Low Expansion Cone 10 Clear Glazes

[Bill Kielb]

3 minutes ago, neilestrick said:

@Bill Kielb I disagree. You can only add so much EPK and silica before the melt changes, and the crazing isn't always fixed when you reach that point. I've been there. And yes, adding in a different flux can change things, but the OP was just looking for a clear and didn't like the original glaze anyway, so there's nothing to lose. I agree that if he can fix it with flint and EPK then that would be a simple solution, but that may or may not work. I think other methods for solving the problem are valid.

I think we are saying similar things. Silica and alumina is the easiest and least likely to change the composition if it works. At its extreme it won’t melt so obviously you cannot get there with the particular  recipe.  My first suggestion was just that to the OP but if you read through everything the popular thought train was to modify one of the recipes and the popular first thought is to add or substitute Lithium, and even boron or any number of low expansion techniques. My comment was simply more complicated is not always better with  a hidden reference to watch durability at least as much as calculated coefficient of expansion.  

I really think we are saying the same thing and as always there are exceptions and many valid ways to solve based on experience.

[Min]

@LeeS, as you can see by this thread there are a couple schools of thought on this how to stop crazing dependent on how severe it is and by trying both methods I'm sure you'll gather some results. Seems we've gone full circle here, let us know how your tests turn out.

- Madeleine

[High Bridge Pottery]

Why do you think adding say 0.1 Li2O and 0.1-0.2 B2O3 decreases durability? I think a good mix chemistry is better for durability in most cases. Too much of any oxide is bad for this so it seems a good option. Sure you can make a nice glaze with only Na2O and CaO but why not add some more flux into the dance that can also lower COE.

I don't know how adding silica or clay is any easier or less complicated than using spodumene and a frit.

Why ignore the calculated COE values when they line up with tests? The 6.5 crazed less than the 6.7. Even if the numbers are for a theoretical final melt they still have value in relation to each other.  

[curt]

This is a very interesting and worthwhile discussion, since we have all probably spent many hours agonizing over how to get a better clear glaze recipe. (I know I have).

In my own testing, particularly with currie tiles, I have seen a few things:

1.The most consistent and predictable crazing is because the simple overall amount of flux in a glaze is simply too high compared to the amount of silica and alumina in that glaze.  This happens particularly in the bottom left, or "C" corner of a typical currie tile, like clockwork.  The implication is that a crazing glaze is quite likely simply an over-fluxed glaze.  So the time-tested remedy of adding more silica or clay - or both - makes sense.  If you don't know what a currie tile is search these forums...

However...

2.  Simply adding more clay and silica in the same ratio quickly becomes impractical, mainly because glazes with too much silica and clay compared to fluxes simply become too viscous and do not melt or heal over well, mostly because silica and alumina are very refractory and stiffen a glaze, something Neil was alluding to above.  The extreme version of this problem, ie when there is far too much clay and silica in the glaze (or possibly the clay - see below), is crawling.  This effect is also very evident, particularly on the upper portions of a typical currie tile.

One approach, if you have too much silica and clay and not enough flux), is simply to apply more heat, ie, fire higher.  And I think this is something that and and should be considered, in addition to changing chemistry, particularly if alkaline earth fluxes such as calcium and magnesium are available,   Their melting power is good, but it takes more heat to crack it.   Oh, and as long as you clay body will take it.

Anyway, if you had to add something to adjust the chemistry, my tests suggest that glazes are much more tolerant of more silica than of more clay.

3.  Once you are beyond the issues outlined in 1 and 2 above (ie, there is not too much, nor too little clay and silica in the glaze), the clay body becomes a very important factor.  Stoneware glazes all produce a glaze-clay body interface layer between the glaze and clay body when fired, where the glaze and the clay body "intermingle" at high temperatures, creating a new , third layer, which has chemistry of its own, distinct from that of either the clay body or the glaze themselves, but more like some intermediate version of the two.

My take is that when glazes are thinly applied (as we like to do with clear glazes), this interface layer has a heavy influence on glaze appearance and behaviour, including crazing.  

