Jump to content
Sign in to follow this  
High Bridge Pottery

Bubble Bubble Toil And Trouble

Recommended Posts

So.

 

Bubbles.

 

As you may know I have been spending time tweaking firing cycles to get rid of glaze bubbles. I thought it was the whiting and dolomite gassing too much. Bought some wollastonite in the hope this would remove the problem.

Here is the whiting dolomite glaze.

gallery_23281_968_22515.jpg

 

The wollastonite glaze

gallery_23281_968_32652.jpg

 

Now, they really don't look that different. Do glazes always bubble if they are thick? I would think no but unsure on that. Could it be my bisque? Go about 100degC per hour to 980. 

Funny thing is I happened to be doing a cone 6 instead of 9 firing for somebody else so put a few tests in to see what happened with a hold at 830degC I think.

gallery_23281_968_38971.jpg

 

Maybe this is not a fair comparison as application is different. Still bubbles there and probably less because of thickness.

 

Here is a firing with controlled cool to see if bubbles went.

gallery_23281_968_94404.jpg

 

Also tried adding 10% more of different fluxes, that didn't change anything either.

 

Should I give up and live with slight bubbles in the glaze? Is it even possible to get rid of them? When fired thin on a tile they are far less noticeable but I want to have a nice thick layer of glass bubble free. 

Share this post


Link to post
Share on other sites

Do you know if the bubbles are coming from the glaze or from the body? 

 

Are you holding your kiln when it bubbles to give the glaze time to bubble out then seal back over? 

 

I had a lot of micro dimple issues going on that I solved with 2 holds on the way down. But I imagine everything is different for different glaze materials.

Share this post


Link to post
Share on other sites

Um...am I weird for thinking that it looks gorgeous?? ^_^ There is just something so great about a speckled pastel pink that I can't resist. I'm soooo jealous you get to experiment so much! :) I wish I had more to offer...maybe try firing it reeeeeally slowly?

Share this post


Link to post
Share on other sites

Min, I really don't think it is over fired but it is interesting the cone 6 had the least bubbles. Why would firing hotter make more bubbles? I think it has the thinnest application of all photos so might be that and nothing else. Doing a melt test showed me that it is very viscous and it feels borderline devitrifying from the crystals in the glass.

 

Here are the two recipes, a little different as one has no Mg but still similar enough.

gallery_23281_1027_68239.png

 

Joseph, I thought it was the glaze as you can see the loss of ignition go from 7.9 to 1.6. That seems to have made no difference so now I am looking at maybe the bisque.

 

My initial glaze firing went 150deC per hour to 1140 then 60degC per hour to 1260 with a 15 min hold. I tried out a 40 min hold around 830degC to allow for gassing in the glaze fire then normal firing150/h with 60/h for last two hours.  I tried the 830 hold with controlled cools either dropping 100 and holding for 15 min or dropping 50, hold then 50 and hold.

 

Guinea the bubbles/photos are beautiful :D but I want to know why and control if the happen or not  :(

Share this post


Link to post
Share on other sites

If the glaze is over firing, it will begin to boil after it melts properly. Kind of like chocolate. This hypothesis would be supported by more bubbles in both the higher fired tests and the one with the hold (more heat work).

Anyone with insight want to check it against limit formulas for cone 6 vs cone 10?

Share this post


Link to post
Share on other sites

This issue has been vexing for me as well. I know a lot has been said on it already in these forums, even in the relatively short time I have been here. If you have not trawled through earlier posts including keyword bubble would suggest it is worthwhile.

 

Would suggest the problem is not overfiring. I have some old photos of bubble formation in a lower fire glaze posted in my gallery that show that the higher you fire the more bubbles disappear. And those same pictures show what a long soak will do to remove even more bubbles. And I have some digital microscope photos like yours in the gallery of a commercial porcelain plate that show that even industry cannot completely get rid of this problem.

