Clay Body Development and Cracking/Dunting

[LeeS]

Hi all,

After my rather unsuccessful experiments with a few commercial slipcasting clay bodies, I've embarked on developing my own. I started with the basic 25-porcelain and I've been working from there. I'm curious both on some overall feedback on my recipes, as well as a specific problem I ran into. This is all at ^10.

So far, I've tested 4 different feldspars; Minspar, Custer, Nepheline, and Mahavir. I found Custer really doesn't flux as much as I would like. Mahavir seems to be creating odd casting problems. Nepheline and Minspar have been great. Nepheline does have a bit of solubility that is staining my molds, but it doesn't seems to be a huge problem.

For clay, I'm using primarily EPK and Grolleg. Mostly to diversify a bit, but also because EPK seems to cast a fair bit better than pure Grolleg while still being low enough iron. There is also a tiny bit of VeeGum to tune plasticity, but only around 0.5%.

Here are the recipes I've narrowed it down to:

B0013 (base - 25% feldspar)
Silica    25
EPK    20
Grolleg Kaolin    30
Minspar 200    10
Nepheline Syenite    15

B0009 (base + custer - 25% feldspar)
Silica    25
EPK    25
Grolleg Kaolin    25
Minspar 200    5
Nepheline Syenite    10
Custer Feldspar    10

B0015 (20% feldspar) 
Silica    25
EPK    25
Grolleg Kaolin    30
Minspar 200    10
Nepheline Syenite    10

B0019 (15% feldspar)
Silica    25
EPK    30
Grolleg Kaolin    30
Minspar 200    7
Nepheline Syenite    8

I haven't tested B0019, but all other seem fully vitrified, I wasn't able to detect any absorption. B0019 was an attempt to make something with _some_ porosity because I was surprised even the 20% feldspar B0015 fully vitrified. I am also firing a set of test bars to ^8 to get a better sense of the maturities. It seems like B0015 will be just about right assuming it doesn't warp too badly.

I would love some overall feedback if anyone has comments.

Lastly, one more specific question. My last test firing was with a few different glazes on all the bodies. The plan was to take them out of the kiln at ~400 F and plunge them into ice water to check for crazing. When I did that, almost all the bottoms of the test cups cracked, some the whole bottom popped off. You can see how the shape of the test cups here: https://imgur.com/a/hGDpkE2.

I rushed things a bit and I _think_ they were close to 400 F, but the kiln was cooling super fast and there is a chance that the base touching the shelf was significantly hotter than the rim. My question... Is this just a screw-up on my part, or could it be a glaze fit issue where the glaze is too low expansion for the body? I didn't fire anything higher expansion then the DigitalFire G1947U (https://digitalfire.com/recipe/g1947u). Next firing I'll include a higher expansion glaze, but I was hoping there is something to be learned here regardless. I can post pictures of the cracks if that would help.

[liambesaw]

Pretty sure most thin porcelain will burst when plunged into cold water at 400 degrees.  That's right around cristabolite inversion and a temperature gradient like that would likely doom most vitrified porcelain.

Just do the freezer to boiling test, it's sufficient.

[glazenerd]

LeeS

Like all things clay: kaolin has a range of particle sizes. An example would be- 0.50 to 10 microns in particle size. The theorem in slip casting is called "card stacking": where particles drop out of suspension and stack like cards. The closer you get to this principle: the denser the piece will be. Cracking off the bottom is an indication that larger particles settled out; with finer particles settling last creating a weak point. You made no mention about deflocculants? While the general theme about deflocculating centers around specific gravity; more to it than that- these additions also suspend particles uniformly .Sodium silicate with a small addition of soda ash is common: Darvan 7 is much easier to use: and provides long term suspension. Your recipes that call for Nep Sy did better because the soluble salts acted as a suspender. The other parameter is temperature: at 68F; the negative particle charge that creates suspension drops by 1/3: which directly effects particle suspension. As the temperature drops below 68F, so does the particle charge. I pull test pieces out at 300F and toss them in cold water: if they do not survive that- testing is a waste of time. I put a kiln shelf over those pieces below it; so that the rush of cool air is minimized.

Tom

 

 

[glazenerd]

LeeS

Sorry for the second post: had to switch over to my IPad to find a picture. I mixed a simple OM4 Terra Sig recipe and added 7 drops of Darvan 7 ( Credit Marcia Selsor for technique) Then placed it in the freezer for 15 minutes to show the effects of temperature on particle size. OM4 ihas a particle distribution below 1 micron. So the effects of settling would increase as particle sizes increased. In addition, inadequate deflocculation would add to this problem even in warmer weather.

