| |
Main
Date: 14 Jun 2006 08:50:05
From: dekaysion
Subject: Espresso based lab experiment for a university class
|
Over at work we are playing with the idea of creating a new lab
experiment for our chemistry and chemical engineering students --
involving making espresso.
>From an ideal standpoint we would like to start with roasting beans (we
have some roasters in our food chemistry department), move on to
grinding (different grinder types and settings), tamping and
extraction.
The aim is to have about 50 groups of students per semester to do
separate conditions and determine characteristic measurements of each
step.
e.g. Particle size distributions of different grinder types and
settings; extraction volume (crema and liquid) at constant time or time
at constant volume; etc. Unfortunatley taste profiles will be difficult
to measure, although we can certainly find a way here, too.
The whole set of experiments will, as each setting and data set is done
by a group, be analyzed using statistical methods and design of
experiments.
Does anyone here have any ideas or suggestions what else might be
useful to do or measure? Would anyone know of a machine setup that
would be suitable? Would anyone know of contacts in the coffee industry
interested in sponsoring or aiding such a project?
Thanks for reading so far ;)
Dan
|
|
| |
Date: 14 Jun 2006 10:26:01
From: Ken Fox
Subject: Re: Espresso based lab experiment for a university class
|
"dekaysion" <daniel.kaesmayr@gmail.com > wrote in message
news:1150300205.790595.133150@h76g2000cwa.googlegroups.com...
> Over at work we are playing with the idea of creating a new lab
> experiment for our chemistry and chemical engineering students --
> involving making espresso.
>
>>From an ideal standpoint we would like to start with roasting beans (we
> have some roasters in our food chemistry department), move on to
> grinding (different grinder types and settings), tamping and
> extraction.
>
> The aim is to have about 50 groups of students per semester to do
> separate conditions and determine characteristic measurements of each
> step.
> e.g. Particle size distributions of different grinder types and
> settings; extraction volume (crema and liquid) at constant time or time
> at constant volume; etc. Unfortunatley taste profiles will be difficult
> to measure, although we can certainly find a way here, too.
> The whole set of experiments will, as each setting and data set is done
> by a group, be analyzed using statistical methods and design of
> experiments.
>
> Does anyone here have any ideas or suggestions what else might be
> useful to do or measure? Would anyone know of a machine setup that
> would be suitable? Would anyone know of contacts in the coffee industry
> interested in sponsoring or aiding such a project?
>
> Thanks for reading so far ;)
>
> Dan
>
I think it depends on what you want to come out of this, i.e. is the purpose
to make practical and novel discoveries and then to publish them, or is the
idea to teach aspects of the scientific method to your students by using a
subject (coffee) that they might find interesting? If the latter, then it
probably doesn't matter much what you study as long as the procedures and
observations are repeatable. If the former, then you need to determine what
you would want to prove or disprove.
The "scientific method" is (as you know) but a tool for proving or
disproving hypotheses. Although serendipidous findings do occur, they
cannot be counted upon and the risk of proceding without an idea of what you
are trying to evaluate is that you will find nothing worth evaluating
further. For example, I'm sure that particle sizes and distribution are
important but the more you think about them the more complicated the issue
is. Taking this example, you could have two different grinders with
different burrs and it could be that one of them makes more fines than the
other but because of the design of the doser the distribution of the fines
in the puck could be entirely different and could negate the chance to
observe any differences in the cup between the two grinders. Also, you have
the issue of tremendous intrashot variability that is part and parcel of
espresso making; throw in a bunch of amateur student baristas and this
variation will increase.
This lab experiment could be popular with the students or it could be a real
dud; there is no way to know until you run a few classes through it, so you
have to be prepared for the fact that it might not be around very long.
So, assuming that you want to discover new and significant things for later
publication, I would suggest choosing some very simple ideas to test and let
the "results" obtained dictate further directions for the research. You
could always start with a simple accepted observation (grind must change as
ambient humidity changes, as an example), then try to find out why this is
the case. Taking this example, you could obtain a quantity of a single
varietal bean (to keep it simple), roast it to two different levels, age the
coffee for several days (to mimic real life usage in a good cafe), then
using a single grinder measure both the time volume characteristics of shots
and look at the grinds in terms of particle size distributions as changed by
humidity in a climate controlled room. Obviously, you'd have to keep all
the other factors constant such as dosing, espresso machine used,
distribution and tamping, etc. There are many other simple observations
whose cause is as yet obscure, and I'd start with this sort of thing rather
than just trying to document something that may or may not be relevant to
making good espresso.
