bIlL BuCkNeR on mon 6 jul 98

Linda: Just be sure you purchase a long thermocouple lead to go with it.
I have seen too many digital pyrometers destroyed by the heat of the kiln
in no time!

Also, here is my "el cheapo" solution to the digital pyrometer dilemma: I
was not concerned about a temperature readout, but rather a indication of
temperature change (up or down) and speed. I went to Radio Shack and
purchased a digital micro volt meter. It comes with alligator clip leads,
which I simply connected to the leads on my analog pyrometer (to use both
meters simultaneously). The analog pyrometer gives me a rough temperature
reading, and the volt meter gives me a quick reference to the change in
temperature (I calculated a temperature conversion factor, but I never use
it). In my firing log, I always just record temperature (from the analog
pyrometer) and the reading from the volt meter. This works well for me,
and for only $20!

-Bill Buckner


At 09:26 AM 7/4/98 EDT, you wrote:
>----------------------------Original message----------------------------
>I am considering the purchase of a portable digital pyrometer, the one
>in the Bailey catalog. Anyone have any recommendations? Where is the
>cheapest price?
>

e-mail: billtom@mindspring.com
web: http://billtom.home.mindspring.com

Why do they put Braille on drive-through bank machines?

Dave Finkelnburg on wed 31 jul 02

Roger,
Your post is an excellent short course on thermocouples and temperature
measurement in kilns. Good work!
Regarding the type K thermocouple, I agree, not everyone thinks that
thermocouple is good at stoneware temperatures. I investigated this
question at length with some industrial instrument technicians and have come
to the conclusion that while the type K thermocouple is not what I would use
for crystal glazes, it is accurate enough up to cone 10 for firing stoneware
glazes. While a type R thermocouple is accurate to a fraction of a degree
at cone 10, a type-K thermocouple is only accurate to within a few
degrees -- still close enough, I figure.
By the way, very nice pottery on your web site!
Regards,
Dave Finkelnburg, listening to a pleasant evening cricket chorus
in Idaho, USA

From: "Roger Graham"
Roger wrote, in part,
> Type K (chromel/alumel)
> Type N (Nisil/Nicrosil)
> Type R (Platinum/PlatinumRhodium)
>
> They all work OK up to stoneware temperatures, though not everybody agrees
> about using type K as hot as that. Type R is expensive, by reason of the
> platinum wire.

Janis Young on wed 31 jul 02

I've decided to spring for a pyrometer for my gas
kiln, accurate to ^10. My local supplier has them for
the equivalent of $240-$260 US$, which seems a lot.
Can anyone recommend something a little cheaper? I'm
in Canada but I don't mind ordering from the US if I
can save some money.

Janis

__________________________________________________
Do You Yahoo!?
Yahoo! Health - Feel better, live better
http://health.yahoo.com

Susan Morissette on wed 31 jul 02

Janis, I purchased an A.I.C. ox. probe here in the states, it is made in
Australia, about 6 years ago. I paid $600.00 u.s. then. three years later
the titanium wire broke and it cost me $300.00 to repair it. This year the
titanium wire broke again, but I will not have it repaired again, too costly.
It taught me a lot about my kiln and its atmosphere, so I don't need it
anymore. I understand you can find cheaper models now. If any one needs an
old A.I.C. requireing repair I'll sell for best offer. Good luck. Susan
from Woodstock, Ct.

Gavin Stairs on thu 1 aug 02

At 11:02 PM 31/07/2002 -0600, Dave Finkelnburg wrote:
>... I investigated this question at length with some industrial instrument
>technicians and have come to the conclusion that while the type K
>thermocouple is not what I would use for crystal glazes, it is accurate
>enough up to cone 10 for firing stoneware glazes. While a type R
>thermocouple is accurate to a fraction of a degree at cone 10, a type-K
>thermocouple is only accurate to within a few degrees -- still close
>enough, I figure.

Hi Dave, a couple of points...

First, types R and S are platinum/platinum alloy thermocouples. They are
rated for high temperatures because platinum is a very refractory metal
that has been used for high temperature labware for a very long
time. However, they both have drawbacks. Platinum is a precious metal,
and it is expensive. So the thermocouples are also expensive. They are
also susceptible to contamination by kiln gasses containing metal salts. A
contaminated probe will report a temperature error. Finally, the
difference in voltage for a degree of temperature (the sensitivity) of R &
S is significantly less than the K and N. So technique is very important
with these probes. The S is a more recent design than the R, and is
generally preferred in a new design or adoption.

