Origin of COE as compatibility measure in studio glass

Pete,
You get two stories mixed up here.
GLASMA was set up towards the end of the 70th to solve the environmental problems around the batch rooms of the Swedish manual glass industry.
Klaus had to close down his colour factory in Germany because the small industrial area got enclosed by housing estates and he had no space to expand into filters et all.
He sold the business to Fridrich & Schebler since they had a factory producing colour bars and rods for buttons and lampworking. The technology there was very primitive compared to the high tech set up Klaus had been running.
More than 10 years ago they invested millions in a new production facility complete with the latest in environmental technology.
They mix and melt there range of glasses there on the premises.
Just as a service to primarily the studio's they have GLASMA produce a pelletised batch that should fit most of their colours most of the times.

Klaus never indicated numbers for compatibility just an indication "Harte W" which you could read as an indication for viscosity because it indicates if it's a hard or soft glass. To bad Klaus passed away last year, but I will try to find out more about the origins.

[Richard Huntrods]

I don't know about fluxing power, but if it's ion size you're after, that's a simple function of the number of protons & neutrons in the nucleus, as well as the number of electrons in the various orbitals.

A periodic table of elements is all that's necessary

http://www.webelements.com/

In essence, down is larger in any column, so it's Lithium (Li), Sodium (Na), Potassium (K), Rubidium (Rb), Cesium (Cs), Francium (Fr). Also, sideways is also larger for a given row. (i.e. Magnesium (Mg) is bigger than Sodium (Na).

Does this gibe with fluxing power (i.e. Sodium larger than lithium, etc)?

Cheers,

-R

Ion diameter and molcular weight might not be related in the way you suggest. Since lithium is a light atom, it's internal forces are weak. I think it is the larger ion even though it is the lighter element. I would need to see some data though to be sure. Can anyone help here?

Originally posted by Bruce Troeh
While Iowa State still has the glass facility it no longer offers glassblowing except through the Glasblowers Guild. I started in 1977 learnign (sp) about formulating batch and calculating COE with Dr. David Martin. At the time I remember one of the grad students had purchased a box of kugler color I dearly wanted to use. Our batch was 111 - 113 COE. I do remeber a few broken pieces.
I've got a box of kugler to play with now but I'm using 104 so it still sits. BTW. Bruce

But these were theoretical CALCULATIONS, yes? Did you ever MEASURE the COE back then? I.e., with a dilatometer?

This is definitely the basis of a lot of the confusion: the difference between calculated and measured COEs. The imagined compatibility between glasses of the same expansion in today's studio/marketplace is based on comparing MEASURED expansions.

And what I'm digging for is WHEN did the measured COE become the standard for compatibility in studio glass?

Originally posted by Durk Valkema
Klaus never indicated numbers for compatibility just an indication "Harte W" which you could read as an indication for viscosity because it indicates if it's a hard or soft glass. To bad Klaus passed away last year, but I will try to find out more about the origins.

Wow, very interesting! Thanks, Durk!

I assume "W" = weich (soft)

So Klaus Kugler was comparing - or categorizing - his colored glasses based on the differences in their viscosities (hart/weich - hard/soft), NOT on their COEs.

Did that information travel with the bars? Does anyone who used Kugler in the '70s or '80s remember these ratings?

[Tom Littleton]

Lani,
From Frank Woolley (data from a Melt at the Studio at Corning in 1997). Softening point 666 degrees centigrade(1231degrees F), Annealing point-492 degrees centigrade (918 degrees F), strain point-455 (851 degrees F), expansion (0-300 degrees) 94.9.

CAVEAT- None of these figures should be taken as gospel, a holy grail, something that everyone is going to get every time they melt. Variations in raw materials, frunaces, melting techniques and other factors will cause differences in the glass. Not to mention Frank told me that the dilatometer measurements were plus or minus one point in accuracy.

Have I put enough qualifiers in here yet?

