thermal shock

Adminstrator's Note:  The message below has been copied from Joseph's comment on the the Understand and Avoid Thermal Shock tutorial. 

Great, thorough, down-to-earth explanation, Paul.  Thank you. 

May I suggest the addition of a page describing the Thermal Shock temperature range and tips on traversing this range on the way up, and the way down.  How do you feel about a short hold at 400F on the way up to stabilize temperatures as one method of avoiding thermal shock on the way up?  I've seen schedules with fast ramp rates (turn the kiln off) below 800F for small pieces, controlled ramps down to 300F (for thicknesses less than 1/4", and controlled ramps down to 100F, or even 75F (for thicknesses of 1/2" or more).  Can you explain some of this that would aid my understanding? 

Paul, I have what I think are thermal shock cracks in 1/2" thick leaded crystal blanks that I am making for a beveled panel.  At least they fit the description and pictures in Boyce Lundstom's books for thermal shock.  I controlled the ramp down from 800F to 300F at 100 DPH.  From Bullseye Tech Notes and System 96 Firing schedules, 100 DPH over this range should have been a safe approach.  If I have deciphered the Labinski schedules in Shop Notes correctly, 75 DPH would be the recommended rate.  Can you make any sense from all this conflicting information?  I am thinking of trying 50 DPH from 800F down to 100F on my next attempt.  Comments?

I should add that I am first doing a full fuse with lead crystal coarse frit to make a clear 1/2" blank.  I have used the 800F to 300F at 100DPH schedule for this first firing without any problems.  I do have "seeds" in the blank since I am not fining it for any length of time at fusing temperature.  I am only holding it long enough to get a smooth even surface.  Once the blank has reached room temperature, I leave it in the fiber blanket dams backed by kiln furniture, and lightly (sparsely) sprinkle the surface with elemental pure copper powder.  I then fire it to 2100F for an hour.  (100 DPH to 400F, hold 30 minutes, 100DPH to 1000F, hold for 1 hr, AFAP to 2100F, hold for 1 hr, AFAP to 1100F, hold for 1 hr,  300DPH to 1000F hold for 2 hrs, 100 DPH to 900F, hold for 3 hrs, 100 DPH to 800F hold for 2 hrs, 100DPH to 300F, kiln off.)  When the kiln reaches room temperature I open it and find broken glass. 

I am getting the copper diffusion I want: the most intense, striking, magnificient blue-green wisps and streaks throughout clear crystal that you have ever seen!!!  (And no seeds in the final result!)  This glass is absolutely STUNNING!!!  At least the half dozen pieces of it are stunning.  Now if I can only get it to come out in one piece.  Any suggestions? 

I should also say that at 2100F the viscosity of the glass is very low, like hot honey is my guess.  (I have to get beyond copper melting temp so the atoms of copper will diffuse into the glass to create the color effects.  This is not a copper inclusion, but a partial diffusion.  Not a full diffusion, like in a crucible melt to make a teal colored glass, but a partial diffusion to create wisps and streaks of intense color dispersed within a clear crystal base.)  I have sprung leaks under my dams, ruined a kiln shelf and have a few golf ball sized divots in the kiln floor that I have to repair.  I have to go to a closed sided mold to contain the glass.  Simple dams are not sufficient for containment.

Paul, I realize that few have tried what I am attempting to do.  There just is not much information to guide me, and it has been a long trail of guesswork, trial and error to reach this point.  I have finally achieved the color effects I envisioned.  The right mold material will solve the containment and kiln damage problem, and I think I have a handle on that for the next experiment.  But the broken glass is disheartening.  Any insights you may have would be appreciated and most welcome.  Thank you.

--Joseph 2bears, Lomita, CA


>> How do you feel about a

>> How do you feel about a short hold at 400F on the way
>> up to stabilize temperatures as one method of avoiding
>> thermal shock on the way up?

With few exceptions (like casting where there are considerations for the mold), I think that holds/soaks below 900F are one of the first warning signs that the schedule might not be a good one. 

The goal (during intial heating) is to heat the glass evenly.  If you have done that, there is no need for a hold.  If you haven't done that, you should have.

>> I have what I think are thermal shock cracks in 1/2" thick leaded crystal
>> blanks that I am making for a beveled panel.  At least they fit the description
>> and pictures in Boyce Lundstom's books for thermal shock.  I controlled the
>> ramp down from 800F to 300F at 100 DPH.  From Bullseye Tech Notes
>> and System 96 Firing schedules, 100 DPH over this range should have been
>> a safe approach.  If I have deciphered the Labinski schedules in Shop
>> Notes correctly, 75 DPH would be the recommended rate.  Can you make
>> any sense from all this conflicting information?  I am thinking of trying 50 DPH
>> from 800F down to 100F on my next attempt.  Comments?

