COE and Thickness

[neilestrick]

I've been making a new line of work this year, and I'e been gradually increasing the thickness of the glaze to get the look I like. The glaze fits very well at normal and double dip thicknesses- it passes the freeze thaw tests with flying colors. I find, however, that it will craze ever so slightly if the glaze gets really thick, which tends to happen on about 10% of my pots. Having not dealt with this issue before, my question to you all is this: can I tweak the COE to work even when it's really thick, or will the super thickness cause crazing regardless of the COE?

[Magnolia Mud Research]

 

Neil,

My thoughts:

My guess is that the 10% failures may be crazing due to being cooled too fast.  

Surface crazing is due to stresses in the surface layers that exceed the strength of the surface material.    Beside the COE differences between the glaze and the clay body, there are a differences is the thermal conductivity and of the specific heat of the glaze and the clay body.     Slowing down the cooling leads to smaller gradients in the surface temperatures and therefore lower stress gradients within the glaze layer. 

Try slowing the cooling cycle at the lower temperatures.  Start by comparing your cooling profile with the cooling profiles needed for thick walled glass work.

I have colleagues working with glass and if they cool (or heat) the glass too fast,  their work gets surface cracks that propagate through the work.  

Bullseye Glass in Portland, Oregon is the technical data source used by my colleagues working with fused glass. They have lots of info on glass cooling cycles. 

LT
 

 

 

[Min]

Neil, I can’t help but think that if the glaze is crazing where thickest then you are right on the edge of it “fitting” your clay.

I used the freeze / boiling water test for a glaze a number of years ago. Multiple test pieces, used test tiles plus real pots, no sign of crazing. Kept some pots with this glaze for our use, they crazed (vitrified clay). Since then I’ve gone back to my more extreme way of testing for crazing.  Test pots in 300F oven for a good 20 - 30 minutes, straight from oven into a sink of cool water. Oven up to 315F and repeat. Oven up to 320F and repeat. Likely more of a temperature difference than the freezer / boiling water test.

Might be worth trying and see what happens to the single and double dipped glaze. 

I'm glazing today and firing in the next couple days, I'll run a test with my bomb proof clear and apply it to double and triple the thickness and craze test it.  I haven't ever had crazing with this glaze so it will be an interesting test. 

[neilestrick]

@min I was thinking the same thing, that maybe it was really close but not quite on the COE. I haven't seen any crazing at all on 'normal' thickness pots after several months of use in the kitchen. Excited to see how your tests come out.

[neilestrick]

1 hour ago, Magnolia Mud Research said:

 

Neil,

My thoughts:

My guess is that the 10% failures may be crazing due to being cooled too fast.  

 

 

I was thinking that was a possibility, too, however the crazing usually doesn't show up for a few days after unloading, which is why I suspected the COE to be the issue. If it is a cooling issue, then I would expect it to craze during normal use, too, which I don't want. If that's the case I'll just have to really watch my glazing.

[Babs]

Ah gives me hope and sustenance on reading you guys actually still have problems:-))))

AND I love reading your methodical informed resolution process.

Thanks.

Babs

[glazenerd]

W.G. Lawrence "Material Science for the Potter" pg.147,148 & 156,157

" delayed crazing (days after) is due to the fact that the glaze did not reach equilibrium conditions during the original cooling process." 

Maximum contraction after cooling 2% of total contraction.  Potash based glazes more prone to delayed crazing due to its higher viscosity at melt. General glass annealing range: 500-600 C pending sodium or potassium based glaze.

micro fissures form during cooling, and after being removed from kiln: absorb atmospheric moisture over several days; expanding the clay/ glaze interface resulting in delayed crazing.

or you can just add a bit more silica, and a tad less spar.  Or remove a touch of spar and replace with gertsley. 

T

 

[neilestrick]

@glazenerd Thank you! But will a very thick glaze tend to craze during use due to heating/cooling stress in the oven/microwave/dishwasher, or with the correct COE will it not be a problem regardless of how thick it is?

