Thermal shock problem, first time in 12 years, pics

[smasty]

Hi Everyone! I've been a fusing hobbiest for over 12 years, this is my very first thermal shock issue that I'm trying to solve. My kiln is a 12-year old Evenheat, it has been a total steady eddy...absolutely no problems at all. It has about 500 hours of firing time I would guess, with nothing done or replaced on it (been reading about relays in the other thread...). I make these cool scrap glass stars, this one though has sharp cracks. I keep firing it to "heal" the cracks, but the cracks won't heal and they stay in the same place. The pieces stay tight together after cracking, I have the pieces spread apart so you can see the crack pattern. This is a thick piece, about 5/8", and tip to tip is about 13". Every time I refire I add a little more glass over the cracks, so it's getting thicker with each firing. I cut apart my star ring so it doesn't pinch the glass...works great. It almost seems like a COE issue, but I've only ever had S96 in my shop. The sharpness of the cracks tells me it's happening after fusing. Obviously there's a mass differential between the center and the legs. I'm not using any fiber between the steel and the glass...I'm wondering if it needs insulation against the steel. This is probably one of the thicker pieces I've done. What's getting me is that regardless of firing schedule, the cracks are always in the same spot, which makes me think COE. Well...give me your best ideas! (the center piece is a "magless" I did many many years ago!) P.S. Isn't it interesting that the cracks are identically shaped on the two legs???
Here's the last schedule I used: 300 to 1000, hold 1:10, 400 to 1150 hold 1:15, 600 to 1250 hold 1:05, 600 to 1460 hold :20, full to 950, hold 3:06, 80 to 850 hold 3:04, 75 to 750 hold 1:00, 75 to 150. This is by far the slowest schedule I've ever used.

[Marty]

Looks like the glass got through the kilnwash at a few points. Add a strip of fiber paper around the perimeter for insurance. Cutting the ring was a good idea.

That schedule is fiction (who wrote it?)- why the holds for 3:06, 3:04? What practical difference is there between 80 dph ramp and 75 dph for annealing?

Try 200 to 250 dph to 1100, hold for bubble squeeze (or you could squeeze during an hour's ramp from 1100 to 1250), etc. Cut your annealing to 50 (or less) dph between the holds at 950 and 750.

[smasty]

It's my own schedule, I should have just put in 3:00 but it's my rush to get close to 3 hours (programming the rampmaster) and what's a few extra minutes. Why does the up-speed matter when the cracking is happening on the downside? Thanks

[Marty]

It's not the few extra minutes. My issue is with schedules that have no rationale for the processes and become gospel.
It's also about understanding and simplifying the whole process.

[smasty]

Marty....you were very observant, thank you! There is indeed a very thick layer of kiln wash adhered to the star legs...the thickest coatings are where the cracks started. I've ground off all the wash and am re-firing right now with your recommended changes to the schedule. I'll update tomorrow, thanks! I also lined with 1/8" fiber strips, including a double fiber at leg intersections for extra insulation.

Sue

BTW, I had my star ring made by a stainless steel fabricator in Denver, cost me about $7.00. It makes great use of scrap--after my first "pinch" I cut it apart.

[Valerie Adams]

The steel cools at a different rate than your glass, which is why they should be lined with fiber.

Is the area that appears clear actually white glass? While I don't use Spectrum glass, I know that repeated, super-high firings to a few of Bullseye's whites (and a couple other colors) can cause them to shift compatibility.

As a side note, would you be willing to share the name/contact info of your steel person?

[smasty]

Thanks Valerie,
I used Colorado Custom Stainless (303.433.2370). They made me a ton of different size casting rings, square, round, star--they would probably be able to do any type of shape (a fish shape would be cool!). Nice heavy gauge. I was able to give them dimensions of what I needed. It goes low on their priority schedule, but the price is right. It took about 2-3 weeks for them to finish up my order. It's not their normal line of business...if you say "casting rings" they might not know what you mean, but just explain it.
Oh....I do have a base of white scrap. This is probably the 4th or 5th firing of this sucker....I'll figure out a secondary use for it if it fails this time.