 

[Sorcery]

Pretty Sure Tom Nailed it. Love this place!

I was thinking firing schedule, since all my problems seem to disappear with a longer firing.

But also plucking. If your porcelain is sticking and not allowing the bottom to shrink well, it could be weakening your bases.

Sorce

[Min]

10 hours ago, LeeS said:

I rushed things a bit and I _think_ they were close to 400 F, but the kiln was cooling super fast and there is a chance that the base touching the shelf was significantly hotter than the rim.

Think this is too extreme a test. Pyrometer will be measuring air temperature, kiln shelf will be hotter. I test for dunting like Liam does with freezing and boiling water.

If you need fast casting speed have you tried including a small amount of Tennessee #1 SGP ball clay? I'm currently using a ^6 grolleg slip with just under 5% of it, I can get 4 casts a day from a mold. Makes a huge difference in how fast it releases. Does change the colour from white to a light cream though.

[LeeS]

Quote

You made no mention about deflocculants?

The slip is deflocculated with Darvan 7, about 1.7 specific gravity (I don't push it too hard for tests), and the mixture is slightly thixotropic.

Your comments about settling are pretty interesting, however, I'm not sure that's the issue here (of course not discounting it either). Reading my original post, I think I misstated the problem a bit. In almost all the tests, the cracks start at the base and work their way up the sides. Only one or two have missing bottoms, but because they look so dramatic I think I fixated on them. A few are also cracked almost perfectly in half. Thanks for all the info though, I will certainly keep  this in mind. Your freezer test was very enlightening.

Based on what everyone is saying it seems like I just stressed the pieces way too hard. The shelves in the kiln could have easily been at 500 F based on how fast the temperature was dropping. I had a bit of a false sense of security because my undervitrified commercial body could handle pretty extreme shocks... with the maturity and glaze fit issues that entails.

Quote

But also plucking. If your porcelain is sticking and not allowing the bottom to shrink well, it could be weakening your bases.

Curiously there is no plucking, even in the over vitrified samples.

Quote

If you need fast casting speed have you tried including a small amount of Tennessee #1 SGP ball clay? I'm currently using a ^6 grolleg slip with just under 5% of it, I can get 4 casts a day from a mold. Makes a huge difference in how fast it releases. Does change the colour from white to a light cream though.

Very interesting! I'm going for maximum translucency (one of my projects is a light fixture) in this round. My casting times are still really good, around 5 turns a day, so I'll probably hold off on trying it, but fantastic suggestion.

 

It seems like in my rush to get everything out I ruined my test :P.

I have the first results from the ^8 test:

B0013 (25% feldspar) - 0% absorption, 9.5% shrinkage
B0015 (20% feldspar) - 0% absorption, 10.5% shrinkage
B0019 (15% feldspar) -  1.6% absorption, forgot to measure shrinkage

They do need to boil + soak a bit longer to confirm these numbers. However, curiously, even the 20% feldspar seems to be fully vitrified at ^8. Maybe the is a sign I should fire lower...  B0019 wasn't  part of the original tests, but I will test fire it to ^10 in the next load.

Thanks again for the advice everyone!

[LeeS]

I was able to test B0019 and a series of glazes. The results are somewhat encouraging, although I definitely have another firing or two before I'll be able to settle on a formulation.

At ^10, test results and recipe:

B0019 (15% feldspar) -- 0% absorption (1.6% absorption at ^8), 10% shrinkage
Silica    25
EPK    30
Grolleg Kaolin    30
Minspar 200    7
Nepheline Syenite    8
VeeGum T   + 0.75

It's very interesting that it's fully vitrified even with only 15% feldspar. I assume it has something to do with the significant amount of flux in the Grolleg, and the relative strength of the Nepheline and Minspar (at least compared to Custer). The VeeGum is just so it will pull away from the mold, and I still need to tune the amount a bit. This body seems to be the way to go. I still need to check how badly it warps, and do some more careful absorption tests, but the maturity at ^8 vs ^10 seems to indicate the flux ratio is suitable.

I didn't have a chance to try any glazes on B0019, but I did try them on the three other body recipes. All were cycled both 400F to ice water, and boiling water to ice water. Here they are [calculated COE in square brackets]:

Tony Hansen's G1947U [6.5] -- No crazing
My take on G2571A, coded G0062 [6.7] -- Crazed slightly on all tests
Matt Katz's CMW Mag Gloss [6.2] -- Weird line at the rim, maybe on the edge of shivering?
A fairly standard gloss I've coded G0006 [6.7] - No crazing

Here are those two recipes:

G0062
Wollastonite    18.5
Custer Feldspar    27.5
EPK    30
Talc    15
Ferro Frit 3124    9

This turned out to be a good start, but it melts a bit much, and the boron seems to cause bubbles that aren't compatible with the fast firing schedule. It also crazes, but with plenty of room for more EPK and less feldspar, that shouldn't be too much of a problem. I also tested a few variations of this on my old clay (Glacia), and they're getting really close.