[neilestrick]

1 hour ago, Babs said:

Ah gives me hope and sustenance on reading you guys actually still have problems:-))))

AND I love reading your methodical informed resolution process.

Thanks.

Babs

Ha!  Ceramics is such a broad subject, and we each work with such specific methods, that when we branch out into new areas there are lots of things we don't know. This is an issue I've never had to deal with before. I've worked with lots of thick glaze applications before, but those were layered glazes where the COE couldn't really be dealt with and crazing wasn't an aesthetic issue for me. But I've never dealt with a single transparent glaze applied this thick where I didn't want crazing. It's out of my experience level, and I though it would be good to share the info here. It's fun to have problems sometimes!

[Babs]

Not boring!

Why are you after a thick layer of transparency?

[glazenerd]

Neil: you are making me read.. Fortunately I have a book shelf full of dictionaries.

Dictionary of Glass by Charles Bray.

the COE does not change, regardless of thickness. What changes is:

1. The rate of cooling,  

 a. As the layer gets thicker, ( above 1/8") then a controlled cooling ramp must be observed during the annealing range of soda/ lime mixtures.  From 1020F (550C) down to 392F ( 200C)  Crystabolite inversion. 

B. Thermal differentials: the area exposed to air compared to the area making surface (shelf) contact.( controlled cooling required)

C. COE differential caused by incomplete melt. ( not all silica incorporated into the melt.)

Or Keith Cummings "Techniques for Kiln Formed Glass" do a 20 minute hold at 1020F (550C), then let it cool naturally. Allows temperature equilibrium to occur before further cooling.

T

[Min]

Hi Neal, test tiles are out and have been through my oven / cold tap water test and then calligraphy ink brushed on, left a minute then rinsed off. No crazing on any of them. Going from left to right, first is my normal thickness, it’s clear and shows the colour of the clay body beneath. Middle tile: glaze at approx 2.5mm thick. Right tile, glaze is 5 mm thick at the top of the glaze mound. Glaze is milky in colour in the 2 mounded thick glaze tests. (sorry about the glare)

Fired in a 7 cubic foot electric kiln, middle shelf. Schedule for the cool down part:

2185F 9999 2085 20 minute soak

2085F 150 1600 kiln off

Glaze: 

[neilestrick]

@Min You're the best! And there we have our answer:

If it fits, it fits!

Thank you Min!

[preeta]

On 8/14/2018 at 11:02 AM, neilestrick said:

my question to you all is this: can I tweak the COE to work even when it's really thick, or will the super thickness cause crazing regardless of the COE?

Uhhhh Neil. I see your excitement so I assume you got your answer. but I don’t understand.  

Does this mean you are going to play with the COE till the extra thick 10% fits? 

In other words was Min’s COE of her bomb clear glaze set so that no matter how thick it always fits. Does her numbers help you figure out her COE.  

Learnt even more here. 

[curt]

Magic Magnesium to the rescue...the anti-crazer of choice! 

...hmmmm, or was it the chunky boron component that fixed the crazing?....  (scratches head).   (Always love how Insight ignores Boron when calculating RO Unity...  - wish I could change that)

Or was it the combination of magnesium AND boron???  (starting to feel light-headed at this point).

Will take a long bath and think about this.

In any case, I like the short soak just off the top, Min.  Such things are much prescribed but generally little reported.  Thank you!

EDIT: I think the milkiness is due (aside from simple thickness) to the boron?

[neilestrick]

9 hours ago, preeta said:

Uhhhh Neil. I see your excitement so I assume you got your answer. but I don’t understand.  

Does this mean you are going to play with the COE till the extra thick 10% fits? 

In other words was Min’s COE of her bomb clear glaze set so that no matter how thick it always fits. Does her numbers help you figure out her COE.  

Learnt even more here. 