Originally posted by Pete VanderLaan
*********************
As to all glasses that have 96 C.O.E. fitting Sp87, no I don't think that but I think it based on the thickness of the two glasses that are present. I make opals at 96 that saw and grind with SP87 just fine if they are under one inch thick. Go over that and they don't fit. Why? my hunch is annealing range, not point.

Pete, the Annealing Point of a glass is a very precisely defined indicator of the viscosity of a glass (and of course, it determines the RANGE). So we are agreeing: the compatibility problems you sometimes encounter when you match measured COEs relates to their VISCOSITY differences.

You can try to compensate for the incompatibility problems of COE-matched glasses by better annealing practice, but you’re still screwing with glasses that are basically badly matched because they haven’t taken viscosity into consideration at the start. And you can very likely get away with it in blowing (or torchwork), but it comes back to bite you in the butt when you get thicker or grind – which is what lots of people are doing currently. At the same time that they’re kidding themselves that their matching COEs guarantee they’re working with compatible glass.

The annealing RANGE of SP87 is about 890 to 945 depending on thickness

Pete, I’m asking for the Annealing POINT. It’s a laboratory measure of viscosity that we can use to compare the relative viscosities of different glasses. It’s not an “about” number and it’s not a range (although I’d guess there might be slight variation depending on the lab used for the measurement).

For Bullseye, for instance, the AP of our clear is 990F, our white opal is 937F, our lead-based gold-pink transparent is 927F. Because of these annealing point differences, we have to adjust the expansion (to 91, 88 and 85 respectively) in order for the glasses to fuse together without visible stress in the daily compatibility tests. If we adjusted all those glasses to the same COE (believe me we’ve done this), they’d break apart.

Originally posted by Tom Littleton
Lani,
From Frank Woolley (data from a Melt at the Studio at Corning in 1997). Softening point 666 degrees centigrade(1231degrees F), Annealing point-492 degrees centigrade (918 degrees F), strain point-455 (851 degrees F), expansion (0-300 degrees) 94.9.

CAVEAT- None of these figures should be taken as gospel, a holy grail, something that everyone is going to get every time they melt. Variations in raw materials, frunaces, melting techniques and other factors will cause differences in the glass. Not to mention Frank told me that the dilatometer measurements were plus or minus one point in accuracy.

Have I put enough qualifiers in here yet?

Perfect! Thanks heaps, Tom! So what you've got - IF all the qualifiers are taken into consideration - is a glass with a 94.9 COE and an AP of 918F.

I'd wager that any colored glasses that fit that base clear (i.e. show little stress in a ring, pull, Trident seal or BE chip/bar test) but have a much lower AP (such as a lead glass would) will NOT have a 94-95 COE. Their measured LEC (0-300C) will be lower.

[Richard Huntrods]

Well, Tom - I'm a fan! I did the bending test when I first set up my annealer, and it resulted in me using my annealer at 915F. I give the glass a 1 hour soak and then ramp 100F/hr down to 50F. So far, so good as they say!

Gotta love that Spruce Pine!

-R

Originally posted by Tom Littleton
Lani,
From Frank Woolley (data from a Melt at the Studio at Corning in 1997). Softening point 666 degrees centigrade(1231degrees F), Annealing point-492 degrees centigrade (918 degrees F), strain point-455 (851 degrees F), expansion (0-300 degrees) 94.9.

CAVEAT- None of these figures should be taken as gospel, a holy grail, something that everyone is going to get every time they melt. Variations in raw materials, frunaces, melting techniques and other factors will cause differences in the glass. Not to mention Frank told me that the dilatometer measurements were plus or minus one point in accuracy.

Have I put enough qualifiers in here yet?

[Dave Bross]

Random bits from the Lithium chapter in Volf's "Chemical Approach to Glasses':

"It's low ionization energy is indicative of the strongly ionic nature of it's bonds. Lithium differs from the other alkalai metals in the properties of it's compounds and it's behavior in glass.

