HomeRoast Digest


Topic: Youse guys/gals with screens???? (9 msgs / 202 lines)
1) From: Michael Horowitz
I'm trying to measure grains size.
What are the pro's/con's in using a series of screens vice something
like a pocket microscope (50x) with a .001" rule/reticule? - MIke
homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast

2) From: jim gundlach
I've put together a set of twelve screens that I am trying to find the 
time to set up and screen several samples but family, work, house 
repairs, and friends have kept me from getting to it.  I've played 
around with it enough to get the protocol down but it looks like it is 
going to take a little more than an hour per sample and I think I have 
about ten samples here that I need to get to.  I went with screens 
because it is the method used by the coffee industry.  I never thought 
about using a microscope and now that I think about it, the microscope 
process could be quicker but it would be difficult to compare the 
results since the screens give a distribution measured by weight and 
the microscope would count the grains by size.  The work I have done on 
this so far suggests that the amount of variation in grain size is the 
most important measure for judging grinders, you would probably need to 
measures a few hundred gains to get a reliable measure but I could see 
the microscope approach taking much less time than the screen method.
I was able to build the screen set for less than $35.00 and  I don't 
know how much the microscope set up would run.
Jim Gundlach
On Sunday, February 2, 2003, at 05:58 AM, Michael Horowitz wrote:
<Snip>
homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast

3) From: Tom Gramila
On Sun, 2 Feb 2003, jim gundlach wrote:
<Snip>
Hey Jim,
	Did anyone with one of the Cunill grinders send you a sample? It 
would be interesting to see how it stacks up ( or maybe shakes out is 
better...) compared to the others.  
	I am in the market for replacing my Solis meastro, count me as 
among those patiently awaiting your results!
		Tom G.
homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast

4) From: jim gundlach
On Sunday, February 2, 2003, at 12:17 PM, Tom Gramila wrote:
<Snip>
Nope, when I saw the discussion of it, I thought of asking for a sample 
but I decided I did not need any more guilt about undone tasks for a 
while.
     Jim Gundlach
<Snip>
homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast

5) From: Rick Farris
The ability to plot the distribution.
-- Rick

6) From: jim gundlach
On Sunday, February 2, 2003, at 02:42 PM, Rick Farris wrote:
<Snip>
I could plot the distribution of both kinds of data.  The issue would 
seem to be what you want to measure.  If you have a sample with a lot 
of dust, anything less than 10 thousandths by the current operational 
definition, you would have to count a lot of quite small particles 
before you got enough in the 15 to 25 thousandths were you would get 
most of your espresso grains, or even more to get a reasonable 
distribution of 35 to 50 thousandths for a drip to French press.  To 
try to estimate the relationship between portion by weight and portion 
by number of grains, let's assume that the grains are cubes.  If you 
had 60% of the weight in 30 to 40 thousandths grains that averaged 35 
thousandths and 5% dust that averaged 5 thousandths, you would end up 
counting about 40 times as many dust grains as 30 to 40 thousandths 
grains.
      If I haven't lost you yet, thinking about it makes me think the 
microscope approach would actually be a lot more work than the screens.
Jim Gundlach
homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast

7) From: Rick Farris
Ok, let me rephrase my original reply:
The practical ability to plot the distribution.
-- Rick

8) From: jim gundlach
On Monday, February 3, 2003, at 12:37 AM, Rick Farris wrote:
<Snip>
Data entry would be substantially different.  Measuring with the 
microscope would require entry of each measurement which would be 
probably be several hundred for each sample.  And studies show that 
people with an IQ above 85 burn out after entering data for a 
relatively short time.  With the screens, there is one entry for the 
weight of the coffee caught by each of the twelve screen.  The commands 
for plotting in JMP are a little different but the program handles both 
kinds of data quite easily.
Jim Gundlach
homeroast mailing listhttp://lists.sweetmarias.com/mailman/listinfo/homeroast

9) From: Rick Farris
Yeah, it was the measurement end I was concerned with.  Counting all the
little particles.  I wouldn't even want to do that if I had a set of screens
to help me segregate the particles for counting.
-- Rick


HomeRoast Digest