Sorry for the confusion. My question was never how to fix the COE of my glaze, but rather whether changing the COE would matter. I was wondering if a really thick glaze would have crazing issues regardless of how well it fits the clay. Min's glaze is known to fit her clay very well, and it didn't craze even when super thick. That tells us that if it fits, it shouldn't have crazing issues at any thickness.

[oldlady]

dear darling min, today is the day i am putting all the colors into the wonderful zinc free clear base glaze recipe you gave me in april.  the tests will be fired in a local studio and i am doing just colors and intensities of stain percentages, looking for pink, red and shades thereof.

is my glaze the bombproof one you are talking about here with neil?  i know the formula above contains the chemical names and you have not posted the recipe but i think it is the one, right???? (somebody's red originally)

[Magnolia Mud Research]

3 hours ago, neilestrick said:

I was wondering if a really thick glaze would have crazing issues regardless of how well it fits the clay. Min's glaze is known to fit her clay very well, and it didn't craze even when super thick. That tells us that if it fits, it shouldn't have crazing issues at any thickness.

Then why does thick homogenous glass (or for that matter cast iron) crack when heated too fast or cooled too fast? 

Me thinks that "shouldn't have crazing issues at any thickness" is an overly optimistic generalization.  

Again, cracks develop because of excessive local stress.   

"Why are these stress occurring where they are?"  is a better problem question. 

Until you have identified ALL the variables, you will not understand the reasons for the excessive stresses.  

Min's experiment show that thicknesses of the glaze drops are in her test pieces is not a significant variable; HOWEVER, is Min's test representative of the glaze and application situation of Neil's forms?  Is the glaze strength in the test glaze similar to the glaze in the cracked glaze (Neil's glaze); are the test tiles representative of Neil's forms, ...?  

My point is that while tweaking Neil's glaze for his application is likely to be successful, our collective understanding of glaze crazing is still in the trial and error stage for most situations because we, as potters, have not yet explicitly acknowledged all the relevant variables.  

Neil's crazing situation is a good candidate for the Root Cause Analysis technique Chris Campbell asked about back in March.  RCA starts with the failure; crazing where not expected based on prior experience.  Questions that might provide insight include, where does this failure occur?   Where does this failure NOT occur?  Does this failure occur on all clay bodies or only on one clay body?  Does the failure occur on all forms or just some forms?    Where in the kiln were the forms placed for each item?  What are the specific differences between the areas, forms, and clay bodies where the failure occurs and where the failure does NOT occur?  
What is the measured COE of the glaze; the estimated (from the glaze calculation software) COE of the glaze; what is the COE of the clay body; How thick was the body where the failure occurred and how thick was the body where the failure did NOT occur ... 

And oh yes, do not forget the question "why does ceramic objects and glaze crack instead of just thinning like a thread of polythene?"  

Answers should be specific, such the failure occurs only on shoulders; the failure NEVER occurs near a lip nor on the inside of the form; the object with failures were never on the top shelf; ....

Yes, there is a probability of being on "a ride down the rabbit hole;"   been there many times; however, after reflection the trip always lead to greater insights into both the specific problems and greater insights into the task of problem solving. 

LT
 

[neilestrick]

@Magnolia Mud Research Of course there are a million variables. I am in no way saying this is foolproof and works in every situation. But  in relation to what I was asking, this is a pretty darn good step forward, and saves me a lot of time doing tests that may have been a big waste of time, and for that I am excited. I apologize that it came off as a blanket statement of fact. I of course will do more testing to make sure it rings true for my particular situation, and I will let you all know how it turns out.

[Min]

5 hours ago, oldlady said:

is my glaze the bombproof one you are talking about here with neil?  i know the formula above contains the chemical names and you have not posted the recipe but i think it is the one, right????

Nope. The one I posted has a very low expansion and would more than likely shiver on your clay. There was no point posting the recipe because what fits on my clay won't necessarily fit on others.

 

5 hours ago, oldlady said:

the tests will be fired in a local studio and i am doing just colors and intensities of stain percentages, looking for pink, red and shades thereof.

Finally!  Looking forward to your results.