In its behavior in glass,Li is more similar to Mg than to Na or K. Lithium and magnesium have similar effective radii,and are functionally substitutable, in particular in glasses with low dielectric losses, and both decrease the liquidus temperature of basic silicate glasses in the range between 2 and 4% of the oxide."

(My note here: lowering the liquidus temp. lowers the annealing point that Lani is talking about up above, and in what I've been melting, way less than 2-4% lithium will drop the anneal temp.)

"Li2O is the lightest oxide component of glasses, being twice as light as Na2O. Replacement of Li2O for Na2O or K2o, or assesment of the effect of Li2O on the physical properties of glasses, should therefore be considered on a molar basis only. The effect of Lithium in glasses is first of all given by its small effective radius, which is the smallest among the monavelent metals. "

" The small size of the lithium ion also facilitates diffusion into the glass by substitution for Na and K both below and above the transformation interval. The process results in a permanent compressive tension in the surface, which has become the basis of chemical strengthening of glass products."

( my note here: You have to wonder if this wasn't in Nick Labino's mind when he suggested the lithium to Tom and family for tuning up the final version of Spruce Pine. Nick knew that studio artists would be attempting to combine all sorts of odd glasses and may have figured some " permanent compressive tension in the surface" would be a very good thing.)

"As a result of a strong attraction to oxygen, lithium tends to contract the free spaces in the silicate network. This contraction, for example, leads to a greater effect of Li on the density of glass than would be expected theoretically.

Owing to its small ionic radius the field strength of Li is the greatest among the alkalai metals. Li is therefore strongly bound in the silicate network, as indicated by its improving effect on the chemical durability with respect to water and acids in both glass and enamels. The properties that increase with increasing bond strength (such as viscosity) thus follow the sequence Li - Na - K, whereas the properties that increase with the loosening of the structure (such as thermal expansion) follow the opposite sequence K Na Li."

" Lithium raw materials generally improve the melting properties of glasses, which is a result of the low melting point of Lithium compounds, of the formation of eutectics and the mobility of the lithium ion."

" The lithium ion resembles in some respects a proton, which has the ability to penetrate deeply into the electron envelope of anions and neutral molecules. In this way, Fajans and Kreidl explain the outstanding efficiency of Li as a flux and its effect on decreasing the viscosity, in spite of it not being an element with a high deformability of its ion. The meltability is likewise contributed to by the great mobility of the small Li ion."

[Pete VanderLaan]

Durk: I understand what you are saying but my understanding of the story is somewhat different in that Klaus got tagged with an overhead photograaph showing clearly a cone emanating from the shop that literally showed death and destuction of vegetation that focused in his yard.

The thing that bothers me about what you say is the incredible variation in L.E.C. in the stick and my best understanding is that it persists to this day. I specifically refer to the soda lime based sticks of bright yellows and red opaques which the last time I measured them were actually about an 85. This would be a group of rod that virtually fits nothing and I have to question why it would be produced if it could actually be controlled. Since most of the lead flourides run around an 89 and the transparants a 94-94.5, what is the possible motivation in not at least standardizing the L.E.C.'s when it is not difficult to do. Lining up the viscosities is another and more substantial problem ( tip of the hat to Lani here)? So my assertion that they have been- at least in the past, stuck with a premixed leaded batch that they added colorants to and that explained the wandering L.E.C.'s I would assume that the lead fluoride batch is a different base formula but I don't know that and the soda lime base for the cadmium selenium glasses has to be a diferent formula. The formula for the lead arsenates could be a simple addition of Arsenic, but I doubt it.

But making glasses that fit nothing continues to baffle me.

Lani, I understand what you are saying and understand the point to which you refer but I do normally look at the range in which annealing can and does take place at varing speeds. We certainly also agree that thickness changes everything. I would note that the 918 falls exactly in the middle of what I suggested the range to be. Once we get out in the field, we find that the equipment is a hell of a lot looser than we would like to see it be while at the same time, the opinions about what is wrong today are wide, varied and unsubstantiated. Even while you refer to the annealing point as 918, I am almost positive that Tom has referrenced 895 as an annealing point at some time in the not too distant past. I assume that would be the lower point.