This interface effect appears most prominent in (relatively) heavily fluxed porcelain clay bodies, where the flux at the surface of the clay body appears to be augmenting the flux in the glaze as it intermingles, resulting in a somewhat over-fluxed final glaze chemistry.  The fine particle size of clay body materials in a porcelain body (and also of the materials in the glaze) is likely also playing a role in the glaze melt, as is the fact that we are usually trying to fire porcelain to porosity of near zero. 

In any case, the thought is that all else equal, if you have a very fluxy, fine-particled clay body that you intend to fire to near-glass state, you can probably get away with even a bit more clay or silica addition than with, say, a coarse, groggy sculpture body (not that you would be putting clear glaze on such a body anyway?). 

So, as Min suggested, looks like I am ending up in the same place as others here, but perhaps for somewhat different reasons.

[Bill Kielb]

2 hours ago, High Bridge Pottery said:

Why do you think adding say 0.1 Li2O and 0.1-0.2 B2O3 decreases durability? I think a good mix chemistry is better for durability in most cases. Too much of any oxide is bad for this so it seems a good option. Sure you can make a nice glaze with only Na2O and CaO but why not add some more flux into the dance that can also lower COE.

I don't know how adding silica or clay is any easier or less complicated than using spodumene and a frit.

Why ignore the calculated COE values when they line up with tests? The 6.5 crazed less than the 6.7. Even if the numbers are for a theoretical final melt they still have value in relation to each other.  

Adding silica and clay moves  the chemistry away from crazing in stull and is very easy and relatively safe because your overall flux is unlikely to change much. A simple progression takes minutes and can be done simply  on an already mixed glaze. Adding other fluxes requires reanalysis and almost always changes the chemistry of the glaze. Adding more than one feldspar and then maybe some boron Frit leads to lots of ingredients and often wastes natural materials. (I have seen glazes with ten or more components ...... and of course some bentonite)

                                                                                   All fine though as each method has its pluses and minuses.

Silica and Clay to me is simple, but durability for functional stuff is super important also, so designing with durability in mind makes me less apt to add products that change this ratio. If I can’t maintain at least 0.2:0.8  to 0.3:0.7 for a flux ratio adding more flux to the dance usually makes this worse and I rarely consider the glaze as a liner. It’s too easy to hit these numbers and then formulate any glaze I want, why add more? 

Again though, everyone has their method and most methods have had some success.

[LeeS]

Quote

I agree that if he can fix it with flint and EPK then that would be a simple solution, but that may or may not work. I think other methods for solving the problem are valid.

Yeah, this is the problem. My basic clay/silica progressions stopped melting before they fixed the crazing. I got close for the higher expansion of the two clays (Glacia) but the lower expansion (Pier) is still crazing really badly. I could (and probably will) try again with a glaze with a more durable flux ratio, but again, I don't think I'll be able to get there without a fundamental change to the chemistry.

The glazes that get close also fail badly in a 300 F:ice water test, so there is much room for improvement.

Quote

let us know how your tests turn out.

Will do! It'll probably be a few weeks but I'll drag this thread back up. I'm bisquing ~40 test pieces today! 

Quote

Why ignore the calculated COE values when they line up with tests? The 6.5 crazed less than the 6.7. Even if the numbers are for a theoretical final melt they still have value in relation to each other.

Yes, so far I have good alignment between the calculated COE and the actual crazing. This would be a different discussion if I was a production potter and fired glaze loads more frequently, because I could use a much more iterative approach over several firings. But for now I need to get as many useful trials into one kiln load as possible and using COE as a guide seems to be the best way to do that.

On that note...

Does anyone have recommendations for how to substitute fluxes? Is starting with 1:1 molar a good choice? Would the same advice apply to boron or do I need to treat it differently?

Obviously testing is the most important, so I'll try some progressions of whatever method I use, but any kind of starting point would be appreciated.

[Bill Kielb]

@LeeS

Just a thought, instead of fitting these recipes, the Katz recipe @min posted has a decent flux ratio and is pretty simple recipe. Might just fit or might be much  easier to fit than trying to make these work.

 

 

 

 

[Min]

@LeeS,I’m just going to pile on a bunch of info that I think is pertinent to your questions.

Simplified Periodic Table, I’m including this to show why I suggested lithia (using spodume) and why magnesium would also be a good flux to look at.