G0006
Custer Feldspar    29.8
Whiting    15.7
Silica    29.8
Dolomite    5.5
EPK    19.2

This was picked because it's higher expansion than G1947U, but it has slightly worse surface texture and clarity so I probably won't be pursuing it.

Overall, I'm pretty happy with how this is going. It seems like my cracking issues last time were entirely my rushed test. If anyone has any general feedback or advice I would love to hear it. I do have one more specific question. The CMW Mag Gloss tests developed a strange crack running around the rim after testing, and I'm curious if anyone has seen this before:

I tried rubbing some ink in but nothing happened. Is there a chance the crack is in the clay and not the glaze? Given my experience comparing glazes on the Glacia, I'm fairly confident it is lower expansion than the others.

As always, thanks for reading and any help!

[Min]

14 hours ago, LeeS said:

All were cycled both 400F to ice water, and boiling water to ice water. Here they are [calculated COE in square brackets]:

400F to ice water is still pretty extreme. What I've been doing for years is use the kitchen oven and start at 300F keep the test at that temp for a good 20 minutes then plunge into ice (or cold) water. If there is no crazing then I bump then temperature up to 315F and repeat. If there is still no crazing then bump it up 325F and repeat. Temperature range of 325F to  approx 40F is 285 degrees, from boiling to cold water is approx 170 degrees which obviously isn't as harsh a test. For dunting testing I  heavily glaze the interior only of a cylinder shape, freezer overnight then boiling water poured inside.

[LeeS]

Quote

400F to ice water is still pretty extreme. What I've been doing for years is use the kitchen oven and start at 300F keep the test at that temp for a good 20 minutes then plunge into ice (or cold) water. If there is no crazing then I bump then temperature up to 315F and repeat. If there is still no crazing then bump it up 325F and repeat. Temperature range of 325F to  approx 40F is 285 degrees, from boiling to cold water is approx 170 degrees which obviously isn't as harsh a test.

I actually left out that I slowly ramped up the test conditions until I got some sort of result. After boiling water, I started at 300F, then tried 350F, then 400F. I found if I'm not at least somewhat aggressive with the testing, my test cups will appear fine but when I glaze something larger it will craze.

Quote

For dunting testing I  heavily glaze the interior only of a cylinder shape, freezer overnight then boiling water poured inside.

This is a great idea, I'll definitely give it a shot next firing. In general I think a thicker glaze layer on my tests would help suss out issues. I glazed this round "properly" like I would for finished ware, but that doesn't give me quite as much data about more marginal fit issues.

[Min]

Have a look at Hansen's 300F to ice water test here. It's interesting that he only goes to 300F and mentions using test pieces that are 3/8" thick as thinner tiles cool too quickly plus not re-using the same tiles after they have been stress tested for crazing.

 

[LeeS]

Thanks Min, very interesting. I was inspired by Hansen's tests, but didn't read the details of the test process too carefully. His reasoning for thicker tiles seems sound.

I'm curious if anyone knows why you should test with a different sample each time? I have a pretty small kiln so that would be a huge pain. I guess once I've moderately confident in my clay body and glaze it could be worth it.

[LeeS]

Very quick update. I got a fancy new (well, new to me) scale and ran significantly more accurate absorption tests, then compiled the data:

Nothing surprising here, just happy I was able to build some confidence in these numbers.

B0013/B0015/B0019 are all very clean and translucent. Getting rid of Custer feldspar helped noticeably, both with vitrification and translucency. Swapping EPK for Grolleg would probably help more still, but I'm happy with where it's at and the diversity of clays will hopefully mitigate batch-to-batch variation.

I also made up a much larger batch of slip, cast, and bisqued some larger cups in B0019. Hopefully in the next few days I can glaze fire again and see how badly it warps on big, thin walled pieces. Given the translucency I don't have high expectations, but it seems to be in the right place on the vitrification curve.

I'm debating testing one more body, with even less feldspar than B0019, to use if the existing tests warp too badly. It'll probably depend on my patience.

[Min]

Have you tried New Zealand Halloysite? I've played around a bit with it in casting slip recipes (cone 6), I found it acts very differently in the bucket and while casting than epk or grolleg. Makes an extremely white casting slip. Just giving you another rabbit hole... 

