
Scientists basically explained why exactly pressurized baskets come with almost all home espresso machines, as such basket holds the pressure not via grinding and tempering and such way allowing better extraction and nice result.
Perhaps the answer to why that one weird pod machine spins is because the centrifugal force prevents channeling.
Really great video, thanks! I was a little lost at the distinction between extraction and strength. I get that extraction is the percentage of the puck that ends up in the shot and I’m guessing strength is the percentage of dissolved coffee to water in the cup? But then he said around 12:30 that the two shots with 4g of dissolved solids in 40g water would be at 9% vs 12% strength and lost me. Is less extracted coffee stronger? Are we talking about a third variable like oils or caffeine or something?
Can someone talk me through it?
This confused me too. I would really appreciate if someone could clarify this.
Extraction% is how much of the dose ends up in the yield (whatever that amound may be)...
Strenght is a % of solids in the yield.
So if you use a 20g dose and yield 40g out at a 20% extraction there will be 4g of solids is the cup (20% of 20g is 4g)... The strength of the espresso will be 10% (4g of solids in a 40g espresso = 10%)
I love the discussion around this. I'm actually really into the minor environmental impacts. My sister went to a coffee farm in Guatemala and they couldn't use their own beans because they were too valuable. And the shop I work at dumps *so* much coffee. And I think about it every time I see a gallon go down the sink. Each little percentage point helps!
Great little video. If the amount of solids in the espresso is 4g for 40g of espresso, how is the concentration different between the 16g & 20g puck? 4g is 40g however it arrives there is 10% solid parts by weight. Why do you say this is going to feel different in the solubality/feel? I'm confused by your math and reasoning. Care to clarify?
12:20 you can tell for sure weak cappuccino tastes more like milk whereas a stronger espresso gives more of that coffee kick aside the milk flavours
“Try this” *immediately starts searching for a better coffee machine just so I can*
Sounds less like a better way to brew espresso than a more efficient way to brew espresso. For large scale brewers or companies with standardized products (Starbucks) that could potentially save them a lot of money.
No tamping preasure as a variable? one would belive the higher the tamping preasure the more channels would be created.
Thanks for your videos. We're gringos about to open a cafe in Mexico and the process they use here has a lot of room to grow. We just ordered a Ceado 37s so we can have the employees start right on grind. If you ever want to come to central Mexico and shoot some video with us about coffee here get in touch, we'll have a bed for you.
Thanks! And I wish you the best of luck with the cafe!
When I lived in Mexico some years ago, the coffee was SO BAD that I celebrated the arrival of Starbucks with glee. I supposed things are better now (I hope!)
Where in Mexico? City?
@@dodopurpura Mate, I had fantastic espresso in CDMX (Saks - San Angel location and Jeronimo also in San Angel - Alvaro Obregon). Amazing espresso from wonderful La Marzocco machines.
@@dodopurpura Well, it depends on where you are. If you are in any major city (Tijuana, Guadalajara, León, Monterrey, Mexico City, Puebla, etc.) you'll surely find at least one specialty coffee bar. This is a relatively recent situation, though, so it depends also on when this was.
Hi James! You know what REALLY surprised me? is that for some reason this is not news to me! In 2016 Socratic Coffee published a paper titled "Exploring the Impact of Particles on Espresso Extraction. Although the math behind Hendon paper is not present in the SC paper, the results are the same! Bigger particle size translated to significant lower times BUT higher EY! like wise, they found that finer particles increased the shot time but decreased EY!
Another example came from Barista Hustle in the article titled "If not Channeling, then what?" where is explained a fluid dynamics principle called no-slip boundary and how finer grounds made it harder for the water surrounding the particles to make it into the main flow, and how coarser grounds resulted in more "turbulence" and by result made it easier for the water to wash of the grounds and make it into the main flow.
Anyways, I had noticed such things in some shots by accident (because it doesn't matter how the shot comes out I ALWAYS taste then) where short shots with a coarser grind where tasty. I've always been pleasantly surprised when they are good, but they don't have the creaminess, syrupy goodness that I many times crave in an espresso.