As to equipment to use that would of course depend on your budget. A good
quality e-61 semi commercial machine and a Mazzer Mini grinder should offer
the sturdyness and dependability you would want, or you could go with
commercial level equipment. You don't want to buy anything that is so
delicate that your students will break it.
As to contacts in the coffee industry, you might have a local roaster you
can work with, and of course there is La Marzocco, a company which seems to
like dealing with consumers more than most.
ken
|
| |
Date: 14 Jun 2006 17:12:40
From: Dan Bollinger
Subject: Re: Espresso based lab experiment for a university class
|
> Over at work we are playing with the idea of creating a new lab
> experiment for our chemistry and chemical engineering students --
> involving making espresso.
>
>>From an ideal standpoint we would like to start with roasting beans (we
> have some roasters in our food chemistry department), move on to
> grinding (different grinder types and settings), tamping and
> extraction.
>
> The aim is to have about 50 groups of students per semester to do
> separate conditions and determine characteristic measurements of each
> step.
> e.g. Particle size distributions of different grinder types and
> settings; extraction volume (crema and liquid) at constant time or time
> at constant volume; etc. Unfortunatley taste profiles will be difficult
> to measure, although we can certainly find a way here, too.
> The whole set of experiments will, as each setting and data set is done
> by a group, be analyzed using statistical methods and design of
> experiments.
>
> Does anyone here have any ideas or suggestions what else might be
> useful to do or measure? Would anyone know of a machine setup that
> would be suitable? Would anyone know of contacts in the coffee industry
> interested in sponsoring or aiding such a project?
>
> Thanks for reading so far ;)
>
> Dan
Dan, I would get the U. to subscribe to the Association Scientifique
Internationale du Café, http://www.asic-cafe.org/
No use reinventing the wheel. And, this is THE peer-reviewed journal for
coffee. These are the article you need to cite. And, eventually you'll want to
publish.
Offhand, I would suggest a semi-automatic 4 group machine. They are available
used.
Also, you will want to buy a large quantity of espresso blend green beans so all
your testing is done using one lot of one blend. I suggest Sweet Marias Monkey
Blend. (Do not get their Liquid Amber blend, it separates because of differing
bean density).
Dan
|
| |
Date: 14 Jun 2006 13:10:28
From: Omniryx@gmail.com
Subject: Re: Espresso based lab experiment for a university class
|
As a teacher of research methods these last 30 years, I concur with
Ken's recommendations entirely. Identify a limited number of variables
and control the others as closely as time and budget will allow. This
is a great opportunity for students to explore the concepts of
systematic and stochastic error, etc.
Good luck. Let us know how it comes out.
Will
dekaysion wrote:
> Over at work we are playing with the idea of creating a new lab
> experiment for our chemistry and chemical engineering students --
> involving making espresso.
>
> >From an ideal standpoint we would like to start with roasting beans (we
> have some roasters in our food chemistry department), move on to
> grinding (different grinder types and settings), tamping and
> extraction.
>
> The aim is to have about 50 groups of students per semester to do
> separate conditions and determine characteristic measurements of each
> step.
> e.g. Particle size distributions of different grinder types and
> settings; extraction volume (crema and liquid) at constant time or time
> at constant volume; etc. Unfortunatley taste profiles will be difficult
> to measure, although we can certainly find a way here, too.
> The whole set of experiments will, as each setting and data set is done
> by a group, be analyzed using statistical methods and design of
> experiments.
>
> Does anyone here have any ideas or suggestions what else might be
> useful to do or measure? Would anyone know of a machine setup that
> would be suitable? Would anyone know of contacts in the coffee industry
> interested in sponsoring or aiding such a project?
>
> Thanks for reading so far ;)
>
> Dan
|
| |
Date: 14 Jun 2006 14:00:30
From: jim schulman
Subject: Re: Espresso based lab experiment for a university class
|
On 14 Jun 2006 08:50:05 -0700, "dekaysion" <daniel.kaesmayr@gmail.com >
wrote:
>Does anyone here have any ideas or suggestions what else might be
>useful to do or measure? Would anyone know of a machine setup that
>would be suitable? Would anyone know of contacts in the coffee industry
>interested in sponsoring or aiding such a project?
Refer to Illy & Vianni eds. 2005; "Espresso Coffee: the science of
quality;" Elsevier 2nd ed. for the techniques currently used in coffee
research. They may be approachable for grants as well.
The pharmacology and food science department at the University of
Navarra is doing quite a bit of research on espresso, trying to
directly correlate taste to its various preparation options. I've put
up some of their papers here: http://tinyurl.com/m28q7, and I'm sure
you can track down the rest.
|
|
