Paul Vernier on wed 6 feb 08
I recently purchased an inexpensive data logger. It is a simple device that
records temperatures at various time intervals for a variety of thermocouple
types. I hook it up to a dedicated thermocouple on my kiln and set it to
record every 5 minutes. This frequency gives you enough data points to
nicely graph the firing profile. It is a great way to record your firing.
You can also by pressing a button record an event/marker on the data that
can correspond to when you make changes, close damper, open, increase gas
pressure, soak, crash cool, etc. I do this and enter the change in my log
with the time. I can then match it against the graph and seen when and where
I made a change and the effect. The unit with software cost a little over $150.
If interested you can find the U12-014 HOBO U12 Thermocouple Data Logger
and BHW-LITE HOBOware Lite Software Kit at http://www.microdaq.com/
Paul Vernier
James and Sherron Bowen on wed 6 feb 08
Check out the Omega Digital thermometers and data loggers. When I bought
mine a couple of years ago they gave me a digital multi-meter as a gift ($50
value) for being a new customer. Don't know if they still do that.
JB
http://www.omega.com/toc_asp/subsectionSC.asp?subsection=L04&book=Temperature
http://www.omega.com/Temperature/pdf/HH506A_HH506RA.pdf
----- Original Message -----
From: "Paul Vernier"
To:
Sent: Wednesday, February 06, 2008 12:40 PM
Subject: Easy way to record your firing profile
>I recently purchased an inexpensive data logger. It is a simple device that
> records temperatures at various time intervals for a variety of
> thermocouple
> types. I hook it up to a dedicated thermocouple on my kiln and set it to
> record every 5 minutes.
Bruce Girrell on thu 7 feb 08
I use the Fluke 54-2 pyrometer that can be set to record results from two
thermocouples at just about any time interval you want. I, too, found a five
minute interval to produce excellent results. With a five minute interval,
the Fluke can record about two days' worth of data from two thermocouples
before running out of memory.
Still, I would love to go the data logger route. In addition to
temperatures, I would want to record outdoor conditions (temp, wind speed,
humidity), gas pressure and flow rates, kiln and/or flue gas concentrations,
...
In there somewhere must be some correlations between the data and the
results (snot shino vs. racer shino).
We (the place where I work) have some nifty math tools to help establish
those kinds of tenuous correlations. They're called neural networks and
they're the kinds of things that stock traders use to establish correlations
between market conditions and stock prices, things that are way too
complicated for humans to grasp all at once.
I would think that neural nets, given proper inputs and training
information, could produce some impressive results. For example, given a
clay body, a glaze recipe, and firing conditions, they should be able to
predict glaze color, surface quality, and stability. The trick, of course,
is those innocuous sounding words "given proper inputs and training
information", which represent a lot of work. Still, a set of fairly well
controlled set of tests, such as Alisa Clausen's glaze tests, could go a
long way toward making a valid training set.
Sorry, I fantasize about this sort of thing from time to time and Paul's
note sent me off on a fantasy journey once again.
Bruce "incorrigible nerd" Girrell
| |
|