 

So I have concluded for the moment that this is not a question of whether or not you will have them, but rather how many there will be and what size. And that matters, because the ceramics industry is selling us stuff that has bubbles, but the bubbles are small enough and few enough that you can not really see with the naked eye, so the simple reality for pretty much everyone looking at those glazed objects is that they do not even have bubbles. Nuf said.

 

Back to your specific issue, it may just be the photos and lighting, but is it possible your glaze is not fully matured, ie possibly not fired high enough and/ or long enough. The background behind your bubbles seems to be somewhat crystalline looking, almost like the melt is not quite complete. Reminds me of certain cells on Currie tiles where either the amount of alumina or silica is too great for it to all be completely absorbed in to the melt.

 

Have you done a currie tile for this glaze? When I was clear glaze testing a while ago various Currie tiles showed me that I could get a much better (ie glossier, cleaner, smoother but still un-crazed) glaze by using much more flux than I had been using with only the limit tables as a guide. If we had the molar analysis for your recipe I think we would see it was very high in silica and alumina on a percentage basis and consequently pretty low on fluxes...

 

Similarly, suggest you might want to assess the viscosity of your glaze ingredients and try to reduce ingredients that are making the glaze to viscous to pass bubbles well. Hamer and Hamer has the best discussion on viscosity I have seen so far.

 

Agree with Tyler that all else equal thinner is definitely better. Simply less place for the bubbles to hang out. However, the quid pro quo for me has been that thinner glaze application requires extra smooth clay surfaces (eg the kind of finish you get on fine grained porcelains) to get a good overall finish, otherwise the glaze can look too thin and ratty.

 

Looking at your recipe (using Insight Live I see ) it seems like the silica alumina ratio is a bit low for a transparent, if that is what you are shooting for. Seems like you have plenty (and maybe even too much?) of silica though, so maybe notch down the alumina a bit? Looks like the Neph Sy has got plenty of alumina so maybe you can dump some of the clay, which is always a sneaky importer of impurities anyway.

 

I know you are sourcing the boron from a frit which is the best way to go, but since boron melts so early and so aggressively I guess there is a possibility it could be involved in your bubble formation...but its benefits may make you want to keep it even if it were the issue.

 

Finally, what clay are you using? My experience has been that it is best to test this kind of glaze on a high quality porcelain because it provides a clean smooth background with minimal impurities so that you see what is going on with the glaze ingredients rather than wondering if the clay is introducing unwanted issues (either visual or chemical) into the test.

 

Whew! Sorry so wordy but hopefully helpful...

Share this post


Link to post
Share on other sites

I think I remember seeing that post a while back with bubbles.

 

Your right that the glaze is borderline too much silica and alumina but it is still very glossy. I did do a melt test and the glaze doesn't budge really, very viscous.

12096549_501805496663676_5560231216250189035_n.jpg.c9ce1e35f8e8b4d8911b6f5e56f0ff23.jpg

This glaze came from the currie tile below. Probably somewhere near the bottom right. The quality of the photo is bad as linked from facebook but all the melted ones have bubbles.

12065738_498255163685376_2785936701770980349_n.jpg.bee353da9ea328c01a803a6c35278e84.jpg

After doing the melt test I thought that it could be far too viscous. I did an arbitrary test adding 10% more flux. Calcium Borate Frit, Spodumene, Potash Feldspar, Zinc and Dolomite across each row.

gallery_23281_912_1359255.jpg

 

Still all with bubbles, I thought an extra 10% frit would make it more fluid for sure but the bubbles still stay :(

I have tested it out on buff stoneware, a darker grittier stoneware and white stoneware. Got to say the glaze looks pretty similar on all the clays no matter the thickness. I think after looking at the glaze tests again today I do need to add a little more flux. Now that I know my whiting and dolomite are not really effecting the bubbles I can add some dolomite back for MgO, maybe a smidge more frit too. 

