I started an important background thread HERE, to discuss what exactly is the role of "fines", whatever those might be, within the puck.

-the Inquisitive bean

The_Mighty_Bean wrote: Matthew - did Chris answer your question?

My question regarding the meaning of this? -

The_Mighty_Bean wrote: It's documented on this site, by certain extremely experienced and well-regarded members, that hand distribution is unreliable and difficult

No, not really.



.

The_Mighty_Bean wrote:
Now Andrea Illy goes and complicates, perhaps even destroys, this whole theory, by telling us this interesting tidbit, on page 215, crudely summarized below:

An espresso puck is a heterogeneous ("plurimodal") mix of coarse and fine particles. The coarse bits create a structure that limits the water flow. The fine bits are what contain the flavor components. Therefore, a distribution that homogenizes the puck or separates it by particle size is going to be undesirable....


You'll still need a light tamp. Illy says so. How hard doesn't really matter, you just need enough pressure to release the oils and stick the particles together, which creates a more even resistance against the water flow. Sorry, no time to go back and find the page but it was contained within the 2nd or 3rd page of the search results for "distribution".

I went back and looked. These concepts are more fully discussed on p277-78 of the Illy book. Currently being discussed on the "role of fines" thread that I spun off of this one.


~tMb

P.S. Matthew, I spent a ton of time searching but I just can't find one of the two quotes I remember. Please forgive me for not linking to this first one, but in the experiment that Ken and Jim ran on the role of preinfusion, Ken stated that Jim did the dosing, because Ken could not imitate his technique.

Okay, when two people as experienced as Ken and Jim agree that their difference in dosing/distribution method is substantial enough that it's too inconsistent to use in a statistical comparison of shots, that tells me that, for sure, you and I and Joe amateur-home-barista are doing different things to the beds of ground coffee we're working with.

In other words, distribution techniques are unreliable as a tool of comparison, and make a difference in flavor.

There's a second post that I couldn't find, and I could swear it was Ken or Jim again, and it was, once again a comparison thread. Might have been somewhere in the Titan grinders mega-discussion. Anyway, the person posting made a complaint to the effect that the distribution was less than ideal because it was inconsistent between testers.

The only way to standardize the compaction aspect of distribution is to get it down to a fixed pre-tamp density. Once it's there, if we use the same tamp (e.g. Espro or Autotamper), then we are working with samples that are identical in that respect.

I also am intuiting that this max-density, evenly-distributed puck is going to consistently taste and pour better.

There's much experimenting to be done, for those inclined to do so, to establish whether there's any merit to my crazy little idea.

~bean

i think based on that... that it is inconsistent from one person to another.

..but i bet the same person would be consistent with them self.

jon

The_Mighty_Bean wrote:I have the same problems with shaking the basket. You end up with a nice dense puck but the distribution is uneven

how do you know?

The_Mighty_Bean wrote: Okay, when two people as experienced as Ken and Jim agree that their difference in dosing/distribution method is substantial enough that it's too inconsistent to use in a statistical comparison of shots, that tells me that, for sure, you and I and Joe amateur-home-barista are doing different things to the beds of ground coffee we're working with.

In other words, distribution techniques are unreliable as a tool of comparison, and make a difference in flavor.

When the variable being tested is not grounds distribution, it makes sense that the distribution being used should be consistent. That does not necessarily mean that interpersonal variations in distribution impact the cup quality, only that it was a test of Variable A and not Variable B, and by holding Variable B constant, they reduce any impact that it might have so as to make the statistics on Variable A valid.

Further, as much as I like Ken and Jim, and they're both good friends, they really aren't that experienced when it comes to making espresso. Perhaps as home enthusiasts, they rank right up there, but I know plenty of folks who make more espresso drinks in a week than those guys make in a year.

I also am intuiting that this max-density, evenly-distributed puck is going to consistently taste and pour better.

but haven't you already declared that this isn't going to be an "evenly-distributed puck" by forcing the fines to separate from the coarses in the matrix through vibration?

The_Mighty_Bean wrote:It's documented on this site, by certain extremely experienced and well-regarded members, that hand distribution is unreliable and difficult

Matthew Brinski wrote:... could someone please help me understand the rationale behind this statement?

At first glance, I was confused - i mean, how difficult is distribution? But then i compared it to most other steps of pulling an espresso, and it seems that there is probably the most "art" and least "science" associated with distribution. Compare it to: grinding (done by a machine), tamping (generally agreed as less important than distribution) and pulling the shot (done by a machine). Even with variables where experience counts - choosing the grind, deciding how hard to tamp, and deciding what temperature and pressure to pull the shot at - are more about decisions (setting the grinder/machine) than actual technique. The prevalence of techniques associated with distribution - the stockfleth move, the WDT, etc. - show how much of an art distribution is.