With my own work on the cut glass pieces- and you saw one when I was up in Portland in 2003, those pieces cannot go into a lehr at 940, let alone 918. I need to put them in at 995 for about an hour and then to lower them slowly to 945 or they simply crack when I do the knife edged grinding that they require.. That says to me that there is something going on in annealing land which I cannot account for except empirically. I trust my controllers and thermocouples.

Dilatometry was also approached back in the '80's by Steve Maslach as well who did a GAS presentation on it as well. I did not ever see the presentation but I did read about it and Steve was really jumping thru hoops to make sure the sample didn't drag on the side of the sample tube. When I asked him about the presentation, he didn't want to talk about it. Now, I have never had any of those problems. I do not think the dilatometer is a holy grail at all. still really only trust the ring test as a final arbiter but the dilatometer gets me right into the ballpark and tells me if I am high or low. If the ring test, or the trident seal is off by much, the sample doesn't survive the test and the information is fairly useless. The trident seal is way more sensitive to self destructing than the ring is which is why I prefer the ring.

I think that Steve O'Day is correct about Theresa at C R LOO being the first to make comparison charts which told everyone about the expansion factor of each glass. I think that really came out as the number of players in the sales field expanded exponentially in the late '80's and a lot of people were just randomly throwing colors together and then happily blaming the vendors when things cracked ( what else is new?).

To simplify for me I have to say that the rules for fit are simply different in blown, cast and slumped ware. People do such bizarre things with the product that creating a homogenous rule of thumb leads me down the path of madness.

You mention AP, great but at Log ç 13.4 ?
At Leerdam the batch calc program spitted out Log ç 2 down to 7 and Log ç 13.4 as AP
Philips mentions Log ç 12.4 Pa.s as AP and 13.5 as SP. The area in between is regarded annealing range in their info.
GLASMA through Glafo talks about transformation temperature at Log ç 13 in Pa.s
So what's the consensus here?

This is a GREAT thread. THANK YOU ALL. Doug Terry

Originally posted by Durk Valkema
You mention AP, great but at Log ç 13.4 ?
At Leerdam the batch calc program spitted out Log ç 2 down to 7 and Log ç 13.4 as AP
Philips mentions Log ç 12.4 Pa.s as AP and 13.5 as SP. The area in between is regarded annealing range in their info.
GLASMA through Glafo talks about transformation temperature at Log ç 13 in Pa.s
So what's the consensus here?

Durk, I'm not sure whether your question is directed at me, Pete or Tom, but the Bullseye annealing points that I provided earlier are at log 13.0. These are measured (not calculated) according to the ASTM test C336.71: "Standard Test Method for Annealing Point and Strain Point of Glass by Fiber Elongation".

We believe that this is Woolley's basis also, but you'd need to ask Tom, or check Frank's book (anyone?) where we think this is published. It's pretty standard in the US.

[Pete VanderLaan]

Originally posted by Lani McGregor


I'd wager that any colored glasses that fit that base clear (i.e. show little stress in a ring, pull, Trident seal or BE chip/bar test) but have a much lower AP (such as a lead glass would) will NOT have a 94-95 COE. Their measured LEC (0-300C) will be lower.

**********************
The spreadsheet I run brings SP87 in at a 94.6 and measures with dilatometry as a 95.9. When I use that spreadsheet on fluorine glasses at 6-7 percent, I bring in a calculated expansion of 72.3 and they almost always fit the first time. I do not know the assumptions in the program being made about the fluorine that makes this so low. They also measure 95.9-96.0. It is also the case that a pulled thread of those glasses has a serious arc. I am no where near the program right now so I cannot say about the annealing Log.

