Daniel Semler on tue 22 mar 05
Hi All,
This is the stoneware I use for raku and larger pieces.
One thing that was interesting about this body is that published values for
both shrinkage and expansion exist. So this seemed like a good way to determine
how close I was getting and how good my execution of the method was.
Again 3 sticks with 100mm lines on each.
Bone dry the aveg length : 93.42 mm
Bisqued the avg length : 92.92 mm
cone 10 oxidation avg. length : 86 mm
That's 14 % shrinkage at cone 10. Laguna's published value is 13.5%
Absorption :
Test piece Absorption
A 1.64%
B 1.27%
C 2.43%
The average absorption being 1.78%. Laguna's published value is 1.75%.
This information is interesting in that I actually did this test twice. The
reason I did this is because of the relatively large range of values from the
three pieces. I wanted to ensure that I had not done something foolish in the
testing. The 1st round results were slightly lower,
Test piece Absorption
A 1.31%
B 1.11%
C 2.08%
The average absorption being 1.5%.
But, note that the variation in results is still present.
So, two things. First, there does appear to be a genuine variation despite the
fact that the test pieces were all fired in the same firing on the same test
disk, a disk of clay about 7-8 inches across supported the test pieces and the
cones. Secondly, the absorption is higher in the second round values. I believe
this was related to my attempting to measure both the Glacia and the Soldate 60
pieces at about the same time and, as I did the Soldate 60 second, they had
longer to evaporate some water. It should be noted that the differences in the
percentages in the two runs are only 1-2 tenths of a gram of actual weight.
I'd be very interested in any comments anyone might have on the above. Jon
Pacini, if you see this, I'd be interested in your comments in view of the fact
that you use a pressure cooker. I did these in the saucepan. Have you ever
compared the two methods ? I think I might.
Thanx
D
----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.
Dave Finkelnburg on tue 22 mar 05
Daniel,
As someone who has used a lot of Soldate 60 I have to say you are doing excellent work! This is what the clay world needs more of...people doing logical tests and reporting the results...AND critically evaluating their own test methods. Nice job!
Second, you may find you will reduce variation by cooling the samples in water and then removing them one at a time, patting them surface dry and placing them on the scale immediately. That way all samples have the same amount of time between being immersed in water and being weighed wet. If your scale isn't accurate enough, consider larger test tiles, so the accuracy limit is a small % of the total sample weight.
As an aside, we should consider the shrinkage reported by anyone, including Laguna, as a number with a range. In reality, when the shrinkage is said to be 13.5%, that's really 13.5 plus or minus 1/4 % or 1/2% or whatever the range is. As reported, Laguna doesn't give us a feel for how much variation exists in the shrinkage of the Soldate 60 they sell. You are seeing the range of your moisture tests is fairly wide.
I, too, have wondered how the sauce pan boiling test compares with the pressure cooker version. If you make that test I look forward to reading the results.
Good potting!
Dave Finkelnburg
Daniel Semler wrote:
One thing that was interesting about this body is that published values for
both shrinkage and expansion exist. So this seemed like a good way to determine
how close I was getting and how good my execution of the method was.
Again 3 sticks with 100mm lines on each.
Bone dry the aveg length : 93.42 mm
Bisqued the avg length : 92.92 mm
cone 10 oxidation avg. length : 86 mm
That's 14 % shrinkage at cone 10. Laguna's published value is 13.5%
Absorption :
Test piece Absorption
A 1.64%
B 1.27%
C 2.43%
The average absorption being 1.78%. Laguna's published value is 1.75%.
This information is interesting in that I actually did this test twice. The
reason I did this is because of the relatively large range of values from the
three pieces. I wanted to ensure that I had not done something foolish in the
testing. The 1st round results were slightly lower,
Test piece Absorption
A 1.31%
B 1.11%
C 2.08%
The average absorption being 1.5%.
But, note that the variation in results is still present.
So, two things. First, there does appear to be a genuine variation despite the
fact that the test pieces were all fired in the same firing on the same test
disk, a disk of clay about 7-8 inches across supported the test pieces and the
cones. Secondly, the absorption is higher in the second round values. I believe
this was related to my attempting to measure both the Glacia and the Soldate 60
pieces at about the same time and, as I did the Soldate 60 second, they had
longer to evaporate some water. It should be noted that the differences in the
percentages in the two runs are only 1-2 tenths of a gram of actual weight.
---------------------------------
Do you Yahoo!?
Yahoo! Mail - now with 250MB free storage. Learn more.
Daniel Semler on wed 23 mar 05
Hi Michael,
Thanx. I'll give this a go. I had wedged a little but perhaps I should have
been more thorough.
Thanx
D
----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.
Jon Pacini on wed 23 mar 05
Greetings All------
Hope everyone had a good time at Nceca, I know I did. It is always nice to
see old friends and put faces with the phone calls and e-mails. For those of
you who didn’t get there, it was not only educational, but also very
entertaining.
Now on to the question at hand---------I got started on the pressure cooker
method working in the Lab at Westwood Ceramic Supply. Prior to that, I
recall I boiled samples in college. Penetration and time were always the
reasons given to me for using the pressure method. If you have very low
absorbency clay, putting it under pressure ensures penetration of the water
in a short time. As Ron has pointed out in the past, the boiling method has
always worked for him, and I’ll buy into that. The important thing is to be
consistent within your methods. Everything should be easily repeatable.