[LeeS]

2 hours ago, Min said:

Have you tried New Zealand Halloysite? I've played around a bit with it in casting slip recipes (cone 6), I found it acts very differently in the bucket and while casting than epk or grolleg. Makes an extremely white casting slip. Just giving you another rabbit hole... 

I've heard of it; and wow is it tempting! You're right that it's another rabbit hole though and I should probably get out of this one first :D.

[Mark C.]

Since you are in Cal. Have  yo tried MC porcelain cone 10 casting slip from Laguna Clay co. I have used tons of it over the decades with zero issues. I did have to add 15% epk but only to stiffen it as I cut out 1/3 of the form away. That said its good stuff-I bought it in dry form.I fired it to soft cone 11 as well.

[Mark C.]

I have some cone 10  porcelain slip formulas as well-I recall they are not as exotic as yours (less ingredients less money) are but they worked fine in college . I'll pots them in a day as I'm out of time now

[LeeS]

Quote

Since you are in Cal. Have  yo tried MC porcelain cone 10 casting slip from Laguna Clay co. I have used tons of it over the decades with zero issues. I did have to add 15% epk but only to stiffen it as I cut out 1/3 of the form away. That said its good stuff-I bought it in dry form.I fired it to soft cone 11 as well.

I couldn't find anything on their website with that name. It seems the only ^10 porcelain casting slips they list are 500L and Toshi?

I haven't tried anything from Laguna though. I considered exploring their stuff before embarking on this experiment but making my own body sounded more fun :). That'll be my next stop if this turns into a failure.

Quote

I have some cone 10  porcelain slip formulas as well-I recall they are not as exotic as yours (less ingredients less money) are but they worked fine in college . I'll pots them in a day as I'm out of time now

I would love to see them! No rush of course.

Yeah I'm developing this without really worrying about cost, because it's just a hobby and I don't go through enough clay for it to really matter. Obviously your production world is a lot different and the cost of raw materials matters, which I completely understand.

I've also ordered some PVA resin to experiment with organic binders, but that's yet another can of worms :).

[Mark C.]

MC porcelain was never on website -call them about it. I do not recall but I think it was a standard mix for them-they can make it dry bagged for you-you will need to contact them on this.Its a slight chance it was custom blend but I do not think so.

my collage porc slip cone 9/10 is

EPK 25

om 4 25

 feldspar (back then we used Del Monte feldspar) custar-25

silica 25

you will also need to add Darvon 7

Back then it was soda ash and soduim silicate -both not good on the molds

 very simple recipe

[neilestrick]

On 1/28/2021 at 11:39 AM, LeeS said:

Maybe the is a sign I should fire lower..

You can make a white, vitrified, translucent body at cone 6 or 8 just as well as at cone 10. If you're firing in an electric kiln, it's a good idea to fire lower. Firing costs will be lower, your elements will last a lot longer, and the kiln itself will last a  lot longer.

[LeeS]

Quote

MC porcelain was never on website -call them about it. I do not recall but I think it was a standard mix for them-they can make it dry bagged for you-you will need to contact them on this.Its a slight chance it was custom blend but I do not think so.

Thanks a lot for sharing Mark. I'll definitely consider that Laguna option if this project doesn't work out.

The recipe looks like a fairly standard 25 porcelain. I started at a pretty similar point, but because it's a casting body I was able to use all kaolin without worrying too much about plasticity. The biggest adventure has been the feldspar. I've had very different results with different feldspars... Well, mostly Custer just doesn't flux that well and you need far closer to 25% compared to the other types.

Yeah I much prefer Darvan to sodium silicate. It's much easier to use as well, the deflocculation curve is very flat.

Quote

You can make a white, vitrified, translucent body at cone 6 or 8 just as well as at cone 10. If you're firing in an electric kiln, it's a good idea to fire lower. Firing costs will be lower, your elements will last a lot longer, and the kiln itself will last a  lot longer.

You're 100% right and this has come up a few times, but I'm obstinate about ^10 for no good reason. I understand it means about half the element life and is much harder on the kiln in general. I don't fire often enough for the electricity to be a notable expense, but it's still a fair point.

I think it's because I enjoy playing around with the chemistry without the added complication of boron, but that's obviously a mediocre reason and there are lots of resources for making ^6 bodies and glazes.

I'm also working on a vacuum kiln design for ~1500 -- 1600 C (^20 -- ^30) to play with sintered zirconia and other engineering ceramics, but that's a ways off. It won't use wound metal elements though.