Why Does a Long Pre-Infusion and Extraction Work? - Page 2

Beginner and pro baristas share tips and tricks.

#11: Post by lucasd »

LR is able to do most steps of it, with exclusion of slow pre-infussion (e.g. 10ml/s).
I did various experiments and so far have not found specific benefits for blooming shot.
The result are not bad, but not better than standard holding 4bar till drop.
Extraction yields are also comparable.

I'm talking about filter scandinavian coffee roast...


#12: Post by CafeVenice »

cskorton wrote:Thanks! Yeah, I guess I'm trying to better understand the why vs the how as far as brewing parameters go. Questions like how long preinfusion? When to know to increase/decrease it? Same goes with brew time, and various pressures. For example, for roast level x and bean type y initially try brew profile z.

I know it's an impossibly complex and deep question, but what I'm trying to do is minimize bean waste so I'm not burning through a half a bag of beans in a day or two trying to optimize taste.
Hi, you're asking great questions. I believe that extraction (body) is determined by the brew pressure experienced by the puck. If you want to increase extraction, increase the dose (don't change the grind level assuming you're at a reasonable grind level). The higher dose puts up more resistance and increases the brew pressure. Play with the yield (grams out) to find the sweet notes. If you repeat this process at a coarser grind, you can dial it in but the shot will be more acidic (which is nice in a different way).

Some baristas online were able to set the pump pressure, and have a grinder, that make the shot look great while extraction is optimal. For my equipment, I get the optimal extraction when the shot is slow and doesn't look that great being pulled. I did pull shots that looks like the pros but they were severely under extracted.

Baratza: skilled in the art of grinding
Sponsored by Baratza
User avatar
Supporter ♡

#13: Post by baldheadracing »

cskorton wrote:... I know it's an impossibly complex and deep question, but what I'm trying to do is minimize bean waste so I'm not burning through a half a bag of beans in a day or two trying to optimize taste.
Forgot to mention that going to a single basket will give you twice the number of shots, maybe more. I make 7g singles.

However, most single baskets can be challenging to use, and can yield different/weird results compared to the same coffee/grind in a larger-capacity basket. I've tried a number of single baskets, and this setup from Tidaka is by far my favourite (site is in German, but Google translate works) https://www.tidaka.net/de/tamper/set-an ... grale.html

(Aside: I actually use a VST 7g basket (which Tidaka also sells) instead of the LF clone of the (no longer available AFAIK) La Marzocco Strada basket that Tidaka supplies with the set. I use the VST so I can switch to a 14g in a VST 15 basket without changing the grind setting.)
What I'm interested in is my worst espresso being fantastic - James Hoffmann

User avatar
Supporter ❤

#14: Post by Peppersass »

Jeff wrote: 5) Ramp down the pressure to keep the flow rate where you want it

#5 is still something of voodoo or religion. I think that "constant flow" is something that emotionally feels good, but I haven't seen any well-controlled studies that say that it is "best". My personal experience is that a "flat, 9-bar" profile doesn't taste as good as a declining-pressure profile (with reasonably constant flow in the later portions), to me, with my coffees, my tastes, and the grinders and machines I've used, both pump-driven and lever.
I agree that there isn't scientific evidence that a constant flow rate is the best profile during the back end of the shot. However, I think it's pretty clear that, as you say, "a declining pressure profile (with reasonably constant flow in the latter portions) usually produces the best taste. I'd add that this is often true for light, light-medium and many medium roasts, but not likely to be true for medium-dark and dark roasts, which are likely to overextract when the flow rate is reduced too much.

I think there's a reasonably simple explanation for why a declining pressure profile works: If you don't ramp down the pressure during the back end of the shot, the flow rate will increase, often dramatically. This is happening because the passage of water through the puck is making it more permeable. I wouldn't go so far as to say the puck is "breaking up" but that the grounds are being pushed apart as the puck swells.

Reducing pressure, and consequently the flow rate, increases the contact time between the water and the grounds, which in turn increases extraction. For light, light-medium and many medium roasts, more extraction produces better taste. It's rare for roasts like that to over-extract.