 

Here is a test on the darker clay. 5% Titanium, 5% rutile, 3% titanium in row with .4% cobalt, 3% titanium in row with 1% copper, 3% rutile in row with 2.5% iron.

gallery_23281_912_506242.jpg

 

The best colours come from thicker glaze but also the bubbles. I am testing the glaze in a bisque firing and see if it has started to melt.

Edited by High Bridge Pottery
Old photos missing, think I put back the right ones.

Share this post


Link to post
Share on other sites

I fail to see where chemistry/glazes and nature part  :huh:  ^_^ but that's just me.

 

The most interesting thing I think from this is that removing most of the rocks that gas I am still left with the same amount of bubbles. I am trying a new bisque firing but I will also try glazing an already fired tile and see if that changes anything.

Share this post


Link to post
Share on other sites

Have you tried increasing the more fluid alkaline fluxes, lithia for starters, for some of the calcium and magnesium to reduce the viscosity? That plus perhaps a decrease in the alumina would be a direction I would go in. Are you aiming for a transparent like in your first test tiles or a matte like in the latter ones?


Share this post


Link to post
Share on other sites

Sorry it is probably my fault for posting the tiles under different lighting conditions. The bottom two tiles are still very gloss and really not far away from the originals on the first tile. I can scrape away at all of them with a knife or key an none are marking so all pretty vitrified. 

 

I have bisqued some more melt testers so time to try out some ways of increasing the melting properties. Would like a transparent but as I say the ones that look matte are very very nearly transparent, just a little too much silica.

 

Put a ball of glaze in the bisque to see if it is melting early. Doesn't look too melted although it is pretty tough. Hit it with a hammer at least 10 times and it didn't break. Porous like bisque and soaks up water.

gallery_23281_912_963067.jpeg

Share this post


Link to post
Share on other sites

In my adventures in bubble technology I decided to remove all clay from the glaze, see if that was causing the bubbles. Looking at the fired results I could have probably skipped adding silica but I like to keep it up near a 4. I also decided to remove zinc as I am looking to reduction fire with this glaze too. Tested the original glaze above and with 10% more frit addition, 005.

 

Glaze 010B really looks like it has too much silica as I put in too much KNaO to have a nice melter and it has hardly moved anywhere. Think you are right Curt about me probably pushing in too much Si/Al.

gallery_23281_1027_32014.png

 

When thin on a glaze tile it actually is very hard to see any bubbles but I want a glaze that can go on thick. My cobalt carb is still spotty after going through a 120 mesh sieve  :rolleyes: I will just have to like the spotty blue. The blue is 005 glaze.

 

gallery_23281_912_413664.jpg

 

Conclusion - I have no idea what causes bubbles in glazes. I also tried firing the glaze on an already cone9 tile, same results.

Share this post


Link to post
Share on other sites

OK, a few more things bubbling up (sorry) for me as I re-read your posts above.  In no particular order.

 

First, awesome pics and many, many thanks for taking the time to put them up, as they provide valuable learning for all of us. 

 

I am having trouble rectifying the silica and alumina numbers your super-imposed on the currie tile with the Si:Al ratios reported for those recipes by Insight in the recipe boxes.  Not meaning to be nit-picky, more that I am concerned that the currie tile you said you chose the glaze from does not match up with the recipe you set out?  Is it possible that the recipe is not the right one?  Only asking to make sure this is not (some of) the source of the problems (or rather just of my confusion!)

 

The glaze ball in the bisque was a good idea.  I suspect it was hard due to the super melty boron frit you were using, which would be well active around bisque temps (and possibly eutectics for the other ingredients but hard to know them all)  You mentioned that it was porous, but how porous exactly?   The % porosity of that ball might give some indication of how matured the glaze was.  I know that is really a clay body testing approach, but might give you some idea of how early you glaze is sealing over.  If it is happening too early, then some body decomposition gases might not be getting out before the glaze surface seals over, and then have to make their way up through the melt. 

 

Following on from this,  I don't think insufficient organic burnout is really the issue (that would cause bloating) but if you think it is suggest put a one hour hold in the bisque (not the glaze, too late then) firing at around 800 Celsius.  