In order to have the least effect on changing the qualities of a glaze when it is necessary to do so on the flux level (rather than with silica and alumina alone) replacing all or part of one flux with another from the same group of elements gives a better chance of success. 

I’m going to use your G1947U glaze as an example of how I look at altering a recipe. I took out some of the custer to start with. The main reason I did this was to reduce the high expansion oxide potassium (bit of sodium in there too). Okay, fine, so I took some of that out but then in order for the glaze to melt I needed to replace it with something. Only other option in the same group of elements, alkali metals, is lithium. I like to to supply lithium by using spodumene. Obviously the spodumene contains more than just lithium, it contains silica and alumina also, but the beauty of glaze calc is it’s very simple to balance out these. Now, when I removed some of the custer in addition to removing some potassium and sodium it reduced the total calcium also but again, if necessary it’s easy to add some back in using calcium carbonate. 

Another way to attack this glaze would be to look at what other high expansion fluxes could be reduced. So, a list of materials from high to low expansion below (I’ve included barium and lead just for completeness but don’t recommend using those): 

HIGHEST Sodium oxide, Potassium oxide, Calcium oxide Strontium oxide, Barium oxide, Titanium oxide Lead oxide, Zinc oxide, Magnesium oxide, Zirconium oxide, Alumina, Silica, Boron LOWEST

Zinc is already in the glaze, one thing to note with it is if you use a glaze that has either zinc and/or lithia then the amount of boron needed will be less due to eutectics. Could add more zinc but keep an eye on the suggested limits (0.36) can have pin holes and other issues with it.

Magnesium is next, I think it’s a very good choice. You can supply it with a frit, dolomite, talc or magnesium carb (though that last material can bring in unwanted issues). So, if you go back to that mini periodic table above you see that magnesium is in group 2, the alkaline earths group. Many glazes contain calcium (also an alkaline earth) but look how much higher up the list of expansion materials it is. If I was going to use some magnesium to lower the COE of a glaze I would take out some of the calcium and rework the recipe to get some magnesium in there, probably using dolomite or talc or a frit. This way you aren't messing up the R2O:RO flux ratios too much.

Boron is next, it used to be thought of as a flux as it gets low and midrange glazes to melt so well but it's a glass former. It isn’t necessary at cone 10 but it can be used. 

One other thing that is sometimes used is zirconium. I know you are looking for a clear glaze but the addition of just 2 or 3% shouldn’t effect the clearness of the glaze and it does help reduce crazing.

I would suggest printing off a copy of the suggested limits for cone 10 (they’re not written in stone but somewhere to start for guidance) then spend some time using Insight and which ever recipe(s) you had the least crazing with and go from there. The CMW MAG Gloss from Matt Katz might work, but it might not be clear enough due to the high talc level. That would be one I would also try, as is but also with some boron and with some spodumene in place of some of the talc. I would run 2 recipes, one with some gerstley borate, one with some spodumene. (you said you didn’t mind a lot of tests ; )

One last thing, COE figures are not absolute, they work best when comparing glazes that use the same ingredients. By adding spodumene to the G1947U glaze I’ve changed the “system” of that glaze so it might be okay with a higher or lower COE than the same glaze without the spodumene. If possible try to change just add one variable (flux) in a glaze at a time. Once you get a glaze that fits then by comparing the COE's of other glazes it will get you in the ballpark of what will fit but it won't be absolute.

Hope I haven’t thoroughly confused you. Play around with Insight, come up with what you think might work and post it here so we can give some thoughts.

edit: re 1:1 substitution of one flux element for another, nope can't do it like this as some have much stronger or weaker fluxing power at certain temps than others.

[LeeS]

Wow Min, that cleared up a ton. Especially how to keep similar R2O:RO ratios, it didn't occur to me to swap fluxes within the same periodic group.

For my experiments, I started with the CMW mag gloss and the G1947U gloss. Here is the general flow:

(note this is with my Insight code numbers to avoid writing out the full name, and COE's are in []'s)

I tried to heed your advice to only swap one thing at a time as much as possible.
(recipes in next post because I'm out of space)

[LeeS]

Here are the new recipes based on the G1947U. Swapping out CaO for MgO (G0024) was surprisingly successful and didn't take a huge change to the original chemistry.