If you are working with lead crystal I would trust Libensky.

Bullseye and Spectrum schedules are based on the Glass Engineer Handbook by Shand which assumes that you are firing from top and bottom.  I don't know about Spectrum, but Bullseye specifies this in the footnote in the technote on annealing thick glass.  If the shelf was a perfect insulator (which it obviously isn't) you would have to anneal for twice the thickness of the glass.  In reality it is somewhere between the published schedule and twice the thickness.  Not much help eh?  As a rule I always anneal for at least "one thickness" more on Bullseye's chart than the actual glass.  That's a lot of extra time for thick glass -- but I haven't lost one yet.

 >> I should add that I am first doing a full fuse with lead crystal coarse frit
>> to make a clear 1/2" blank...[snip]...lightly (sparsely) sprinkle the surface with
>> elemental pure copper powder. ...[snip]...When the kiln reaches room temperature
>> I open it and find broken glass. 

Are you not letting the glass cool to room temp before applying the powder?  Could the glass simply be breaking from not letting it cool sufficiently during the first firing?  Is it possible it isn't annealed properly from the first firing (it would be more likely to thermal shock on the 2nd firing.  Is it breaking on the way up or down (are the edges sharp or fire polished)?  Pure metal can create a heat sink which might thermal shock the glass.  Also, at 2100F I have no idea what the copper might be doing to the compatibility of the glass.

I'd love to see pictures of the glass.

Helios Kiln Glass Studio


Cracks in Lead Crystal with Copper Diffusion

Thank you, Paul.  I know how difficult it is to diagnose a problem without being able to see it and with only limited information.  I had the occassion to discuss the problem and show the glass to John Williams, owner of Pacific Art Glass, this weekend.  His conclusion was "Compatibility!"

He thought my firing schedule was OK.  What he meant by "compatibility" was that the glass was not compatible with the kiln furniture dams and kiln shelf (all mullite).  I had done the right thing in lining the dams with fiber blanket, in that the fiber would provide the needed cushion for expansion and contraction against the fixed position of the mullite pieces which expand and contract at a much lower rate and amount.  But my dams had leaked in several places and everything was "glued" together by the solidifying glass.  As the crystal blank contracted during cooling, it was being pulled apart by these "anchors".  Something had to give, and it was my crystal blank. 

As I thought about this it started to make sense.  I knew exactly where the dams had leaked, since I had to clean up the mess.  Every fracture and crack in the glass corresponded precisely where those leaks had occurred.  At this point, I think John Williams has nailed it!

BTW, all of the breaks were sharp and straight.  This is what had me thinking they were due to thermal shock.  I see now that the fractures are a symptom, not THE problem.  The problem I have is containing the glass at the high temperatures I am working at.   And the kiln was at room temperature after the first successful firing to make the blank, before it was opened and the copper powder applied.  That fact does seem to support John Williams assessment.  The glass was contained by the dams while full fusing at 1500 F to create the blank, and I had no fractures.  The leakage occurred during the second firing to 2100 F.  The fact that the breaks were sharp, indicate that the fractures occurred during cooling, after annealing, when everything was "glued" together on the kiln shelf. 

First I have to repair my kiln.  I procured some cement for that today.  Then I have to prepare a mold that will contain the glass.  I am going to try some "700" Vermiculite board, 2" thick.  With a plunge router I will mill out a well 1" deep, then cut 5 to 6 degree draft angles on the edges with a hand chisel.  I have some high temperature kiln wash used in glazing ceramics that is good to beyond cone 10 that I will use to coat the mold.  That should solve the problem, I hope.  Whether or not the vermiculite board will withstand more than one firing up to 2100 F is just something I will have to try and to discover. 

It will probably be a week or two until I have anything further to report, but I will follow up and let you know how it goes.  I will be taking photos of successful glass and I will share those with you when I have them. 

Thank you for all your help, comments, and suggestions.  They are greatly appreciated.

--Joseph 2bears, Lomita, CA

Joseph - I think you are

Joseph - I think you are over complicating the solution Smile

All you need for the dams is an old cut up kiln shelf and some fiber paper (thicker is better).

Here's a picture of the setup as viewed from above:

dam for glass

Not shown is that you want a piece of fiber paper under the whole setup.  The result is that the fibre paper works as a gasket of sorts between the kiln shelf slices (or other appropriate kiln furniture).

To reinforce the sides, simply put something with heft against the dams.  Additional pieces of a cut-up kiln shelf works great.

I've used this setup countless times, including for 2" thick paperweights.  Done right it never fails.

