Grinder data

Cuprajake

https://coffeeadastra.com/2023/09/21/what-i-learned-from-analyzing-300-particle-size-distributions-for-24-espresso-grinders/

MediumRoastSteam

Cuprajake - I’ll need a PHD and a whisky to read all of that. 😉

DavecUK

MediumRoastSteam I’ll need a PHD and a whisky to read all of that. 😉

Felt the same…the guy isn’t going to really help many people with that…

delta76

it’s as @Cuprajake said - just data. From data to analysis, linking particles distribution with espresso quality (if any) and finally recommendation, is a long, long way

MWJB

delta76 Do you mean, ‘a long way off.’?

delta76

MWJB I just said “a long way”

most people don’t care about particles distribution, they care about espresso quality. the tricky part is that espresso quality is subjective, so you will need to do blind test with large enough sample size, to see if particles distribution has anything to do with perceived quality. then you have different type of coffees to test for - does light roast perform the same as medium, or does it do differently on same distribution curve etc.

I’m all for objective, measurable data, but this is just a start.

MWJB

delta76 The Kaffeemacher youtube article & blog post is better start point. Go there first, or scroll to the TL:DR in Gagne’s article.

If you don’t care about particle distributions then use a blade grinder or pestle & mortar :-)(e.g. something abnormal)

The TL:DR is that grinders make different distributions but that a single aspect/quality can’t point you to Nirvana - grinders have a wide range of grind attributes, but considering a normal sample, a range of different grinders work re. tightness of distribution, burr size & speed, preference etc.

…which, for me, was pretty much where we were before this.

delta76

MWJB I actually read Kaffeemacher reviews of grinder quite intensively (limited by the translation). that and the article in question still do not have a strong model for particles distribution and espresso quality. Don’t get me wrong, it is very nice to have such data, it is many steps into right direction, but it is not there yet.

Take audio, for headphones we have Harman curve that most people will prefer than other curves. Will there (ever) be a curve for particles distribution that most people will prefer than other curve? Things like that.

JonWoo187

I gladly say that espresso is like roasting in that it’s ⅓ skill, ⅓ art form and ⅓ science. Those that try putting it all (or mostly) on science are simply missing out. Technology can be quite awesome in many ways, but I’ll never trust numbers over my senses at the end of the day. That approach has never let me down although I’m missing a sense now, bwahaha…

DavecUK

JonWoo187 I gladly say that espresso is like roasting in that it’s ⅓ skill, ⅓ art form and ⅓ science. Those that try putting it all (or mostly) on science are simply missing out.

Completely agree!

MWJB

delta76 I agree with you in essence. The data is what it is, it’s neither nice nor otherwise though. If none of it (distribution) mattered at all you could get comparable results by wrapping beans in a tea towel & smash them with a hammer (rather than using normally functioning, tested & well used grinders making expected distributions that achieve predictable goals).

Basically, there is a range that works through experience & design, distributions can be different (measurably) and still work. For distribution to be something that needs no consideration, anything would then work. But we don’t use anything, we use tools designed for the job and they work. Who’d a thunk? :-)

Loonster

I would expect that the shape of the ground particles would have a large impact. I have not seen anything analyzing the different shapes that grinders produce.

MWJB

Loonster There was some study into that with the development of the Mythos Clima Pro, look for a presentation by Colin Harmon on youtube.

Ultimately though consider that grinders make particles of many different shapes in the same grind and that there are so many particles at each grind size interval, that they likely average out (to normal expectation) at each grind size interval?

Cuprajake

MWJB dr mr harmon present a curve?

MWJB

Cuprajake Unfortunately the video seems to have been taken down now, previously hosted on Tamper Tantrum.

Yes, he (or, more precisely the scientists employed by Nuovo Simonelli) provided curves, as there was more than one burr set considered and compared. Be aware that laser diffraction curves are extremely difficult to compare if carried out on differing bits of kit, ideally you should compare to curves on the same model diffractor.

In summary, I seem to recall the flat burrs chosen were seen to produce ‘rounder’ particles rather than than more rugby ball shaped particles from conicals after imaging analysis.

Cuprajake

MWJB

Giphy - Jimmy Fallon I Had No Idea GIF by The Tonight Show Starring Jimmy Fallon

i was making a audio joke

MWJB

Found it:

http://www.tampertantrum.com/developing-the-mythos/

Loonster

I haven’t watched the video yet, but will do so soon.

My thoughts on particle shape for coffee being important is based on readings of grain shape for gunpowder. Just replace burned with extracted and the parallels will be more clear.

For ball shaped gunpowder, the exposed surface area decreases as the propellant is burned. These are called regressive grains.

For tube shaped gunpowder, the exposed surface area stays relatively constant as it is burned. These are called neutral grains. (Note: interestingly blood cells also have a relatively neutral shape).

There can also be progressive shapes, but I don’t see how they can be created in coffee.

Flat disks would also be regressive, but not as much as balls.

Ideally, I think we would want a more neutral grain shape for coffee. Failing that, I think we would want to adjust flow in relation to how much extraction is taking place (hence the popularity of decreasing pressure profiles).