But as always, thanks for you amazing videos and starting this type of discussions for us! Cheers from the Dominican Republic!
Just wondering. How did the puck look? And did you change the basket size?
I literally do this for almost 10 years and everyone tell me I am wrong , no I can show this to everyone I will translate this to 5 languages and hang it on my own coffee , I only trust to my taste until now science with me too.
I've tried this out on my Silvia/rocky setup with 15g in and 36 out it consistently takes 13-15 seconds and the espresso is much more enjoyable to me. I enjoy the looser texture and clarity and its never sour which is amazing. I would have inconsistency before where I can imagine what was being referenced towards too fine of a grind being counter to what we want. I've switched to using this technique on my daily I rather enjoy it.
Which brand of 15h basket are you using?
I have been watching your videos for a while and am very surprised at one point.
You are very precise in measuring every variable, temperature, time, pressure, coffe-water ratio BUT when talking about grinding, you are working pretty blind. Just with whatever numbers grinding machines give, but not a real standarized measure of grain sizes.
Measuring fine grains, and its proportions in a mix after breaking into very fine pieces any material, is something the industry and engineers do in many ways.
Check the abrasive material world and you will get an idea of what I am talking about.
Diameter of the grains is simply measured as “mesh” numbers, you have touched this item when using a sieve to separate smaller pieces from coarser ones.
Well, this sieves are catalogued by number of stings per inch, and you can perfectly separate a grinding into its portions of super small, small, not so small.....etc etc portions. Consistently and accurately, and NUMBERED (200 grain, 160....etc etc).
You never spoke about this topic -as far as I know- which may be really really important to get an even flow. You mentioned finer grains need higher pressure, and coaser grains won’t extract well, but the coffee in the container technically speaking is a filter, and filters may work better with uneven diametres.
If you were expecting for new ideas, try a not an evenly grinded measure, but some mix of diameters. Smaller pieces mixed with some coarser ones will prevent channelling, allowing the water to flow more evenly through the cake.
Also I expect a more precise grinding size guide, than “between salt and sugar size will be ok”.
I repeat you are extremely precise in all and every aspect ....except grinding size, which, I believe you, is extremely important.
It's the one variable that you usually change when dialing in a coffee so there's much less advantage to having precise information on it compared to just the setting used on the grinder.
Agree with you about the imprecise measurements we often use with espresso grinding. Still, if you google "coffee particle size distribution" you will find extensive material on this subject.
Sieves have been used for a long time for drip coffee grinding. For espresso grinds, however, they don't work well -- the particle size is extremely small and the coffee particles are oily and sticky. Espresso researchers use laser particle size analyzers, but their accuracy is limited since small particles stick together just enough to fool the analyzer into thinking a lot of small particles are actually one larger particle.
@@andrewschecter555 Thanks very much for your answer. Really interesting other people have thought in this too. Some things are still hiding their secrets to science, remaining in pure craftmanship. Still human :)
I'm disappointed that I somehow missed seeing this JH vid earlier... but there are quite a few! Anyway, coincidentally, I've been inadvertently trialling this in my own brewing journey. I presently use a Flair 58 manual espresso machine, which allows easy and dynamic adjustment of pressure in the face of excessive flow when the grind or other variables conspire against me. I often reduce from a nominal 8bars to 5-6bars if the flow is fast, to preserve a longer extraction time and (in my head) a chance to preserve sufficient contact time between water and puck to save what would be an otherwise under-extracted coffee. This doesn't always work... but now I know that there is merit in the method, I will try and be more methodical. I'm also trialling decaf for some of my daily coffee, and this requires a much finer grind to achieve nominal extraction time/flow rate. Being able to test a faster extraction without guilt will be interesting! :)
i thought a little too much about 'nominal extraction time/flow rate'... i first thought about you meaning to devide extraction time with flow rate ? you mean 'normal', right? i don't know how to interpret it otherwise
After graduating university I made a promise to myself to keep reading academic work, but I hadn't really kept myself to it. If there's more work like this out there however, that should no longer pose a problem. I don't have access to an espresso machine myself at the moment, but once I do, I can't wait to test all this out!