The CMW mag gloss was a bit more tricky because I didn't have any soda feldspar on hand, so I reformulated it with custer first, to make G0021. I also had to add some dolomite to fix the MgO/CaO ratio which changed with the feldspar substitution. Both new recipes ended up with a bit more CaO, which isn't ideal in terms of testing, but hopefully will result in a glossier glaze. I also shifted them to a slightly higher Si:Al ratio. G0026 attempted to have a similar flux/Al/Si ratio, while G0027 added boron and then significantly more Al/Si. All oxides are fairly well within limits except for MgO.

I'm not as thrilled with these because they ended up a bit further from the original chemistry, but they are reasonably inside the limits at least.

I also found a lithium glaze on ceramic arts which looks interesting:

Here are all of them plugging into Glazy:

I still haven't decided how to do the testing. I have ~20x test pieces each of two different clays, one with much lower expansion than the other. I'll probably try some progressions between the original glaze and each of my new attempts...

[Min]

Some thoughts, G0023 I'ld bump the zinc up by 0.50, room for more silica in the glaze. I'ld try it like this then take 100 grams of this base and add 2 grams of silica, dip and tile and repeat up to 6 grams of extra silica. Nice low LOI on this one, looks good.

G0024, looks good although the LOI is a perhaps a bit high. Again I would try this as is but also with 2,4,6 more silica to lower the COE a bit more, more gloss but should be okay.

G0025, If this melts okay it should be good.

G0021, hmmm, you've increased the COE from the original. Don't forget that by using a pot spar instead of a soda spar you've basically flipped the amounts of K2O and Na2O, this will have an effect on the melt, might not matter but it could be less melted now. Think I'ld skip this one.

G0026, try as is but also with 3% more silica. Given the MgO levels plus Si:Al ratio is in the range where this could be a semigloss so don't slow cool the kiln. (same with the CMW original glaze)

G0027, if it melts okay this should be good.

G0020, COE looks good, room for more silica and alumina to make the COE even lower if necessary. Just as an aside, every batch of spodumene I've used in the past 10 years has resulted in bubbles in the glaze from detergent residue from the processing of it. I try and remember to weigh out the spodumene as the first ingredient for a recipe then "wash" it. Add a fair bit of water and stir like crazy (water goes frothy) dump it out and repeat then just continue with the recipe. Spodumene is very slightly soluble so just don't leave it sitting around for days in the water.

I think you have a lot of good possible recipes here, I'm sure you already know this but I'ld fire all the tests on waster wafers just in case you get a runny one. If/when you get one that doesn't craze then I'ld check it for dunting to make sure you haven't gone too low on the COE. (thin walled clay cylinder, heavy glaze on the inside only, freeze it then pour boiling water into it)

 

[LeeS]

Quote

Some thoughts, G0023 I'ld bump the zinc up by 0.50, room for more silica in the glaze. I'ld try it like this then take 100 grams of this base and add 2 grams of silica, dip and tile and repeat up to 6 grams of extra silica. Nice low LOI on this one, looks good.

Bumped the zinc to 2.5. I'll try the silica progression for sure. Yeah, the LOI is quite low, no carbonates to decompose.

Quote

G0024, looks good although the LOI is a perhaps a bit high.

It's from the dolomite. If it works out I'll try to switch it for talc and adjust the wollastonite/silica to suit. I just don't have much talc on hand right now.

Quote

 G0021, hmmm, you've increased the COE from the original. Don't forget that by using a pot spar instead of a soda spar you've basically flipped the amounts of K2O and Na2O, this will have an effect on the melt, might not matter but it could be less melted now. Think I'ld skip this one.

Agreed, it's not a great recipe. This was just the closest I could get to the original with the materials I have right now, but understood it's a big-ish change... The plan was to run it essentially as a control against the modified versions (G0026/27), because they also use potash instead of soda spar, but I'll drop it if I run out of test space in the kiln.

Quote

don't slow cool the kiln.

Good tip. My tiny kiln cools super fast so this won't be a problem, but I'll make sure not to program a down-ramp.