The K and N are base metal probes, employing Chrome-Aluminum and Nickel
respectively. These metals are significantly less refractory than
Platinum, but also significantly less expensive. They are also susceptible
to degradation and thermal drift. The principal cause in this case is
alloy drift from high temperature diffusion, evaporation, etc in the case
of K, and all of the above in the case of the N. However, since they are
both relatively cheap, you can better afford to replace them. The K
thermocouple is only plotted (and standardized) to somewhere around the
middle of the stoneware cone range, so use of it at cone10 or so is
extrapolating it into the "not recommended" range. Extrapolation is known
to be less accurate than interpolation, which is what you do over most of
the range, in the midst of all the numbers. The N however, is standardized
to the cone 10 region, but not much above.

All thermocouples are available in several precision grades. The better
grades are selected from batches by measurement, and conform to their
published charts better (within a stated range at a set of given
temperatures). Since they are selected from the run of production, the
highly precise results may be missing from the lesser grade lots, resulting
in a skewed, double peak distribution for the lesser grades. What this
means is that you are more likely to have a probe with approximately the
rated divergence, plus or minus, from actual temperature than you are to
have one with the actual temperature. You have to take this into
consideration when deciding what strategy to employ.

So why do most potters use the K? First, because it is cheap and almost
universally available in standard industrial grades. Second, because it is
good enough. It lasts long enough to be useful. The best practice with a
K thermocouple (or any of the others, for that matter) is to calibrate the
probe initially, and then track its changes over the life of the
probe. The life of the probe in this case usually means for as long as the
probe continues to report a voltage (ie as long as it isn't actually
broken). During this lifetime, the temperature discrepancy will commonly
drift by a few degrees. The calibration is against cones or rings, or
perhaps against a newer probe. In the lab, it is against temperature
standards in a calibration apparatus. What most potters measure,
therefore, is not actual temperature, but the relationship between the
temperature in the kiln they are observing vs. a similar temperature in a
previous (successful) firing.

If you are using a thermocouple in an automatic controller, you need to
adjust the set points from time to time to account for this drift. So,
when you use thermocouples this way, you are not reliant on the absolute
accuracy of the probe, and do not need to spend more for the precise
probes. Also, you can use the cheap K instead of the expensive S. The K
is actually preferable because of its greater sensitivity (higher voltage
for a given temperature).

If you want to rely on the thermocouple to report an accurate temperature,
then you will need to use the higher grades, and replace them long before
they break, much before you think you ought to. The S thermocouples might
be more useful in this case, because of their higher maximum rating, which
means you will get a more accurate reading by interpolation, rather than
extrapolation. The same can be said for the N.

So the answer to which thermocouple type to use depends on they way you
intend to use it. Most potters use it as a relative temperature
comparator, and for this use the K is quite adequate, provided its
calibration is adjusted from time to time, to compensate for drift. Used
in this way, the K costs less than the S by a factor of 10 or more. If it
is being used as a pyrometer, or pyro-comparator, it can also be used with
a very cheap millivolt meter rather than a relatively more costly pyrometer
readout. There is no need to convert to degrees: just compare the
voltages this time vs the last successful firing. You will soon get used
to the ramps and end points in millivolts rather than degrees. The real
end point indicator is the cone pack, not the pyrometer. Use cones.

Gavin

Edouard Bastarache on thu 1 aug 02

" At 11:02 PM 31/07/2002 -0600, Dave Finkelnburg wrote:
... I investigated this question at length with some industrial instrumen=
t
technicians and have come to the conclusion that while the type K
thermocouple is not what I would use for crystal glazes, it is accurate
enough up to cone 10 for firing stoneware glazes. While a type R
thermocouple is accurate to a fraction of a degree at cone 10, a type-K
thermocouple is only accurate to within a few degrees -- still close
enough, I figure."


Hello Dave,

I have not found the "K" thermocouples I own to be good to be used
at stoneware temperatures. When new they seemed good up to c/4
but over time they became less accurate above 2000 F.
Any thermocouple will become inaccurate over time because, as for
kiln elements, each firing cause the deposition of a microscopic coat
of oxidized material on the wires the thermocouple is made up of, and
makes it less efficient.

"S" thermocouples are best suited for stoneware temperatures,
so says Mr. Andr=E9 Rajotte, engineer in physics and ceramicist,
who runs 9 large reduction kilns to produce iron and titanium
slags from ilmenite; these kilns vary from 90 to 200 tons in
capacity and he uses all sorts of temperature measuring devices.

During each firing I plot what the digital readout gives against what
the cones give, and with increasing temperature the drift becomes
more important. To me this drift represents the "time factor" of
"heat work". So, the reading on the digital readout plus the drift
calculated from the preceding firing tells me nearly exactly when
a particular cone will go down, better than astrology, no?
(Hehehe).

