Espresso crema and flow profiling

Beginner and pro baristas share tips and tricks for making espresso.
Molina
Posts: 28
Joined: 5 months ago

#1: Post by Molina »

Hi all,

Lately, I've been playing around with flow profiling on my Bianca. I'm learning a lot and understand the relation between pressure, (puck) resistance and flow. So far I'm getting interesting results with blooming style espressos. Certainly not perfect but my last shots seem to have reduced astringency and are more balanced overall.

I noticed that I'm losing the thick crema on top compared to the shots I made with a standard 9 bar profile. Not complaining since I enjoy the taste but wondering what's causing the thin, sometimes non existent, layer of crema. It's more noticeable when I make 2 cups at once in the morning with a spouted portafilter. The bottomless portafilter for 1 cup seems to add more crema.

Overal I saturate the puck at start. Quick buildup of pressure (3-6 bar) then reduce flow completely. Playing around with blooming time (currently between 5-20s). Then I ramp up to full flow (measured around 6 g/s) at 9 bar with 2 g/s output flow. Lastly, I'll reduce flow to keep 2 g/s output flow. Pressure will reduce and I end up between 4-7 bar.

Personally I like what I'm getting out of the machine more than without flow profiling but don't understand why I'm losing the crema. Pressure is still there right? The same beans will create thick crema when pulled with 9 bar throughout the shot.

LucaFg
Posts: 95
Joined: 4 months ago

#2: Post by LucaFg »

A rough relation between Resistance, Flow and Pressure is the following:

R = P/F^2 , Resistance = Pressure/square Flow. The square is useful because R is not really constant during extraction, so the square flow compensates for. Obviously there are more complex math models using differential equations, but this is the easiest and help to understand what's happening during extraction.

Having lot of crema is a matter of difference of pressure between pressure on the puck and pressure of the air and how much such difference persists. The more the difference and the more it persists, the more CO2 is first compressed and then released with more energy, the more and consistent and durable the cream. Obviously if difference is too much, you risk to have too big bubbles. But it's not happening in the 9-12 bar range, typical of espresso machine today.

Obviously it depends on roast and kind of coffee. The more roasted the beans (but not burnt beans), the more cream. Usually robusta gives more cream than arabica. Light roasts give less crema.

The blooming or turbo shot style coffee is well known to have very low crema.

JRising
Team HB
Posts: 3723
Joined: 5 years ago

#3: Post by JRising »

From what is the CO2 released?
Is the CO2 trapped in the grounds before the pressure somehow releases it rather than forcing it to be further trapped?

Capuchin Monk
Posts: 1282
Joined: 15 years ago

#4: Post by Capuchin Monk »

Not just CO2 but air in general.

User avatar
Jeff
Team HB
Posts: 6929
Joined: 19 years ago

#5: Post by Jeff »

Before parroting Darcy's law, you should catch up on a few years of discussion. There was quite some discussion over if the term should be squared or not in relation to espresso extraction. JoeD has done some interesting work that suggests that neither holds well. The "puck resistance" line has been accepted as pretty much bogus for much more than watching its general trends during the shot.

Edit: "The square is useful because R is not really constant during extraction, so the square flow compensates for." would be called "hallucination" if you were a generative AI bot.

Crema is little more than espresso foam. Its presence or lack is no longer believed to be related to the quality of the shot, except for those that doggedly adhere to the Italian marketing definitions and those that derived from them. In fact, it tends to be bitter and, for many, detracts from the overall taste.

LucaFg
Posts: 95
Joined: 4 months ago

#6: Post by LucaFg »

For what I know, the reactions during the roast create the CO2 (I think it's called carbonization?), that's why the more roasted beans come with more crema.

Hot high pressured water extract CO2 and oils from the ground coffee; before exiting the PF, the extracted coffee is at high pressure, so the CO2 is compressed with the oils (emulsion). When it exits the PF, CO2 expands and starts to diffuse, however viscous oils get some CO2 trapped, that's how foam is formed. You have more crema if you grind your beans finer (so you can extract more oils and CO2), if you use hotter water, if you reach higher pressure for more time (it helps to extract oils and CO2 and compress the emulsion), if your beans are fresh (after roast, CO2 in the roasted beans starts to diffuse into the air).

Lately I managed to have crema not at all bitter but rounded and balanced. I love thick coffee, the intense mouthfeel, the silky sensation on tongue and palate!

LucaFg
Posts: 95
Joined: 4 months ago

#7: Post by LucaFg »

Jeff wrote:"The square is useful because R is not really constant during extraction, so the square flow compensates for." would be called "hallucination" if you were a generative AI bot.
Oh I just talked to someone working at Decent that suggested this simple model and told me what you call hallucination is used to draw the resistance graph in Decent machines; obviously it's a simplified model, as I told. Just to understand what's happening at basic level. I don't think the OP is searching more complicated models based on differential equations.

User avatar
HB
Admin
Posts: 22028
Joined: 19 years ago

#8: Post by HB »

LucaFg wrote:Hot high pressured water extract CO2 and oils from the ground coffee; before exiting the PF, the extracted coffee is at high pressure, so the CO2 is compressed with the oils (emulsion). When it exits the PF, CO2 expands and start to diffuse, however viscous oils get some CO2 trapped, that's how foam is formed.
I am skeptical that CO2 plays a significant part, if only because degassed coffee can still produce a lot of crema, if the pressure is high enough. To see what I mean, try grinding some coffee, letting it sit for 5 minutes to allow any CO2 to dissipate, then see how much crema is produced. I'm not a food scientist, but I've always assumed the crema is an emulsion of oil and water.

Robusta coffees in particular can produce mountains of crema, even if they are months post-roast.

UPDATE to mention that James Hoffmann's video corrects my erroneous assumptions above. Nevermind! :lol:
Dan Kehn

LucaFg
Posts: 95
Joined: 4 months ago

#9: Post by LucaFg »

In my experience I noticed:
The same beans give less and less crema month after month.
Some kinds of robusta give more crema. That's the kind of robusta I'm hunting lately! :-)
The lower the pressure, the lower the crema. I can see the difference between 9 and 6 bar on the Decent.
Light roasts give few vanishing not too thick crema.

I found similar results and convincing explanations in reliable sources on the web. However I think there is still to study and learn about crema and coffee in general. Yes, for sure oils (emulsion) have the lion part in the process, you're absolutely right! That's why some kinds of robusta give more crema!

Happy new year Dan!

User avatar
RapidCoffee
Team HB
Posts: 5017
Joined: 18 years ago

#10: Post by RapidCoffee »

Molina wrote:I noticed that I'm losing the thick crema on top compared to the shots I made with a standard 9 bar profile...
I like what I'm getting out of the machine more than without flow profiling but don't understand why I'm losing the crema. Pressure is still there right? The same beans will create thick crema when pulled with 9 bar throughout the shot.
Hi Mark. In my experience, for a given bean, rapid puck saturation and high max pressure tend to enhance crema, while slow puck saturation, low max pressure, and blooming profiles tend to reduce crema. I still get plenty of crema from lever (declining pressure) profiles. Based on this, I'm guessing your reduced crema is due to the blooming step. Eliminate the bloom and see what happens. Here are a couple of profiles to try:
1) straight lever profile (no bloom): fast saturation, gradually decrease pressure from 9->6 bar during the extraction
2) Jim Schulman's Bianca profile: fast saturation, reduce flow briefly as soon as pressure rises to 9 bar, increase flow back to 9 bar when you see the first drops appear, gradually reduce pressure for a slowly declining flow

End of the day, taste wins. If you enjoy your profile, don't worry about crema!
John