 

Quote

Just as an aside, every batch of spodumene I've used in the past 10 years has resulted in bubbles in the glaze from detergent residue from the processing of it. I try and remember to weigh out the spodumene as the first ingredient for a recipe then "wash" it. Add a fair bit of water and stir like crazy (water goes frothy) dump it out and repeat then just continue with the recipe. Spodumene is very slightly soluble so just don't leave it sitting around for days in the water.

Great tip, I've heard about that happening but didn't know how to deal with it.

Quote

If/when you get one that doesn't craze then I'ld check it for dunting to make sure you haven't gone too low on the COE. (thin walled clay cylinder, heavy glaze on the inside only, freeze it then pour boiling water into it)

Ah interesting way to test. I'll try that for sure, because I generally glaze pretty thin and wouldn't notice if the glaze was under too much compression unless it was obviously shivering.

I can't thank you enough Min, this has been such an enormous help. Your guidance and explanations are invaluable.

[Min]

Somewhere to start, let us know how your tests turn out. Might take a bit more tweaking but I think you're heading in the right direction.

[High Bridge Pottery]

What's the problem with having high LOI? I never found it causes any problems, especially in cone10 glazes. Always found whiting the best CaO flux source. I hate trying to sieve wollastonite so don't use it, clogs up the mesh. and takes me 10x longer.

The Silica and Alumina look too high for me but interesting to see what results you get. 

[LeeS]

Quote

What's the problem with having high LOI? I never found it causes any problems, especially in cone10 glazes. Always found whiting the best CaO flux source. I hate trying to sieve wollastonite so don't use it, clogs up the mesh. and takes me 10x longer.

It's probably not that important and wollastonite is for sure annoying, but eventually I would like to fire very fast (the goal is ~5 hours from cold to ^10) so it'll probably matter more at some point.

Of course, the best way to fire faster would probably be to switch to a lower temp... but that's a problem for another time...

[LeeS]

All my tests came out of the kiln yesterday with somewhat surprising and mixed results. One shelf was an ideal ^10 and the other was a little over. I also ran a 300 F :: Ice Water test to check for crazing, and all the results below are after that. I'll itemize it to make things easier to follow:

 

[modification from base glaze in brackets]

G1947U Based Glazes

(1) G0023 [lithium] -- Crazed badly. Progression showed that larger portions of G0023 actually increased crazing against the base glaze. Silica progressions (3% and 6%) showed poor melting and increased crazing.

(2) G0024 [magnesium] -- Crazed slightly. Progression against base glaze showed it helped, but it is already starting to matte and stiffen the melt substantially.

(3) G0025 [boron] -- Didn't craze on glacia, crazed slightly on pier. This seems like the best of the G1947U based glazed, but it took a ton of boron to melt. I ran a progression from 0.1<->0.2 boron and only the 0.2 melted well and didn't craze. Less boron crazed more.

CMW Low Expansion Gloss Based Glazes

(4) G0026 [lithium] -- Crazed slightly on the glacia, none on the pier. This is a weird one because the crazing is opposite of everything else on the two clay bodies. The surface is also pretty awful, lots of little half bubbles but it's not properly matte.

(5) G0027 [boron] -- Didn't craze on any clay. I ran a progression from 0.1<->0.2 boron as well. The 0.1 boron has the best surface character and is well melted. The 0.2 is a bit bubbly and coarser.

Misc

(6) G0020 (Ceramic Arts Lithium B) -- Crazed worse than I've ever seen. I didn't even thermally stress this one because it was so bad already. May make a fun crackle glaze?

 

Some interesting results here:

• The lithium seemed to make the crazing worse in all cases. I'm not sure how to explain this given it's reputation for making low expansion glazes. G0026 on the pier is a bit of a weird exception, but it's an awful glaze visually and to the touch.
• Nothing is as glossy as G1947U, but I'm somewhat ok with that. Super high gloss isn't really the goal.
• G0023 definitely couldn't take more silica, at least not 3% more.

For future testing, it seems like adding boron is the only option that's working for me. The low boron (0.1) version of G0027 actually feels like the best bet right now, other than the not awesome surface texture.

I'm probably going to try a few things:

1. G1947U with both boron and magnesia additions, but not as much extra silica/alumina.
2. G1947U with both boron and lithium additions, probably not as much silica/alumina.
3. G1947U with both magnesia and lithium additions, probably not as much silica/alumina.
4. G0027 with small changes to improve the surface texture. Not quite sure what yet.
5. G0027 with small amounts of lithium (0.05).