I'm waiting for the perfect steak, and the best burger too!
If you reduce your coffee in from 20 to 15 grams you may have to change to a smaller basket. From a technical perspective using a 14/15 gram basket with a dose of coarsely grounded coffee at 6 bars could be interesting for improving water to coffee contact. Pucks with coarsely grounded coffee tend to grow big, dry and hardish saturating the space between basket and shower. I am definitely intrigued to try it!
did the paper discuss mixing two different grounds to achieve stability and surface area? (I think it would be interesting to try just like you need small and large particles and stones in soil to get plants nutrients and drainage)
Is this still relevant today?
Years ago.. Michael Cameron and Matt from full court press found out that 4-6 bar made far more consistent espresso than 9 bar.
For convenience's sake, we bought a superauto machine that pulls shots quicker than I'd like, but which, while not out of this world, and perfectly drinkable and serve our purposes on busy weekday mornings. I've felt somewhat guilty as a coffee nerd for enjoying these shots, perhaps this study affirms that you can get a decent shot even if it's not a 30 second pull from a very fine grind.
AGAIN with the dope and funky intro. The wholesome coffee content I need. Keep up the great work sir I owe my tasty morning coffee to you.
"They used X" - " I had one of those lying around at Square Mile" - classic British understatement. Seriously though, this is a place where science and experience meet.
hi everyone!
i work at tailored coffee in eugene, oregon where this model is being used everyday.
if you’re doubting the method, come on in and check it out!
Nicolas Ryan Brown can I ask how long you have been using this technique? I’ve had quite a few shots from tailored original location and have always thought there was some difference I couldn’t 100% put my finger on
@@peterthompson8806 about 2 years.
I originally watched this video a few months ago and thought it was interesting, but quickly forgot about it. Tonight I was recommended Lance Hendrick's video on the topic and one of the things he focused on a lot was how much easier it is to get _repeatable_ shots using the coarser grind + lower pressure method the paper described. It's something I definitely want to try when my Flair arrives, as I'm sure my hand grinders won't consistently produce uniform enough grinds for a more traditional shot.
+1 lmao, same story, save for the fact that I sold my Flair and instead plan on making this with my newly bought Barista Max. I'll have to adjust the grind size to emulate the 6 bars I guess
@@mikairu2944 Actually one of the takeaways of the paper is that adjusting the grind size to emulate 6 bars would take you to an altogether different place.
@@hshooker5761 I see, to bad then.
Thank you! I have been so skeptical about these articles!
They're worth reading. And testing!
Hi James I own a coffee shop in South Africa we stumbled on to this method by chance we do 18g in 36g out in 18 seconds and yet to find a origin of coffee that does not taste good this way we use a Wega Atlas machine and a expobar grinder at 9 bar at peak. We also use a tamper that is depth control so all baristas use the same pressure so our espresso is more consistent from barista to barista. Here inS.A we have much less espresso drinkers than overseas. 70% of our drinks sold is milk based.
Regarding 3:00, I’m 99% certain a higher “Extraction Yield” will absolutely correlate to a more EVEN extraction if achieved with same dose, and final beverage mass. This is because each particle of coffee gives up the most solubles when it is first exposed to water, and decreases with time. If you’ve got two identical shots, but one has a higher TDS, you know it’s because the water interacted more fully with every coffee particle.
Now, who’s to say an even extraction is the be all end all? Maybe you want a bit more acidic or a little extra bitter. Complexity sort of sounds like the opposite of “completely even extraction”.
Hi James,
I was wondering: did you change the basket when moving from a 20 gr dose to a16 gr one? If not it might be the case that the shots were channeling due to an oversize basket for the dose
Looking forward to hear if you did and your thoughts about this possibility:)
you have a find solution?
I knew the moment I saw those papers that I would have some great videos from the coffee authorities to tune into. Thank you James.
