On Mon, 21 Mar 2005 10:09:32 -0800, Brett Mason wrote:
Electronics and computers are big on my interest list. From my
observations of the hobbyists in those areas I think that those of you
experimenting with roaster designs could go far stealing from the
basics of those disciplines. My dad is an industrial electrician and
I spent some time dabbling in the 80's and 90's with homebrew computer
gear with PCs and the T.I. calculators.
I have seen comments to the effect of 'you should have seen the
controls on the commercial roaster I just saw'. I am not trying to
belittle anyone for being not being fully versed in everything
engineering has to offer, but the kinds of process control required
for small, single batch roasting is not rocket science. Sivetz might
be a heretic in some circles, but I have read his patents and seen
his products. He obviously knows something about coffee, and not just
the continuous, industrial kind.
Imagine an air roaster with a variably blower, variable heater,
airflow sensor, cold air TC, hot air TC, discharge air TC, and idle
bean TC. From a computing standpoint controlling this overgrown
monster could easily be handled by a microcontroler (20 year old core)
and some I/O hardware (relays, adc, motor controlled variac, etc).
The motor controlled variac might be the only 'expensive' part - if
you have the developer kit for the micro you choose. Control loops
would make sure that the air flow gets lowered as the bean density
falls, with voltage to the heater also figuring in. The oft mentioned
PID is an algorithm first, after all.
Food for thought hopefully,