I'm happy to post pictures if anyone is interested in any particular glazes. It's just a bit much to do all of them and still see the surface well.

I would love a bit more guidance on what to try next. I may use the 0.1 boron G0027 as-is, but it could use some improvements and I'm willing to do another test fire.

[Bill Kielb]

8 hours ago, LeeS said:

All my tests came out of the kiln yesterday with somewhat surprising and mixed results. One shelf was an ideal ^10 and the other was a little over. I also ran a 300 F :: Ice Water test to check for crazing, and all the results below are after that. I'll itemize it to make things easier to follow:

 

[modification from base glaze in brackets]

G1947U Based Glazes

(1) G0023 [lithium] -- Crazed badly. Progression showed that larger portions of G0023 actually increased crazing against the base glaze. Silica progressions (3% and 6%) showed poor melting and increased crazing.

(2) G0024 [magnesium] -- Crazed slightly. Progression against base glaze showed it helped, but it is already starting to matte and stiffen the melt substantially.

(3) G0025 [boron] -- Didn't craze on glacia, crazed slightly on pier. This seems like the best of the G1947U based glazed, but it took a ton of boron to melt. I ran a progression from 0.1<->0.2 boron and only the 0.2 melted well and didn't craze. Less boron crazed more.

CMW Low Expansion Gloss Based Glazes

(4) G0026 [lithium] -- Crazed slightly on the glacia, none on the pier. This is a weird one because the crazing is opposite of everything else on the two clay bodies. The surface is also pretty awful, lots of little half bubbles but it's not properly matte.

(5) G0027 [boron] -- Didn't craze on any clay. I ran a progression from 0.1<->0.2 boron as well. The 0.1 boron has the best surface character and is well melted. The 0.2 is a bit bubbly and coarser.

Misc

(6) G0020 (Ceramic Arts Lithium B) -- Crazed worse than I've ever seen. I didn't even thermally stress this one because it was so bad already. May make a fun crackle glaze?

 

Some interesting results here:

• The lithium seemed to make the crazing worse in all cases. I'm not sure how to explain this given it's reputation for making low expansion glazes. G0026 on the pier is a bit of a weird exception, but it's an awful glaze visually and to the touch.
• Nothing is as glossy as G1947U, but I'm somewhat ok with that. Super high gloss isn't really the goal.
• G0023 definitely couldn't take more silica, at least not 3% more.

For future testing, it seems like adding boron is the only option that's working for me. The low boron (0.1) version of G0027 actually feels like the best bet right now, other than the not awesome surface texture.

I'm probably going to try a few things:

1. G1947U with both boron and magnesia additions, but not as much extra silica/alumina.
2. G1947U with both boron and lithium additions, probably not as much silica/alumina.
3. G1947U with both magnesia and lithium additions, probably not as much silica/alumina.
4. G0027 with small changes to improve the surface texture. Not quite sure what yet.
5. G0027 with small amounts of lithium (0.05).

I'm happy to post pictures if anyone is interested in any particular glazes. It's just a bit much to do all of them and still see the surface well.

I would love a bit more guidance on what to try next. I may use the 0.1 boron G0027 as-is, but it could use some improvements and I'm willing to do another test fire.

So how did calculated COE work out in the end?

[High Bridge Pottery]

With G0023 you were just swapping out custer for spodumene? Strange results.

If the G0027 is the closest to what you want then I would go from there. Can you post a picture of the 0.1 and 0.2 boron ones?

Maybe G0026 too

[Min]

I agree with you and Joel, strange results with the lithium. It's like it's not actually spodumene. The Ceramics Arts Lithium B glaze has the highest lithium amount and yet it crazed the worst is very bizarre. Any chance you have some lithium carb? It's not great for using for glazes long term as it's soluble and will crystallize out but for tests you could use sub it in for the lithia in the spod. If a lithium carb glaze acts differently than the spod maybe there's an issue there.

The unmelted CMW ones might just have too much alumina in both and silica in the G0027 version. How fast are you firing the last 2-300 degrees?

edit: one of the most difficult glazes to make is a clear glaze that doesn't cloud, or craze, especially one for low expansion clay, it takes some work.

