[Homeroast] Homeroast Digest, Vol 25, Issue 23

Michael Dhabolt michael.dhabolt at gmail.com
Wed Feb 24 15:54:22 CST 2010

 J.W.Bullfrog wrote:
> You just have to get the probe to a 'representative' location. Getting
> a 'representative ' local is a bit like bear hunting. Sometimes you get the
> bear; sometines the bear gets you.

Absolutely Correct!

I ended up building quite a number of (Ubber Popper) clones.  I had an
interest in establishing a consistent reference temperature from one
machine to the next.  This would hopefully allow PID profile (program)
transportability between machines.

The P1 with a tubular glass chimney (the configuration I use) has an
effective roast chamber measuring 3" (three inches) in diameter with a
flat bottom and a initial bean depth (greens) at the start of roast in
excess of 3",  I decided (somewhat arbitrarily) that a Thermocouple
(TC) junction located 1.5 inches from the sides and 1.5 inches from
the bottom would provide me with a fairly uniform bean temperature
measurement location.  I continue to be pretty happy with the
decision. I have been consistent with the TC junction (Omega probe),
TC cabling (Omega wire, flex cables and connectors) and
Indication/Controller (Fuji PXR3) on all but personal and test
machines.  I've found less than 1° F variability between machines.

A number of machines produced as individual projects by some other
folks using somewhat less exacting tools resulted in a bit less
consistent TC probe placement.  This resulted in a bit less
temperature consistency between machines (up to 5° F).  The radial
placement seems to be less important than the vertical placement with
respect to consistency.  Even at this data displacement, poking a
temperature offset of an appropriate amount into the PID brings the
system back to an acceptable ability to utilize profiles (programs)
developed on different machines.

In other words a long pedantic way of saying "Location, Location, ....

Mike (just plain)

More information about the Homeroast mailing list