I do know that opal lead fluorides from kugler have ( or at least had the last time I tested) an expansion coefficient of about 89 . They however do not fit SP87 at all. It is why Tom created SP83 which I always thought had a measured L.E.C. of about 90.5 That would be within the 1.5 ten thousandths tolerance that has been accepted in the past as the maximum outside tolerance for expansion coefficient fit. .

I am confused though about the viscosity issue. Any Kugler rod will really run on the surface of a piece made with most commercial clear glasses indicating a lower viscosity than the host clear. This could be , either cullets or batches. One of the main reasons people really liked my black as an example was firstly the density of the color but also the fact that it does not bleed at all and stays absolutely where you put it. . It measures a 96.0 but has a close viscosity match to SP87 and still contains a substantial amount of lead.. It presents no strain in a ring test at all. I find bringing all of the gut features of a glass body to be very contradictory especially when one is trying to match up radically different formulations. I am thinking out loud here and will probably get an earful from someone.

So where does The Littleton Point come in? I was reading the interview from the oral history for Harvey K. Littleton and it mentions his father coming up with it but doesn't really date it. http://www.aaa.si.edu/oralhist/little01.htm

MR. LITTLETON: And so it was his job to get rid of the dark of the moon as a factor in making glass, among many other things. He defined annealing, the softening point of glass, which became an international standard.

MS. BYRD: The Littleton Point.

MR. LITTLETON: The Littleton Point, one way of determining what is a glass and differentiating one glass from another. There are literally an infinite number of glass compositions possible. And so, it's difficult to say precisely what one glass is without some kind of comparison to another glass.

A quick google tells me that it measured the softening point to compare glasses but I was just wondering how that fit in the whole picture. sounds like viscosity being measured to me but what do I know? This has been a cool thread.
Thanks!
Kevin Bethea

[Pete VanderLaan]

I had no idea that Harvey started Paoli Clay. Thanks for posting that. It takes quite a long time to read but it's really engrossing.

[Dave Bross]

Yes indeed! A must read. I love his take on art.

Now, from Scholes "Modern Glass Practice":

"The softening point of a glass, defined by Littleton as the temperature at which a filament of glass of specified diameter and length, heated at a given rate, elongates under its own weight one millimeter per minute, is a definite physical quality. According to Lillie, the viscosity at the Littleton softening point is 4.5 x 10 to the seventh power poises, or log n = 7.65.

The Littleton method is illustrated on this page. The small electric furnace contains a vertical iron core, four inches in length and one inch in diameter, with a 1/4 inch hole drilled through it lengthwise and another small hole drilled two inches deep paralell to the central hole. This permits the insertion of a thermo element at a point opposite the center of the rod to be tested. The specimen is a round filament of glass, .50 to .75 mm diameter, varying not more than .05 throughout its length of 22.8 cm. This is suspended in the central hole in the core of the furnace with its projecting lower end opposite a scale ( my note: "scale" meaning a ruler), against which the length of the freely suspended filament can be observed. Length in mm is plotted against time in minutes in the same units, and when the slope of the resulting curve reaches 45 degrees the temperature at the observed time is accepted as the softening point. The softening point of a glass runs about 200 degrees higher than its annealing temperature. The range for the commercial glasses is from 500 to 800 Centigrade (900 to 1500 farenheit) ."

For those suffering from “log” jam (like me), Morey mentions that Littleton initially chose the term “softening point” to describe the viscosity at which tubing bent in the flame of a lampworker’s torch.

Dave, great info on lithium. Another book I'll have to find I guess (Volk). Interesting that the small ion actually makes the glass more dense and more fluxed. Good to have an actual reference rather than what I think I remember someone saying one time. ("Don't trust half of what you think!" and maybe less of what you remember!)
As a side, I always liked my batch formulas with a percentage of magnesium from dolomite better than straight calcium. Interesting to see this compared to lithium. Magnesium containing batches seemed more even in their softening than with calcium only, which makes glass harder on the cold end and softer on the hot end. Some high calcium glasses have seemed rather sudden in softening when heating.
Nick was a smart man. It seems he included lots of non-essential but very desirable components in the SP batch. Zinc for colorant solubility, lithium for durability, density, and ease of melting, a small amount of fluorine which helps with the phosphate opals.