- Paul

Helios Kiln Glass Studio


Thank you Paul.  What you

Thank you Paul.  What you have depicted is EXACTLY the method I use in all my firings for thicknesses greater than 1/4".  This method worked very well to fuse the crystal frit into a 1/2" slab at 1500 F.  I used 1/8" fiber blanket and first cut a piece to cover the kiln shelf.  I used pieces of 1" thick kiln shelf cut into 1" wide strips to form a box on top of the underlying fiber blanket.  Then I cut 1" wide strips of the 1/8" fiber blanket to line the inside edges of the box.  I cut and fitted the peices carefully so there were no gaps or holes at the seams.  I have found that this method ALWAYS works for full fuse firings. 

But when I fired the crystal to 2100 F, it did not contain the glass.  Glass ran through the corner seams and glued the ends of the shelf strips together.  Glass ran under the bottom seam, across the edge of the underlying fiber blanket, onto the kiln shelf, and over the edge onto the floor of the kiln.  It was as if you had poured a cup of water into the box.  Well, not quite as dramatic, but there were multiple leaks.  This was in spite of the fact that the whole containment system was carefully constructed so there would be no gaps in it. 

At full fuse temperatures glass has a fairly high viscosity.  It will move, but it moves slowly.  It has enough surface tension to generally hold it in place within the dams.  If there is a gap in the fiber blanket lining, it will enter that gap and form a ridge on the edge of the glass, but generally it will not flow through a 1/16" gap in the 15 to 30 minutes time at full fuse.  Even at raking temperature, 1650 F, the viscosity of the glass is fairly stiff.  If you have ever pulled a rake through it, you know just how stiff the glass is at even that temperature. 

We know that different compositions of glass have widely varying viscosities at different temperatures and generally the higher the temperature, the lower the viscosity.  Your assessment that the copper diffusion may be changing the composition of the lead crystal is probably correct.  My guess from observation is that it dramatically lowers the viscosity.  I have also observed that the copper tends to sink to the bottom.  Partially diffused copper on the bottom is EXTREMELY caustic.  It eats right through the fiber blanket and into the mullite kiln shelf.  The effect is like dropping a drop of vinegar from an eye dropper onto a layer of baking soda.

Through a series of disastrous experiments I have learned to control the thickness of glass and copper penetration (sinking) based on particle size and distribution so that each particle is completely diffused into the glass before it reaches the bottom and starts eating the underlying fiber blanket.  I have also learned to keep copper particles at least 1/4" from the edge of the crystal blank for the same reason. 

When the copper particles sink into the glass, it is not a straight descent.  From the diffusion traces they appear to swirl in random paths.  This is what creates the beautiful color effects I am seeking.  But it is quite possible that some of these particles do get to the edge where they would eat holes in the fiber blanket.  That may be a contributing factor to some of the leaks, but I don't think that is what happened in the last experiment.  I did not have the characteristic "knots" on the edge or bottom of the tile where this has happened in earlier experiments.  The "happy" factor of the last experiment is that I think I have finally developed a good recipe for glass thickness, particle size and particle distribution that will create the exact effects I want. 

The "unhappy" factor is containment.  I had 4 leaks that each dribbled out about one to two teaspoons of molten glass.  There were another 3 leaks of about 1/4 teaspoon that ran under the mullite dam strips, but did not make it to the kiln shelf.  My best guess is that the viscosity of the glass with the added copper at 2100 F is about that of hot honey.  That is pretty runny, and acting more like water than molasses (which is how glass acts at fusing temps). 

That, Paul, is why I am "complicating" things and going to a fully contained mold.  If anyone has ever tried this, they are not admitting it.  And they probably gave up on the idea following 2 or 3 disasters, destroying a couple of kiln shelfs, assorted kiln furniture, and suffering a week of down time to repair the kiln.  I'm not bright enough to accept the fact that "it cannot be done".   I have suceedded in achieving the color effects I envisioned when I started this with the last disasterous firing.  I am so close to ultimate success that it has heightened my excitement and I just can't abandon the quest.  I have to go on. 

Thanks for all your help.

--Joseph 2bears, Lomita, CA

Have you tried a thicker

Have you tried a thicker fiber paper and adding more weight behind and on top of the dams?

You can put some shelf strips across the tops of the dams (like bridges) and weigh them down.

With sufficient weight the only way the glass is going to run is if it seeps through the fiber paper.  I know lead crystal is very viscous at those temps -- I don't think it is that thin.

Helios Kiln Glass Studio


No Paul, I have not tried

No Paul, I have not tried weighting the dams.  I am working in a small 8" Paragon Caldera and there isn't much room.  I'm damming on the 7" shelf to make the largest blank I can, which ends up being 4-3/4" square.  My larger kiln is a fiber blanket model, which will only go to 1700 F.

