CraftWEB Hot Glass Talk

CraftWEB Hot Glass Talk (
-   Color work from rods and batch glasses (
-   -   trident seal (

Bradley Howes 03-19-2019 11:12 AM

trident seal
Hello All,

Now that I'm getting deeper into making color, I'm getting asked if I've done any compatibility tests.
I've heard trident seals are one of the best ways to test compatibility. However, I don't know how to prepare this test. Can any of you share this with me?
Thank you.

Pete VanderLaan 03-19-2019 12:43 PM

I can't recall if the paper Hagy wrote surfaced in Glassnotes IV, a book you should have. If not, it can probably be gotten from The Rakow Library at Corning as a photocopy.

I'm mildly surprised that you aren't doing some form of testing since one comes up against the issue so rapidly.

Being brief, the test involves making canes of your two glasses in question at about 4mm diameter and about 1.5 inches long. The glass in question should be placed with the two other host glasses laid along side of them. Normally, the outside glass would be your clear you regularly use but it could be something else. Assume it's your clear.
The ends of the canes in question need to be fused together with a small torch. Only fuse the ends. Then anneal the piece. Once annealed and cooled, the sample needs to be placed in a polarimeter that is preferably graded by degrees totalling 360. Starting at the point of least retardation between the filters in question, the cane is observed as the dial is rotated until the retardation is at it's most intense. Reading the degree increments, one can infer the variances in linear expansion between the glasses in question.
Gaffer uses the trident deal for its testing and John is quite keen on it. I have issues with it because the tool I just described costs over $2500 dollars these days and if the glasses in question are not reasonably close in expansion initially, the seal won't survive the annealer and you learn nothing. Alfred really should own a strainoptics polarimeter and I'd be surprised if they didn't.

So I use a ring test even though I do own the strainoptics tool and use it. A ring test involves gathering the color in question and casing it in your clear. It
is nice to get reasonably even amounts of both glasses in question but it isn't critical in what you will learn.

Blowing out that cup as a cylinder and then annealing it, take your sample cup and slice a ring off on a diamond saw about 3/4 inch wide. If the sample survives this test, you already have a clear indicator that it is likely compatible. In incompatible tests, the cup just breaks right away. So, if you have this ring, then take it and make a vertical score across the width with an ordinary glass cutter. Tap that score open with the tool just like you were cutting sheet glass. The opening will go in one of two directions. Either the ring will spring open somewhat around 1/8th inch or less, or the ring will try to close on itself. Those two events are telling you whether your interior color has a higher or a lower expansion than the host clear. If you try this test with most German color rods, they are not likely to fit well and many just totally fail. Gaffer colors tend to do better against a true 96 expansion glass. If the ring pops open, the glass on the inside has a lower expansion . If it closes the gap, the expansion is higher. In batch glasses, adjust with either silica or sodium. You can't really change cullets.

The last reliable test is not a comparison test. It utilizes a tool where a rod of glass 4.000 inch is placed inside a quartz tube with a quartz push rod that in turn pushes on a dial indicator measuring in ten thousandths. The sample is heated from 19C to 300C and the expansion is shown on the dial indicator. This is a touchy feely test and I sometimes may run fresh samples three times to get a sense of the glass. Building a dilatometer is not cheap. Buying one is way worse.
The final simplest and most completely unreliable test is a simply cane pull. On a punty take a small gather of your clear and wed to it a similar size piece of your color in question. Heat it and try to pull it absolutely straight holding it in tension as it cools. Then taking an 18 inch sample of that rod, lay it across a straight edge. If the arch is more than 1/2 inch, it's going to give you trouble. Your work may hold together up until you grind it which changes the surface tension and the piece will break. Don't trust pull tests.

In all of these tests, you are really looking at indicators. There's no one thing that tells the tale but in all cases, the indicators should be pointing you in the right direction.

If you are going to make color, you simply have to test.

John Croucher 03-20-2019 12:36 AM

Bradley- the original source of Hagy's paper is to be found at Hagy, H.E.
Journal American Ceramic Society Vol 62, [1-2] 60-62 (1979)
The title of the paper is "The Trident Seal-A Rapid and Accurate Expansion Differential Test"
The retardation mismatch is read through the central cane which is your master clear. The outer canes of the seal will be the color you want to compare.
Alfred will have back copies of this journal and the glass engineering dept should have an optical bench with a polarimeter.

Pete VanderLaan 03-20-2019 09:00 AM

I should have been clearer. Most of the time I compare colors to colors. It helps me see the effects of specific changes in the oxides used. John is quite correct about the placement of the clear if that's what you're doing.

Where I have my greatest difficulties is comparing two colors in a thick piece of work where there's no clear at all. A lot of why I prefer the ring test.

All times are GMT -5. The time now is 11:24 AM.

Powered by vBulletin® Version 3.7.2
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd. Opportunity Network. 2008. All Rights Reserved.