Hope this helps.


Later,



Edouard Bastarache
Irreductible Quebecois
Indomitable Quebeker
Sorel-Tracy
Quebec
edouardb@sorel-tracy.qc.ca
http://sorel-tracy.qc.ca/~edouardb/
http://perso.wanadoo.fr/smart2000/index.htm

Roger Graham on thu 1 aug 02

El cheapo suggestion from Australia, re pyrometer. This in reply to Janis,
who was asking. If you can find a supplier for just the pyrometer probe
alone (no readout) the price is not so alarming. If you already know all
this, please be patient. But if not, here's a crash course:

Pyrometer probes consist of two strands of wire, made of different metals,
enclosed in a refractory sheath, usually made of alumina. At the kiln end of
the probe, the two wires are melted together as one. At the outboard end,
the two wires emerge at some kind of terminal block, where you can connect
some kind of meter. When the "hot junction" of the probe is heated,
electrons travel more readily one way across the junction, than they do in
the opposite direction. The result is a small electrical voltage which can
be measured at the other end of the probe. The hotter the "hot junction",
the bigger the voltage.

How big a voltage? Not much. So small that it's usually expressed in
millivolts (thousandths of a volt). Typically 20 to 50 millivolts at cone
10, depending on the kind of probe. Here (Australia) electronic supply shops
are selling small digital multimeters, capable of measuring this kind of
voltage accurately, for prices like $15 or $20 Australian. I have three of
these attached to various kiln probes, cheap but so accurate. All you need
is a "lookup table" showing what temperature corresponds to what voltage.
Read the meter in millivolts. Look up the table. Takes only a few seconds.

Now, what kind of probe? They're identified by letter names, or by the names
of the metal alloys used in the wires. Three common kinds are:

Type K (chromel/alumel)
Type N (Nisil/Nicrosil)
Type R (Platinum/PlatinumRhodium)

They all work OK up to stoneware temperatures, though not everybody agrees
about using type K as hot as that. Type R is expensive, by reason of the
platinum wire.

If it helps, I can provide prints of the necessary lookup tables, as Word 97
files. Each table fits neatly on an A4 page. Different table for each type
of probe, of course. Email me off list if you'd like a copy, and say what
kind of probe.

Roger Graham

http://members.optusnet.com.au/~rogergraham

Brian Molanphy on sun 4 aug 02

edouard wrote:

'Any thermocouple will become inaccurate over time because, as for
kiln elements, each firing cause the deposition of a microscopic coat
of oxidized material on the wires the thermocouple is made up of, and
makes it less efficient.'

edouard, what if the thermocouple is protected by a ceramic (refractory)
sheath?

-brian

David Woodin on sun 4 aug 02

Type K thermocouples have tables that read to 2400 deg F. Omega makes a
low drift type K in a 1/4" inconel protecting tube. I get well over 100
firing from this type of thermocouple. The temperature is read at the tip
and if it is not close to the cone pack it may not seem to read correctly.
The rate of temperature rise has to be known for the themocouple to read
close to what the cone pack is saying. Orton has a disc out that shows
temperature equivalents for any ramp rate of temperature. The standard
cone tables are based on 27 deg F, 108 and 270 deg F temperature ramps this
is why pyrometers seem to read wrong. If the temperature ramp is know than
the pyrometer is very accurate. Industry would not use type K at the
temperaures potters fire at but would use type R or S. Most potters can
not afford the cost of type R or S and use type K.
David

Edouard Bastarache on mon 5 aug 02

Hello Brian,


this statement comes from Andr=E9 Rajotte, engineer in physics.
I will check with him again if it is still true for those protected by
any kind of sleeve.
I hate physics and he hates chemisrty, so I take what he says
for cash money.

Later,



----- Original Message -----
From: Brian Molanphy
To:
Sent: Sunday, August 04, 2002 10:32 PM
Subject: Re: Pyrometer advice


edouard wrote:

'Any thermocouple will become inaccurate over time because, as for
kiln elements, each firing cause the deposition of a microscopic coat
of oxidized material on the wires the thermocouple is made up of, and
makes it less efficient.'

edouard, what if the thermocouple is protected by a ceramic (refractory)
sheath?

-brian

_________________________________________________________________________=
___
__
Send postings to clayart@lsv.ceramics.org

You may look at the archives for the list or change your subscription
settings from http://www.ceramics.org/clayart/

Moderator of the list is Mel Jacobson who may be reached at
melpots@pclink.com.