Was reading v2 of your coffee atlas and in page 8 you use the American spelling for socialise ;( my heart is broken
I have brought a great shame to my family.
I'm not smart enough, nor do I have the equipment to do anything with this, but it's fascinating stuff and you asked me to comment so I did.
It ain't rocket science. You're just running hot water through ground coffee beans.
@@l.rongardner2150 you might be right. If you are, why does this bird have this TH-cam channel and why are you here?
Always love the hard science/food crossovers
Thanks for another well balanced and thoughtful explanation of your understanding of the original paper and your extensive experience, @jimseven
I wish I could like this twice. Oh yeah, Patreon!
Just saw someone on the tube watching this video; shout out to that guy for letting me know to check my notifications.
@James Hoffmann Try preventing channel formation by wedging the coffee between disks of coffee filter paper. It is called mechanical stabilization: th-cam.com/video/0olpSN6_TCc/w-d-xo.html and it might counteract the seepage: th-cam.com/video/eImtYyuQCZ8/w-d-xo.html . You could also try putting the coffee in an archimedes screw if that doesn't work, this would act as a cutoff wall as described in the video on dam seepage.
Any recommendations for how we can test this without a TDS meter? I sip my espresso while looking at the barista hustle compass and get a little overwhelmed haha.
But, if there's a dark roast I don't particularly like, I guess dialing in for 15s is now on the radar for a cleaner, sweeter version of it. Thanks for the video!
You could do a sudo TDS meter, given a constant and reproducible light source. Here is my idea: 1. brew coffee 2. use a eye dropper to place drop of coffee on a flat white surface 3. take picture of drop of coffee with reproducible exposure and aperture settings from reproducible angle and distance from coffee drop and light source 4. crop image in post to just the drop of coffee 5. do a signal count in black and white 6. lower signal = higher tds
@james: I was asking for this exact video on twitter, thank you so much!! Now I only started with espresso 1yr ago and still struggle as Randy said he is. BUT I WANT TO DO THIS AT HOME! Is it fair to say: grind coarser until the tasted strength of the shot drops, then fixate the grind setting & adjust brew ratio to your liking? Thanks for your help!
Also need to reduce pressure to 6 bars from 9
That was exactly my question. I'd definitely love to do these tests at home, with my medium tier grinder and machine but IIRC refractometers are really expensive, right?
I met one of the authors when I went to Oregon!!! Super cool to see this vid!
purely anecdotal and accidental: just before watching this I made an espresso with my robot using a slightly coarser grind that I usually do (exact repeatability is difficult with a stepless hand grinder, unfortunately) and it was obvious from the start of pressing my shot that it was going to be blurghie - the pressure was too low and finished in about 20 seconds. I was afraid to taste it, but it turned out not just better than I expected, it was actually tasty! This may be more a function of the robot itself, and cafelat does recommend pulling a lower pressure of 6-7 bars, but it was surprising and interesting nonetheless. Great coincidence that maybe this study explains it, at least in part?
Same applies to some extent to th ebrewing process of a Moka stove top. I prefer the taste from a much courser grind than is recommended. surprising but true.
_tastey point_
When you have so much to talk about coffee but none of your pals won't listen to you so you end up filming videos for youtube 😄
Anyone try this but without adjusting the brew pressure? Keeping pressure at 8-9 bars and brewing less coffee or more coarsely ground coffee for 15-ish seconds instead of 30 ish?
I just realized something during morning coffee that seems important... needs to be tested.
What about the bed size/depth in relation to the extraction?
I imagine the model uses a standard like 58mm or something as a constant.
We need to build differing sized systems to play with, see how really fine grinds do with say a wider 80mm or maybe a deeper but thinner 20mm puck.
I will need to read the article again but this sounds interesting.
(The thought was thinking about waters ability to extract being overcome by the width of the chamber ... seems almost more... arg!! English man? Parallel? (computer scientist here)
With Titus-Grinder and La Marzocco Strada EP-1 at the findest setting that still got a "drink", i had:
16% EY / with 16% TDS (but BR was 1:1) and i EXTRACTED KOFFEEOIL !