[smasty]

After a 26-hour firing cycle, I took a peek (at 175 degrees)...it looks like perfection, just tiny tiny bits of devit since this is the 5th firing. Given the mass, I will wait for further cooling before opening for a new picture. It's a Christmas Miracle! Thanks everyone.

[Bert Weiss]

My opinion is that 950 is too hot for the soak, Bullseye suggests 900. If I am correct, you wasted 3 hours soaking above the anneal range and then the anneal actually took place during the drop to 850, which was speedy. The soaks on the way up were a monumental waste of time and energy. I would just take a steady rate from room temp to top temp. Then anneal. I suggest you use the Bullseye schedule for thick glass, regardless of whether the glass is Bullseye or 96.

That said, I don't know what they say, but if it were me, I would soak at 920 for 3 hours, then take 3 hours down to 820, then take 2.25 hours down to 620 and then 2.25 hours down to 220. MY controller doesn't program by degrees per hour, so I don't think that way. I go from temp to temp over a programmed amount of time. That said, I would probably advise using the BE schedule.

[smasty]

Hi Bert, Thank you very much for your comments. You are right, any time the kiln is running unnecessarily it's a huge waste on multiple levels. Annealing at a lower temp makes a lot of sense given the mass of thicker glass. I will study more about this. I've seen you guys recommend the Graham Stone book....but it appears to be out of print? There's a few copies on Amazon for high prices.

My star, BTW, appears to be perfect. The fiber lining left some edges that need to be cold worked, but all the old cracks have healed and not a new crack to be found anywhere. It packs a punch at 3 pounds.

Sue

[Bert Weiss]

Hi Bert, Thank you very much for your comments. You are right, any time the kiln is running unnecessarily it's a huge waste on multiple levels. Annealing at a lower temp makes a lot of sense given the mass of thicker glass. I will study more about this. I've seen you guys recommend the Graham Stone book....but it appears to be out of print? There's a few copies on Amazon for high prices.

My star, BTW, appears to be perfect. The fiber lining left some edges that need to be cold worked, but all the old cracks have healed and not a new crack to be found anywhere. It packs a punch at 3 pounds.

Sue

http://www.hisglassworks.com/shop/firin ... -html.html This is the lowest price I found. I recommend Stone, not so much for his schedules but for his explanations about his schedules. I prefer Bullseye's schedules to Grahams.

I once asked the top experts, if my schedules (as I posted) or Stone's were better. We both spend about the same amount of time, but spend it at different temperature points. I was advised that mine were better. This was by the top guys at Bullseye. Several years later, they revised their own schedules to reflect the strategy of spending more time at lower temps.

[Barb R]

My opinion is that 950 is too hot for the soak, Bullseye suggests 900.

Hey Bert - if she's using System 96 - why shouldn't she soak at 950, as recommended by Spectrum?

Barb

[Bert Weiss]

My opinion is that 950 is too hot for the soak, Bullseye suggests 900.

Hey Bert - if she's using System 96 - why shouldn't she soak at 950, as recommended by Spectrum?

Barb

I wonder why Spectrum suggests this in the first place? Back in the 1980's I tested both Bullseye and 96 for anneal point characteristics, and both came out, for me, at 920ºF. At one point, I asked Ray Alghren who designed System 96, about my anneal strategy and he said it was a good one. He was also a designer and original owner at Bullseye. At the time I was asking about this, Bullseye had a soak, I think at 950, and then a ramp through the anneal range and down past the strain point. He told me then, that glass anneals quicker at the bottom of the range than the top, and my strategy was the better one. Several years later Bullseye adjusted their published anneal schedules with even lower temperatures than I was using. Spectrum never changed theirs.

Over the years, I have worked with a glassblower who told me that 20 or 30º off can ruin an anneal schedule. I also know that thermocouples can easily drift off 20 or 30º. So, my common sense tells me that 950 might be too hot, at which point the actual anneal is happening on the next step of the program. For many schedules, this might work. The piece in question here is 5/8" thick, which requires a good schedule.