1 hour ago, Bill Kielb said:

So how did calculated COE work out in the end?

@Bill Kielb, as you realize it's a method used by many and it's very useful for looking at glazes within a system. It's another tool in the toolbox to use just as the Stull chart is.

[Bill Kielb]

@Min

no worries, these were all sort of designed around calculated COE or it was a significant part of the design objective so I was  just wondering how it correlated.  There might be a trend, just asking. I agree, this is often a difficult problem to solve and can consume a bunch of time testing. Hats off to the effort to  final solution

[LeeS]

First, pictures:

https://imgur.com/a/hGDpkE2

 

 

 

 

 

 

I tried to take them at an angle that would show the surface texture, and I embedded some ink in the crazing in some of them.

Quote

I agree with you and Joel, strange results with the lithium. It's like it's not actually spodumene. The Ceramics Arts Lithium B glaze has the highest lithium amount and yet it crazed the worst is very bizarre. Any chance you have some lithium carb? It's not great for using for glazes long term as it's soluble and will crystallize out but for tests you could use sub it in for the lithia in the spod. If a lithium carb glaze acts differently than the spod maybe there's an issue there.

I don't have lithium carb because my local supplier charges $55 for 1 lbs :/. I could try petalite maybe? Or another supplier or just suck up the money? I'll explore this a bit more.

Quote

 The unmelted CMW ones might just have too much alumina in both and silica in the G0027 version. How fast are you firing the last 2-300 degrees?

120 F/hr from 2150 F -> 2350 F, then 20 minute hold at 2350 F. ~500 F/hr before that.

Quote

So how did calculated COE work out in the end?

It provided some useful characterization within the same flux group for sure, as Min mentioned. There was some inaccuracy as the principle fluxes changed substantially (adding/removing zinc for instance). I would say I had useful results within +/- 0.2 COE.

Once I figure out if my spod is actually spod or if it has enough lithium I'll give you a better answer :).

[Min]

US Pigments price for lithium carb is $14- for 1/2 lb. What happens when you mix some water with your spodumene? Does it get frothy bubbles?

https://uspigment.com/product/lithium-carbonate/

https://uspigment.com/product/spodumene/

https://uspigment.com/product/petalite/

The only highfire clay I have right now is some ^10 B-mix. Was thinking I could run one or two of your recipes and fire them in my test kiln but apart from seeing how they melt it isn't going to help with the crazing problem. Let me know if you think this would help either rule in or rule out the spodumene maybe being an issue. 

 

[LeeS]

Quote

What happens when you mix some water with your spodumene? Does it get frothy bubbles?

I got a tiny bit of froth, but I didn't need to pre-wash it. Surprisingly little froth compared to what most people describe.

The texture is a bit weird too. It's very free-flowing (total opposite of wollastonite), and somewhat granular. Quite similar to table salt. I'm wondering if maybe I used too fine of a screen and it got caught? That doesn't make a ton of sense though because I probably would have noticed before I rinsed my screen off. Next time I'm at the community center studio I'll ask the tech there to see his spod.

Quote

The only highfire clay I have right now is some ^10 B-mix. Was thinking I could run one or two of your recipes and fire them in my test kiln but apart from seeing how they melt it isn't going to help with the crazing problem. Let me know if you think this would help either rule in or rule out the spodumene maybe being an issue. 

I really appreciate the offer Min, but don't worry about it. I'll keep working through this and let you know the results :). Of course you're always welcome to try for curiosity's sake, and I would love to know what you get!

[Min]

22 minutes ago, LeeS said:

It's very free-flowing (total opposite of wollastonite), and somewhat granular. Quite similar to table salt.

That does sound like spodumene.

[High Bridge Pottery]

G0026 is a weird one, doesn't even really look like the same glaze? Maybe the ink is throwing me off.

Did you dip them at the same time?  Not sure why the worse clay is a lot better.

G0027 seems like the one to go with. I would try reducing the silica and/or alumina and see if that reduces the orange peel texture.  The high SiAl is probably doing a lot for how it fits though so may start crazing again. If it does it would be interesting to try your spodumene again and see if adding a little makes it worse or better.