 

Similarly, the LOI numbers on their own can seem important, but I think it is probably just as important to know when is that LOI occurring?  even low LOI can be bad if it is happening during the wrong time (ie, late) in the firing when the glaze is actively melting.

 

I think the elephant in the room is still the clay bodies.  Do you know the chemical composition of your clay bodies underneath these glazes?  Increasingly in my view this is at least as important as the glaze recipe.  The clay body will represent about 95% of the weight of materials in your pot.  Their off-gassing profile will be just as important as the glaze ingredients, not least of all because gases from the clay body trying to get out of your pot have the farthest to travel to escape!

 

Again, I don't think there is going to be any silver bullet solution to the bubble problem.   It is more an issue of managing it down to a reasonable level.

 

If time and money were no issue I would think about ball milling the glaze down to a, say, 200 mesh level and seeing if that helped.  I am increasingly suspecting that the finer individual glaze ingredients are, the more easily - and the earlier - they decompose.  This avoids big chunks of individual ingredients having to be broken down by heatwork and mechanical forces during the height of the melt, only to release their gasses too late in the day to make their way safely out of the glaze without leaving bubble traces behind.   I have also read that ball milling is common in industry, and may be why they can reduce - but not completely eliminate - the bubbles as a visual artefact.

Share this post


Link to post
Share on other sites

I am not sure what limits you are using, but at cone 6 it barely fits Matrix and Green & Cooper.  At cone 9 Insight, at cone 10 Insight, Food safe, Matrix, Green& Cooper, Cushing Glossy Clear.  Since you probably want cone 6 I would suggest changing your Alumina Silica ratio to get into the cone 6 limits for Insight or better yet food safe limits. This could help the bubbles problem.

David

Share this post


Link to post
Share on other sites

The first two recipes posted with the whiting/dolomite and wollastonite swap match up with somewhere around square 14 and 19 on the tile, giving around 4 silica and .5 alumina. All the glazes are somewhere vaguely in that area. Not sure how porous the ball was but it soaked up water like any other bisque thing so I think it was pretty good.

I was supposed to add the frit by itself to the bisque but being forgetful I forgot  :huh:

 

I did a different bisque firing with those little squares but not a hold for 1 hour. Going to keep experimenting with that. Looked into whiting and dolomite LOI and had a few glaze firing experiments with holds between 800 and 900.

 

I really don't know much about the chemistry of the clay body and after subbing in wollastonite I think I will have to start studying  :lol: it is interesting that I have used three different bodies and the glazes are all pretty much the same. I think it is as you say lots of small things will add up to a glaze with few bubbles as possible. I do want to make a ball mill so maybe now I will get the motivation to build one in the quest of the bubbles.

 

 

 

 

David I do fire to about a cone9 and will want to fire cone 10 gas with this too. The glaze actually looked pretty good when I tested at cone 6, if a little under fired. The reason for cone6 was because of firing some work for another person and sneaking in a few testers.

Share this post


Link to post
Share on other sites

Have you read Hansen’s article on bubbles in glazes? Second to last set of pictures looks like your issue, look at the correlation between melt fluidity and bubbles. Still think you need to reduce the viscosity of your glaze.

 

Just in case you haven't already read it, http://digitalfire.com/4sight/glossary/glossary_glaze_bubbles.html

 

oops, its the last picture I meant.

Edited by Min

Share this post


Link to post
Share on other sites

+1 on the viscosity issue. And Hamer and Hamer ranks oxides by their viscosity which I have not seen elsewhere. A nearby and somewhat related issue is surface tension if you really want to make it interesting.

 

i also think the backside of the firing schedule is important as mentioned above, ie when and where you soak. I used to soak only at top temp in the glaze firing, but now I am starting to include holds after hitting top temps when the temp has dropped off a good 100 degrees or so. Too early to tell if it is making a big difference yet though, but I am hopeful.