So, comparatively, i would argue that hand distribution (as opposed to the proposed vibratory distribution) is unreliable and difficult. Perfect distribution is very complex. Even our rudimentary measurements of distribution - through the absence of channels, through taste, or through the use of a bottomless portafilter - are just that, rudimentary.

Undoubtedly, most of us are quite good at distribution. But we've gotten there through much trial and error. And none of us can claim to be perfect at it. In fact, further improvements are elusive, because we are dealing with very small particles (microns in size?) with very rough tools (fingers and needles). If distribution is currently an art, we are very good artists. But there's no need to be afraid of science. Let science proceed, and those who wish to use it's tools do. Personally, i keep my fingers crossed. As long as my art is better than the science of the superautomatic, I think the end of the barista is nowhere near.

RegulatorJohnson wrote:i think based on that... that it is inconsistent from one person to another.

..but i bet the same person would be consistent with them self.

jon

As we attorneys (well, future attorneys in my case) like to say in our pleadings, "I hereby incorporate by reference" what Popeye just said.

We are dealing with very small particles, with huge and clumsy tools. You're never going to push around a hundred thousand microparticles in the exact same way.

A practiced, competitive barista will be most consistent, a skilled home user less so, and so on down the line. Not one of them will be consistent with another.

I've been at this for over a year, I make several shots a day and it is patently obvious to me, each and every time I make a shot, that sometimes there's a few more grinds pushed into a "hole" that I need to fill, and sometimes a few less.

So there's definitely variance in distribution. The question is how much variance makes a difference that you can see in the pour and, more to the point, taste?

But, if I may flog the dead horse once again, if you get that puck to max tapped density, then that variance is drastically reduced, because the tapped density is supposedly constant for a given substance.

Once you have that consistent puck, it's easy to test and adjust for flavor issues.

tmb

barry wrote:how do you know?

It's visible on inspection. The grounds in the basket heap up on one side or form peaks and valleys, and grounds in the shaken basket settle in lower than if it had not been shaken. It's like tapping- it leaves visible room for updosing.

barry wrote: When the variable being tested is not grounds distribution, it makes sense that the distribution being used should be consistent. That does not necessarily mean that interpersonal variations in distribution impact the cup quality

.

The test was for effect of preinfusion on cup quality. If individual distribution technique wasn't at least thought to possibly have an outcome on the final cup, there would have been no need to standardize it. If its not a variable, then there's nothing to worry about. --edit-- You are right, I stated it above as a fact. It -is- an established fact if you talk about a newbie's distribution attempts versus an accomplished barista. One you get to higher and higher levels of competence in the art, the differences grow smaller and possibly at some point become negligible. Again I refer to Popeye's post.

Further, as much as I like Ken and Jim, and they're both good friends, they really aren't that experienced when it comes to making espresso. Perhaps as home enthusiasts, they rank right up there, but I know plenty of folks who make more espresso drinks in a week than those guys make in a year.

True, but if it takes the likes of Chris Tacy to get consistent distribution by hand then it's probably not a bad idea to look for new methods that can give folks like you and me a helping hand.

barry wrote:but haven't you already declared that this isn't going to be an "evenly-distributed puck" by forcing the fines to separate from the coarses in the matrix through vibration?

No, well, maybe. LOL. I'm figuring this thing out as I go, not declaring laws. If I declare something as a rule, then I've cited a source for it (I hope).

I'm trying to figure out if there's an "ideal" intra-puck distribution of fines, which is why I spun off that other thread. And I have also suggested a way to get consistently dense pucks. I'm not sure how the fines would be distributed at max tapped density.

There's a whole lot I don't know, and some of it may be a lot easier to solve experimentally. Thanks for your thought-provoking response. Also, I need to get my hands on that Illy book.


-bean, confusedly

The_Mighty_Bean wrote:True, but if it takes the likes of Chris Tacy to get consistent distribution by hand then it's probably not a bad idea to look for new methods that can give folks like you and me a helping hand.

Um, dude ...

Not that it automatically makes him correct, but do you understand that you're having a conversation with an established USBC judge? I just thought I would point it out after reading the "give folks like you and me a helping hand". (And I've been told that he's pulled a few shots)

... he's gotta be pretty cool, 'cause he has designated helicopter parking at his coffee shop.


Matt