Flicker-Pics:
flic.kr/p/2inDWQH
flic.kr/p/2inBqS7
flic.kr/p/2inBqvv
flic.kr/p/2inF6qT
so it confirms: finer has it's EY limitation, and EY drops after a certain finess
@13:54 But the mathematical equation section is the real fun part :)
I believe that Adler of Aeropress holds the same idea. I watched one of Adler's very old videos and he proposes less time and less water with his Aeropress. I have recently watched Tetsu-san of 4:6 ratio V60 recipe and he said similar with his newest V60 Switch recipe, that is, to extract more, and to extract faster at the beginning of extraction by pouring water in circles through the open V60 and only close the switch after 40% of the water has been poured.
It sure seems as though many coffee experimenters are finding that the best flavours come from rapid first pass with as little water as possible.
I bet Matt Perger is feeling pretty good at the moment 😀.
Differential equations+multivariable calc+hoffman = meaning in my life
Saw this study when it came out, with all the fanfare James mentions, and told my friends to ignore it; the "model" developed in the paper was intended, from the start, to find a way to reduce coffee use in extracting espresso, not in making the best shot possible. James sees much more promise in this than I. If it pushes the discussion toward designing extraction methods that provide more homogeneous flow, and therefore more effective extraction and better flavor, then it will be useful.
i dont understand anything but somehow i can smell coffee
HI JAMES! this sound! Is perfect 👌, keep it this way, the gain .. Usually your video have inconsistent gain, make this video your benchmark
I will be looking into this myself with my flat burr grinders (I have two commercial units) and my commercial lever machine. Like one of the other commenters noted, Illy has standards that vary dramatically from what I hear so many people talk about in the US - namely the 7 gr shot. I was directed twenty years ago by the Illy rep in Poland to dial in my machine based on a 7-8 gram portion of coffee. I have stuck with this ever since and consistently get nice, rich, syrupy espresso, usually better than any local cafe serves.
Some thoughts to maintaining pressure on the puck would be to use slightly undersized baskets to maintain preload on the puck as it dissolves or to redesign the portafilter to incorporate a spring loaded, floating basket which could maintain an even upwards pressure on the basket as solids dissolve away.
Thanks for a great video!
Great vid James! Will you be at the WBC this year?
Probably not, I'm afraid.
This is cool to me but not very surprising. As simple as it seems this is a very complex topic. Not only do you have mass transfer factors at play here you also have fluid dynamics at play. Everything plays a role and in the end trial and error is really the only way to determine what is the best.
Incredible video!
Couldn't it be possible that the drop in pressure and the increase in flow rate is due to the puc providing less resistance as soluble particles are extracted, rather than the uniformity of the puc changes and channeling occurs.
Of course this is only the case when we are talking about a perfectly prepared puc.
This is how I am interpreting this phenomenon.
This is one of the most 'English' things imaginable.
What is the data outcome after two years of testing this? What have we learned? Can we apply something from the data? A follow up video would be appreciated.
Maybe a problem is that the puck in espresso machines is too big, i.e. has too much surface area, so it can break easily in many places. Maybe instead of one puck, there should be several smaller ones- a basket with several "chambers" instead of just one...
Thank you.
I would love to see this method in the dialing in espresso series. I have read the paper and while your explanation makes some sense as does others on HB, I would really like to see it; both with and without a refractometer.
Also I am not sure if the narrower Flair Pro portafilter effects the ideal dose. I know some of the maths focused on the a 58mm coffee bed.
This was a great video! As someone who knows more about science than coffee, your explanation of how science is done, and what we should learn from one paper is great!
This really reminds me a lot of something we (the coffee industry) seemed to stumble upon when EK43's were the new hot thing. At least in Melbourne Circa 2014/15, there was lots of excitement around pushing brew ratios. Recipes of 19/20 grams in, 50-70 grams out in 20 seconds. Some of the best espressos I ever had were made like this.