Share this post


Link to post
Share on other sites

+1 on the viscosity issue. And Hamer and Hamer ranks oxides by their viscosity which I have not seen elsewhere. A nearby and somewhat related issue is surface tension if you really want to make it interesting.

 

i also think the backside of the firing schedule is important as mentioned above, ie when and where you soak. I used to soak only at top temp in the glaze firing, but now I am starting to include holds after hitting top temps when the temp has dropped off a good 100 degrees or so. Too early to tell if it is making a big difference yet though, but I am hopeful.

 

+1 on the soak on the way down. The last year or so I have been dropping by 100F from the end of the soak at the top temp then soaking for 30 mins. Makes a difference on one of my glazes that is prone to pinholes. In my kiln it adds minimal to nil more heatwork to the cones.

Share this post


Link to post
Share on other sites

I have tried a few holds on the way down, dropping 50 or 100 from top temp and holding for half an hour. I even tried re-firing some glaze tests to simulate firing hotter, see if that changed anything.

 

It is a really viscous glaze so I think working on a few more tests in that area will be a good idea anyway.

Share this post


Link to post
Share on other sites

I think I remember seeing that post a while back with bubbles.

 

Your right that the glaze is borderline too much silica and alumina but it is still very glossy. I did do a melt test and the glaze doesn't budge really, very viscous.

12096549_501805496663676_556023121625018

 

This glaze came from the currie tile below. Probably somewhere near the bottom right. The quality of the photo is bad as linked from facebook but all the melted ones have bubbles.

12065738_498255163685376_278593670177098

 

After doing the melt test I thought that it could be far too viscous. I did an arbitrary test adding 10% more flux. Calcium Borate Frit, Spodumene, Potash Feldspar, Zinc and Dolomite across each row.

gallery_23281_912_1359255.jpg

 

Still all with bubbles, I thought an extra 10% frit would make it more fluid for sure but the bubbles still stay :(

I have tested it out on buff stoneware, a darker grittier stoneware and white stoneware. Got to say the glaze looks pretty similar on all the clays no matter the thickness. I think after looking at the glaze tests again today I do need to add a little more flux. Now that I know my whiting and dolomite are not really effecting the bubbles I can add some dolomite back for MgO, maybe a smidge more frit too. 

 

Here is a test on the darker clay. 5% Titanium, 5% rutile, 3% titanium in row with .4% cobalt, 3% titanium in row with 1% copper, 3% rutile in row with 2.5% iron.

gallery_23281_912_506242.jpg

 

The best colours come from thicker glaze but also the bubbles. I am testing the glaze in a bisque firing and see if it has started to melt.

Your grid is 5x5, Currie grid is 5x7 is that your intent?  Have you compensated for this?

David

Share this post


Link to post
Share on other sites

Yes, I decided I didn't need all 7 and a 5 by 5 would be better. I have a spreadsheet that works out grams per ml, how much of each glaze and each material in each glaze is in each square then spits out a recipe/ratio. Bit crude and could probably be done a lot better for easy manipulation for different tile size.

Share this post


Link to post
Share on other sites

Was wondering the same thing, ie why you left off the bottom two rows of the currie tile.   

 

Those two rows represent about 30% of the information on the alumina in a standard currie tile, which to my mind is quite a bit.   Even small changes in the alumina seem to have an outsized impact on all sorts of things in the finished glaze, more so than silica.  People are pretty relaxed (confident?) about adjusting glazes by just whacking in some more silica, particularly as an oft-prescribed cure-all for crazing.  But when was the last time you heard someone say "oh, to fix that problem just add more alumina" ??   Not often I am guessing. 

 

Have you simply removed the bottom two rows of a standard currie tile (in which case your top five rows follow exactly the same recipe ratios as a normal standard currie tile), or have you adjusted the middle three rows on your tile so that each one now has a different recipe than the middle three rows of a standard currie tile?  Just curious so I know what I am looking at.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.