My question at the start of this thread was “When/where did COE come to be synonymous with compatibility in the studio glass scene?”

I could have spared everyone the pain (ok, so some of you like pain) of English & Turner, Winkelman & Schott etc if I’d written MEASURED, not calculated, COE – since it is the measured COE that is nowadays claimed and generally understood to equate to the “fit” of different glasses. And – no arguments here? – we arrive at this COE (0-300C, or 25-300C, or 17-300C!) measurement via a laboratory test using a dilatometer.

Where do we find this – that the measured COE equates to compatibility - in the technical literature? Is there any mention in Scholes, Morey, Volf? If so, can someone direct me to it? We can’t find it.

Was matching COEs to determine compatibility EVER the practice of glass factories in the US or Europe? Not to our knowledge. We, and everyone else we know (thanks to Durk & David Hopper for input on the Europeans; thank you, Tom, for the notes on SPB history), have been doing compatibility testing based on thread tests, ring tests, Trident seal tests, bar tests, and chip/bar tests. None of those tests are measurements of COE or LEC.

Where the COE = compatibility thing IS found is in the book that Dan & Boyce wrote in 1983, while they still thought that matching COEs – or being fairly close – was critical to compatibility. They had reservations at that time and expressed them with disclaimers like “Coefficient of expansion numbers should be regarded as starting points from which to experiment rather than definitive numbers not to be questioned” and “…some glasses that have the same coefficient number, as determined by a laboratory, do not always fit each other when fused together.” (pg. 42, Glass Fusing Book One, 1983)

But in spite of those disclaimers, our industry (colored glass dealers and other manufacturers, and eventually users) grabbed this easy “code” and started using it to define compatibility. Terms like “Tested Compatible to 90.0” (even though the experts will tell you that you can’t even measure to a .0 of accuracy in a dilatometric test) and “90 COE glass” came to be accepted to mean “this glass will fit other glasses of this same COE”.

When Bullseye published its actual COEs that showed a COE (0-300C) range of 88 for our white and 91 for our clear, it was interpreted to mean that we allowed a mismatch of plus or minus 1.5 COE points. We never said that (but this also seems to have come down in the lore - note Pete's comment earlier).

When we declined to guarantee that our glasses would fit other glasses with a COE of “90” we were accused of trying to corner the market and refusing to be “cooperative” with other manufacturers.

The ball was rolling. Dealers were looking for easy ways to inventory, categorize and sell. (I remember the work that Therese did at C & R Loo to try to categorize compatible glasses. It was a supremely noble effort. But I can guarantee that she didn’t have COE tests run on those glasses. At $150 per sample, does anyone really think that Mr Loo let her spend that kind of money to find out what the COE of K61 was, much less the entire Kugler line? I suspect she called around and ASKED. If she then put COE numbers on the answers, it was pure speculation and trust in her sources.)

(apparently I’ve exceeded the ability of the blather-o-meter to track my rant at this point, so this is continued in the next post…)

I originally asked this question (started in the prior post) because I honestly don’t know for sure whether Glass Fusing Book One is the original source of this bad science. But from what I’m reading on this board, this (early 1980s in the fusing sector) is where the measured COE as a synonym for compatibility first showed up in studio glass.

I would LOVE for someone to find evidence to the contrary, because we’d really like to blame all this crap on someone else. I can’t find it.

Now it’s everywhere. It’s accepted as “biblical” even though it runs counter to users’ own experience.

This is what amazes me here in the Kingdom of Blowing: I can understand why kilnworkers have a hard time understanding viscosity and how it is an equally important factor in compatibility. We put stuff in a box, turn on the heat, close the door and usually don’t have a clue what happens until we open the door and it’s broken or not. Pete, why do you say that blowers’ eyes “glaze over” at the mention of viscosity? Who the hell can’t feel/see the difference between a hard and a soft glass? Why would any blower think that the most OBVIOUS characteristic of a hot glass would not be at least as important as the characteristic he can’t even experience (the expansion between 0-300C)???