But does the flow rate have to be constant? Why not keep reducing the flow rate? In theory, that should maximize contact time. But in practice I find that as I try to push the flow rate lower and lower, below the nominal constant flow rate I like (about 50-60 ml/min for a double, 20-30 ml/min for a single), the stream from the puck turns into a series of slower and slower drips. Certainly the contact time is greater, but I haven't found that the extraction increases appreciably. So I just try to maintain a constant flow rate as best I can.
★ Helpful


#15: Post by jmotzi »

A helpful explanation is provided here, hidden away in a machine related thread, but not machine dependent at all :) "A unified theory of espresso making recipes" Decent Espresso news
LMWDP #662

cskorton (original poster)

#16: Post by cskorton (original poster) »

Thank you all for your insights. And that Decent thread is an excellent resource to read and watch through. I just wish John would go more I depth about the theory and execution behind the Londinium profile, haha.

I'll try summarize what I learned and see if I'm missing the forest in the trees:

It all boils down to a bean's roast level. Lighter roasts can handle and need longer preinfusion and brew times. That's the reason why my shot was so good at such outside of the box brew parameters, it must be a medium light roast (I thought it was medium). So my initial shot I mentioned landed somewhere in between a traditional lever and Jeff's blooming shot he mentioned.

The same thing wouldn't work with more traditional dark roasted Italian blends. These blends are easier to extract and are of lower quality (mostly). The harder you push these beans, you may go into over extraction territory, producing bitter, ashy tastes. That, and the genera lower quality means they have less "good" taste to give compared to lighter roasts.

Further, most machines are only capable of pulling shots that preinfuse for up to 8-10 seconds, then extract either at a full 9 bar or declining pressure profile. This is simply because of tradition and it's what works best for the medium to dark roasts that were traditionally used. Higher quality beans at lower roast levels require different parameters.

I think that's in it in the nutshell, though I still feel like I'm not sure how to put this into action in the real world outside of very vague terms. I was hoping to find more concrete guidance. Like, decrease preinfusion by x seconds of roast level goes up by y units or something.


#17: Post by jmotzi »

Since you are located in Philadelphia, you might want to try Elixr. They are a great roaster and typically have someone in the annual national barista competitions. They don't do the dark stuff however if that is your preference. Before social distancing they were offering an "Omakase" tasting bar at Elixr Roastery (N 12th St) on Saturday mornings. Hosted by one of their expert baristas, it's a great way to learn and taste with 5 other people. Anyway they have great coffees.

Also worth checking out is Brandywine down in Wilmington. I don't think they have a café but they also do great things with coffee.

I like to give business to the local roasters.

LMWDP #662

Versalab: maker and supplier of finest espresso equipment
Sponsored by Versalab
User avatar
Supporter ♡

#18: Post by baldheadracing »

So ... it was bugging me that Mr. Gunson's recommendations were the opposite of mine for pre-infusion pressure. So I tried it his way. Here's what I think - again, just on lever machines (as both pre-infusion time and pre-infusion pressure will affect peak brewing pressure in a spring lever), and I am assuming high-grown hard-bean specialty-grade coffee:

For the usual pre-infusion time of 5-10 seconds, the lighter the roast, the higher the pre-infusion pressure, for example, 6 bar on light roasts, 1 bar on 'traditional' espresso blends.

For a long pre-infusion of, say, 30 seconds, a higher pre-infusion pressure like 6 bar needs an incredibly fine grind to last 30 seconds without starting extraction (beyond sweating), which leads to what is called "back-end blow-out" where the visible flow greatly increases early in the extraction, early blonding, etc. On the other hand 30 seconds at a lower pre-infusion pressure like 1 bar does not need as fine a grind as a higher pressure, helping to prevent back-end blow-out. It also results in a lower peak pressure on my Strega, which also is desirable on lighter roasts; I don't know about the Londinium.
What I'm interested in is my worst espresso being fantastic - James Hoffmann


#19: Post by okmed »

Craig, is your Strega supply plumbed in and are you using the pump ?

User avatar
Supporter ♡

#20: Post by baldheadracing » replying to okmed »

It isn't plumbed in.

I have a pressure gauge installed on the empty port in the feed between the HX and the group, so I know the pre-infusion pressure. (I don't have extraction pressure once the lever is lifted to the point that the flow from the HX to the group cylinder is cutoff.)

I do have the pump on a dimmer control, but I have ended up using the dimmer as a glorified pump on/off switch. The same thing can be done by lifting the lever just enough to turn the pump off; I just find it easier using a separate control.
What I'm interested in is my worst espresso being fantastic - James Hoffmann