If you are familiar with the small Caldera, the elements are in the side walls below the middle of the kiln depth and the thermocouple is above the elements sticking into the space about 2 inches.  In the last experiment, I did use 1" fiber blanket on the bottom, but then ran into an interference problem with the dams and thermocouple.  The only way around this obstacle was to put the shelf on 2" posts to get it above the thermocouple.  By the time I had it all put together, I only had 1/2" clearance below the lid, and only 1/2 clearance around the shelf for heat to rise to the glass.  Since my shelf strips were right on the edge of the 1" fiber blanket, the weight of them wanted to compress the fluffy blanket and they would roll over and off.  I had to put additional support between the shelf strips and the kiln walls to hold everything in place, which further restricted heat flow.  I extended my firing times through trial and error, and actually had to fire the base blank twice to get a full fuse.  I used that experience to estimate the firing time for the high temperature firing time with the copper. 

When this came out of the kiln, the original 1" fiber blanket was 1/2".  That may be part of the problem.  I've been using what I have in the shop, which are refractory materials meant for glass below 1700 F.  Maybe I should have started with, and should be using higher temperature refractory materials.  In my original research I did look at this and spent some time at the Zircar website.  But those materials are much more expensive, not readily available, and minimum purchase quantities are way beyond the little I need right now.  After talking to a number of artists who have pushed the envelope here and there in their endeavors, the consensus of opinion and experience was that the fiber blanket used by glass artisans was good for at least one firing above 1700 F.  Since that is what I had in the shop, that is the choice I made. 

I talked to Petra Kaiser.  Same story.  The Kaiser-Lee board is good for multiple uses below 1700, but it will maintain its integrity above 2000 F for one firing.  Above 2000 F it does degrade and will crumble apart when the glass is removed.  The Kaiser-Lee board will stick to the glass, but can be scrubbed off easily.  That has been my experience with the fiber blanket.  But Kaiser-Lee board is another expensive option for a one-shot use.  That is why I have decided to try the Vermiculite board.  It is much less expensive. 

So you see, Paul, aside from trying to do the impossible, I am further compounding the problem with an inadequte kiln, and inadequate refractory materials, with inadequate financial resources, in an undersized shop space.  And most folks think Don Quixote was nuts.  Well...the dear Don has got nothing on me!

--Joseph 2bears, Lomita, CA

Trial and re-trial...the learning curve

Joseph, I am a fused glass artist in the LA area as well. I also know John at Pacific Artglass, and get most of my supplies from him.

I finished my MFA in warm glass in 1990. At the time, no-one was doing more than melting Wasser glass in small enameling kilns. My work was huge, heavy, deep slumps, thick casts....basically, impossible. I was also working with copper in my glass. My equipment consisted of one home made, over-used, underbuilt, RIT student monster we called the Kiln of the Immaculate Conception. Enough said.

I made many costly mistakes in my endeavors. Lots of "flowing" leaks, full melt-downs, the whole disaster thing. I documented in detail every one of them, and made many more, but never the same, mistakes again. After countless replaced elements and glass soaked brick pried out and redone,  I finally succeeded against all odds. My 110 page dissertation is a testament to insane perseverance. You can see the results in my retro gallery under "thesis" at

My point is, if this is what you really want to accomplish, then keep trying. Don't let equipment or lack of money stop you. Document, document, document, and ask anyone who will listen to you. They won't all have the right answers, of course.

My solution of the meltdowns was:1) to do a solid pour (needed to buy an engine hoist to do this :*{ ) because at the high range, the mold materials striate (split apart) at the seams if you don't have a continuous pour; 2)reinforce your mold with something like chicken wire. Make sure to not have any exposed to the glass or the air, though; 3) Additives in the mold mix help.  I can share my mold mix recipe (not commercially available, but you can get the stuff and make it yourself...very old-school) if you email me at I can give you my number, too, if you want to talk about ideas. My humble mix held up to 2200 + without failing, and that is containing 350 pounds of lead crystal cullet. I melted down poor little Fenton crystal teddy bears and birds who were Fenton rejects...their little heads blurring together in veiled swirls. So sad. Also slumped Bullseye clear for the figures...much easier.

Here's my newest piece from my Winter collection. Less insane, for certain. Hope you like it.Fused glass with inclusions 14"x11"

Anyway, best of luck! Let me know how it goes. Maybe we'll meet up around town...Pacific has a Sweetheart sale coming in Feb. that I might go do.

Happy Baking!

Cynthia Ann Swan, Terra Pax Studio, Los Angeles, CA

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.