David has pointed out some possible points of variation. If the sample is
steaming hot, some water may evaporate, or if you dry one sample more
thoroughly than another it will affect the weight or even a simple thing as
the variation in the scale itself can play into this situation. However if
the samples were thoroughly dried prior to boiling, boiled for the same
period and dried and weighed in the same manner, I’d expect the results to
come out the same.
Clays that are heavily textured such as Soldate seem to be the toughest to
get a constant value for though. I think you may see surface water on the
sample as being a problem.
There is no doubt about variations existing from published numbers. The
biggest difficulty here is that my firing methods may vary from many of
those used in pottery studios. You know the old tale of one glaze turning
out a dozen different ways depending on who fired it. Methods used in
forming and firing will certainly affect test results. And in reality, the
published numbers are meant to be a guide so you can see the differences
from one clay to another. They are not intended to be definitive, note the +
or – 2% at the top of the characteristic chart.
I know it seems pretty silly to put such a number up there, but it’s there
to accentuate the fact that it’s a guide and not definitive.
So---I prepare the samples the same way on all the clays, fire them as
consistently as the equipment allows and test them as similarly as the
equipment allows and then publish those numbers. If you really need to know
how a clay reacts under your conditions, you have to fire it under your
conditions.
I do have a question for you Daniel-- you list 3 different samples you
fired. You have said that they were all fired together. Did they come from
the same block of clay, the same ton or the same run? Or were these samples
from several sources. I may have missed your explanation from an earlier
post.
Best regards
Jon Pacini
Clay Manager
Laguna Clay Co.
Daniel Semler on wed 23 mar 05
Hi Jon,
Thanx for this. All make sense.
The differences in firings and so on is fine and understood. I was mainly
concerned with variable results from the same samples, ie. some sort of method
error. The rough surface of Soldate 60 is an interesting point. I had thought
more about the shrinkage line in the tile but had not considered the body
surface.
Your point on consistency hits precisely at what I'm attempting to do. Develop
a method I can use and have faith in the repeatability of. Its all fine and good
to read about how its done, but it all changes a bit, and kinks reveal
themselves when one does it for oneself. Hence the questions and thanx for the
help.
What was bugging me in my test was the variance in the samples absorptions.
Doing it twice, showed that the variance still existed and that there was one
sample that was noticeably more absorbent that the others. Why ? Lots of
possibilities, firing (all fired together on the same chip), cracks (can't find
any), different surface texture leading to more surface water (doesn't seem so,
tiles are all similar - the second test I tried to make sure I'd patted the
surface well dry), different batches - your question (no, all from the same
bag). So its interesting and a mystery perhaps but its not so wildly out that I
can't get to sleep :) These things vary a bit. I expect as I get my method down
and test a few more things I'll be able to understand it better.
> They are not intended to be definitive, note the +
> or – 2% at the top of the characteristic chart.
>
> I know it seems pretty silly to put such a number up there, but it’s there
> to accentuate the fact that it’s a guide and not definitive.
Quick question about this. I assume that its +- 2% of the quoted value, not the
sample weight, ie. for the 1.75% quoted, the range could be expected to be
1.715 to 1.785 %
Thanx for the info. Hopefully I'll have this down soon.
D
----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.
Michael Wendt on wed 23 mar 05
Daniel,
Another factor that may affect finished porosity is initial moisture
differential. If the clay is from a different part of the pug, it may have
more or less water and when the water exits during firing, that may be a
factor in absorbency variability.
Consider stack and slam wedging the clay, then roll out the slab and cut the
test bars from a sample that has been laminated 30 times or more to assure a
more uniform moisture and grain structure. If you still detect variation in
shrinkage and absorbency, then some other explanation must be sought.
I always do my tests with uniformly stack and slam wire wedged clay and
there is no detectable variation in either shrinkage or absorbency in the
bars if they also have the same thermal treatment.
Regards,
Michael Wendt
Wendt Pottery
2729 Clearwater Ave
Lewiston, Idaho 83501
USA
wendtpot@lewiston.com
www.wendtpottery.com
Daniel wrote in part:
What was bugging me in my test was the variance in the samples
absorptions.
Doing it twice, showed that the variance still existed and that there was
one
sample that was noticeably more absorbent that the others. Why ? Lots of
possibilities, firing (all fired together on the same chip), cracks (can't
find
any), different surface texture leading to more surface water (doesn't seem
so,
tiles are all similar - the second test I tried to make sure I'd patted the
surface well dry), different batches - your question (no, all from the same
bag). So its interesting and a mystery perhaps but its not so wildly out
that I
can't get to sleep :) These things vary a bit. I expect as I get my method
down
and test a few more things I'll be able to understand it better.
> They are not intended to be definitive, note the +
> or - 2% at the top of the characteristic chart.
>
> I know it seems pretty silly to put such a number up there, but it's there
> to accentuate the fact that it's a guide and not definitive.
Quick question about this. I assume that its +- 2% of the quoted value, not
the
sample weight, ie. for the 1.75% quoted, the range could be expected to be
1.715 to 1.785 %
Thanx for the info. Hopefully I'll have this down soon.
D
| |
|