Contrary to what we’ve been accused of (thank god you’ve banned David from this board or I’d never have gotten this far), Bullseye is not out to BASH the COE. Jeeezus, how sick does that make us? What next? Do we go mug a Tg? We’re only arguing that divorcing it from viscosity seriously hinders the understanding of what makes glasses fit.

But I never intended this to be an “argument” anyway. I was just hunting for evidence that Bullseye didn’t start this COE = Compatibility mess in the first place. So far, I haven’t found any.


--------

[Steven O'Day]

Lani- I am pretty sure that Therese had the base glasses tested, but agree that they probably used the manufacturer's numbers for the color.

I was at Bendheim at the time and the two main problems were bad quality color and compatability problems. It seemed like some bars had more stones than glass.

Some customers were screaming for guidance on compatability. We had a lot of personal experience and anectodal help but no empirical proof. And so the COE number. I don't think that anyone ever represented it at the time as the end all, but a starting point. We always recomended testing with the actual base glass being used.

We were familiar with Boyce and Dan's book and had been around workshops that Boyce did in the early 80's. This was one source, I was also looking at Scholes. I believe Huchthausen had a lecture on it at Appalachian Craft and I remeber that lecture that Maslach gave. I don't remeber a eureka moment but rather a whole lot of sources saying the same thing.

Did that information travel with the bars? Does anyone who used Kugler in the '70s or '80s remember these ratings?

Yes, Kugler had the words Harte W on the labels but no numbers. They also seemed to be color coded (the labels) but I assumed it had to do with the color of the bar. And the glue was water soluable so it was easy to get the label off!

Originally posted by Steven O'Day
Lani- I am pretty sure that Therese had the base glasses tested, ...

But what were the "base glasses"? And I'd still be shocked if Mr Loo OK'd dilatometric lab tests, considering the expense.

Some customers were screaming for guidance on compatability.

I remember the pain.

We had a lot of personal experience and anectodal help but no empirical proof. And so the COE number. I don't think that anyone ever represented it at the time as the end all, but a starting point. We always recomended testing with the actual base glass being used.

I believe you. Nevertheless, the information has morphed over the years from "if it has the same COE it's (probably) compatible" to "if it's compatible it must have the same COE". Hence, people in Germany are fusing Bullseye (that famous "90" glass) to Desag Artista and because most of it isn't cracking wildly, dealers are listing Desag as a "90" glass - contrary to what the manufacture lists as their measured COE (which Desag lists as 94)

We were familiar with Boyce and Dan's book and had been around workshops that Boyce did in the early 80's. This was one source, I was also looking at Scholes.

Did you ever find any reference in Scholes that equates a measured COE with compatibility?

I believe Huchthausen had a lecture on it at Appalachian Craft and I remeber that lecture that Maslach gave. I don't remeber a eureka moment but rather a whole lot of sources saying the same thing.

But what exactly was that "same thing"?

Thanks for the input. It's great to hear from someone who was at Bendheim back then. I sure wish I'd held on to catalogs! I just KNEW that cleaning house we fundamentally WRONG .

[Peter Bowles]

[quote]Originally posted by Lani McGregor
[b]My question at the start of this thread was “When/where did COE come to be synonymous with compatibility in the studio glass scene?”



Its probably more to do with the technology that the potters and ceramicists brought with them when they moved over to working with glass than something that has been promulgated by the glass industries.

Certainly from a UK perspective, many of the key educators in glass have come from a background in ceramics, and with an understanding of 'fit' and 'compatibility' between ceramic bodies and glazes as an integral part of their knowledge and practice. Ceramic glazes are almost always made to fit by shifting the COE- a little bit more or less Silica will solve most glaze fit problems. I reckon some of this same thinking has just been transposed onto the 'new' materials of glass and taught into the system.