HomeRoast Digest


Topic: SCAA and roaster controller group project (4 msgs / 212 lines)
1) From: Jeffrey Pawlan
I am very surprised that some of you seem so negative about the SCAA. Yes they
have problems and yes I too have made some waves there and a few enemies.
But the basic concept is a good one and I think you should support it. I think
Folgers is a member of NCA, not SCAA. But Starbucks is a big contributor. We may
not like their coffee but they have put 'specialty coffee' on the map and I
think made coffee better by inspiring competition.
An important clarification about the research session:
The research is not sponsored or being paid for by the SCAA. Not even close.
This was entirely my brainstorm and I put it together. I met Marty in Charlotte
and we seem to work together well. So we have teamed up to do this. It was going
to be here in the Bay area but that did not work out. I am now a professional
member of the SCAA because I sell roasting equipment. I got permission to use
their lab for the research because I am a member. It is a nice facility with
real PID controlled coffeehouse size brewers and lots of grinders. They
also have plenty of tables and chairs and space which I do not have. Trying to
do this in my house would not be possible. I had put on a tasting event at my
house previously and assisted other Bay area gatherings.
I am also a member of the Roasters Guild. My focus in this research project is
to further the body of knowledge about roasting. Personally I do hope that the
art of roasting will adopt computer control. Some roasters, don't really want
that to happen.
Scott,
There WAS a public homeroaster group development project but it died from lack
of interest. The core participants were the founder: Jeffrey Bertoia plus Ben
Treichel, Bob Yellin, and myself.  The webpages are now gone but I will include
a written description of what they had done. They even had some PCBs made but I
did not care for the design so I did not buy one and instead I went a different
route. If you want to start such a group, you are very welcome to but you will
likely find it frustrating and a big timesink.  Your idea of a ubiquitous
controller is not unlike mine; mine can do all that.  But you do not have the
engineering background to understand the thermal dynamics of different roasters
and thus you have an idealistic view. I have often been asked to put my
controller on an existing 50 year old roaster. The contoller will work just
fine but you can't expect a giant Mack truck to have the maneuverability of a
sports car so you won't get a profile no matter how good the controller is, and
software tries.
Here is the post that I had saved from 2003. You will see that I had started
with a Hearthware Precision air roaster.
Date: Sun, 07 Dec 2003 18:54:07 -0500
From: Jeffrey A. Bertoia 
To: Jeffrey Pawlan , roasterdesign
Subject: Re: roaster controller
Jeffrey
I am crossposting this to the roaster design mailing list.
Jeffrey Pawlan wrote:
<Snip>
Ben Treichel  is working on the theory of operation for the entire
board.  I am working on the theory
of operation for the phase control circuits.
<Snip>
Each PIC is the heart of a phase control circuit one for the fan and one
for the heater.  I do not know
what your backgroud is but by phase controlling the AC Line we can
effectively control voltage.
<Snip>
I haven't bothered to generate a parts list yet and an interface
document should be available in Bens
package, when complete.
You can get more information now by subscribing to the mailing list and
going through the
archives.
<Snip>
....  OK quick summary.  The design is actually a one wire IO board
containing all IO required
to control any roaster.
The process controller can be anything that is capable of 'speaking' one
wire.  Presently we
are expecting to use a Dallas Semiconductor TINI board.  This will allow
us to program in JAVA
so we can develop on PCs or similar.  The Human Interface can be through
a webserver or
one of several one wire displays.
<Snip>
The HWP should be no problem for this controller.
<Snip>
Yes it is essentially a 2 speed control.
<Snip>
I did not see this as a problem on my HWP.  My roasts always seemed very
even.  But the system is less
than sophisticated.  THe fan switches speed based on a single set point
and only has 2 speeds.
<Snip>
This controller will have the capability to change both fan speed and
heater voltage.  THe intention
is to use a  'driver', if you will for each roaster, that will tell the
controller how to deal with it.
Fan speed and heater voltage will be continuosly variable from
essentially 0 - 100% .
Whether we allow the algorithms to actually make those changes is
another question.
However, one must keep in mind that with a fluid bed roaster one wastes
a lot of energy
because of the large airflow.  In fact with small loads the roast can
stall because there
is so much air flow that it never achieves temperature.  I think we will
need to do
some fan control.

2) From: Brett Mason
Jeff,
I hope you gain audience for computer control development.  Your innovation
is tremendous, and very highly regarded here.  My instinct is your
experience will be missed by higher paying people...
I may be all wet, dead wrong.
I hope I am - I can apoloize.
In the meantime, all those folks still brew coffee that sucks.
And I get fresh roast tomorrow mornin - they have never tried such...
Brett
On 9/25/06, Jeffrey Pawlan  wrote:
<Snip>
-- 
Cheers,
Bretthttp://homeroast.freeservers.com

3) From: Scott Marquardt
Thanks for this info. Very interesting indeed. FWIW, though I have no formal
engineering background I'm aware of some of the problems. As an inactive
pilot of small aircraft, landing wide body jets in sims is the same kind of
thing.  ;-)
However, in many areas where I've successfully contributed worthwhile
innovation, it's generally been my outsider perspective that encouraged
optimistic exploration among colaborators bogged down by insider despair. In
areas where I've been the one in despair, it's often been the most naive
questions of the uninitiated which, patiently tolerated through some
obviousnesses, spark an idea.
But yeah, the time sink.    ::sigh::
- Scott
On 9/25/06, Jeffrey Pawlan  wrote:
<Snip>

4) From: Dan Bollinger
This is a multi-part message in MIME format.
Scott,  You don't have to be an engineer to have a vision, but it does =
help in the follow.  :)    
Jeffrey,  Good luck with your tests. I'd be there if I could.  Dan
  Thanks for this info. Very interesting indeed. FWIW, though I have no =
formal engineering background I'm aware of some of the problems. As an =
inactive pilot of small aircraft, landing wide body jets in sims is the =
same kind of thing.  ;-) 
  However, in many areas where I've successfully contributed worthwhile =
innovation, it's generally been my outsider perspective that encouraged =
optimistic exploration among colaborators bogged down by insider =
despair. In areas where I've been the one in despair, it's often been =
the most naive questions of the uninitiated which, patiently tolerated =
through some obviousnesses, spark an idea. 
  But yeah, the time sink.    ::sigh::
  - Scott
   
  On 9/25/06, Jeffrey Pawlan  wrote: 
    Scott,
    There WAS a public homeroaster group development project but it died =
from lack
    of interest. The core participants were the founder: Jeffrey Bertoia =
plus Ben 
    Treichel, Bob Yellin, and myself.  The webpages are now gone but I =
will include
    a written description of what they had done. They even had some PCBs =
made but I
    did not care for the design so I did not buy one and instead I went =
a different 
    route. If you want to start such a group, you are very welcome to =
but you will
    likely find it frustrating and a big timesink.  Your idea of a =
ubiquitous
    controller is not unlike mine; mine can do all that.  But you do not =
have the 
    engineering background to understand the thermal dynamics of =
different roasters
    and thus you have an idealistic view. I have often been asked to put =
my
    controller on an existing 50 year old roaster. The contoller will =
work just 
    fine but you can't expect a giant Mack truck to have the =
maneuverability of a
    sports car so you won't get a profile no matter how good the =
controller is, and
    software tries.


HomeRoast Digest