Adding glass during a slump ?

[John Barrett]

*** utter newb alert ***

This may seem a little strange here since it seems most of the posts are about art glass, but I need some experienced help with ideas that I want to try for making telescope mirrors. I've got projects in the works that involve both casting and slumping. The casting part seems simple enough, but there is one thing I want to try with slumping that has a high potential to cause huge problems, so I decided to ask before destroying my kiln

Some background for anyone who is interested in astronomy... I'm trying to make meniscus mirrors slumped over a parabolic mold with a foamed glass backing. Making the mold and slumping the front surface of the mirror should not be a problem, the issues I want to address are related to the foam glass backing. Making the foam glass is simple enough... 88% finely ground glass plus 12% carbon powder by weight, heat to approximately slumping temps, soak for 30-60 minutes, and cool with annealing as for normal glass.

Here is the problem: Since the solid front surface plate is on the bottom against the slump mold, it is going to heat up and slump last, which will overheat the foam glass material if I try to "stack" everything in the mold while cold. Also since the foam glass is getting heated first, it might try to work its way around the surface plate, causing problems on the front surface.

How much of as nightmare am I looking at if I try to add the powdered glass/carbon mixture to the mold AFTER the front plate has slumped.. while the kiln is still hot -- (1400-1600F)....

If that idea is just a recipe for disaster, I've got a backup plan.. doing the process in 2 heats.. but I would much prefer to do everything in one shot

[Laurie Spray]

John.......you are so far beyond my glass experience.....but....all i can say is that i do occasionally add glass (in frit not powder) to a piece that is fusing. The worse case scenerio is that as you are adding it some glass scatters and gets into the coils on you red hot kiln. That is an easy way to burn out an element. You may make it thru that firing but the glass will eat into the coil until it fails which could be one firing more or ten. Many times you can't even see it until you remove the dead coil. Unfortunately i know this from experience.
Other than that......i can't imagine what you are doing there!!!

[John Barrett]

TY so much for the reply, Laurie.. I was starting to get a little worried The idea may be beyond your experience, but I'm not.. I have yet to cast a single piece of glass... still working on the containment walls and base plate for my molds.. I couldn't get anything large enough in a one piece kiln shelf so I'm using a plate of fireplace insulation cut to size and cutting firebrick is a very slow process !!

It's good to hear that you can add glass in mid process, and your warning about glass in the coils is going to cause me to think real hard about particle size. The paper I found on making foam glass talked about particles on the order of 0.003" for the glass, and I can see that size being rather uncontrollable in terms of dusts getting into the air as the powder mix is being added to the mold. Just took a peek at a glass supply web site to get a better understanding of frit types and I may want to screen my crushed glass and just keep the larger particles, saving the powder for other uses that don't involve adding to a hot kiln.

I'm in the process of adding a downdraft vent to the kiln, and the problems of glass in the coils may cause me to turn off the bottom 1 or 2 elements and cover them with ceramic paper to protect those coils. That may not be a bad idea in any case since I've got a pottery kiln and all the heat is from the sides. Getting more of the heat from the top so the glass heats evenly may be worth the slower ramp up time.

I'm going to be adding the mix using a 1-2" inside diameter steel tube with a funnel on top bent and shaped so the funnel is not over the kiln in case there is a spill, so I shouldn't have that much problem with dust higher up in the kiln. Hopefully any that is created from the drop between the tube and the mold will be caught up by the down draft vent and quickly removed from the kiln.

Do keep it coming though. Any more comments or ideas are very much welcome from you or anyone at this point !!

[Kevin Midgley]

Throwing a combustible in a powder format into a hot anything!!!!!!???????
or is someone not familiar the dangers?????

[John Barrett]

Read the MSDS for carbon powder (Activated Charcoal)

Flammability rating > SLIGHT
Melting Point > 3550C 6422F

As with most organic solids, fire is possible at elevated temperatures or by contact with an ignition source.
Activated carbon is difficult to ignite and tends to burn slowly (smolder) without producing smoke or flame.
Wet activated carbon depletes oxygen from the air. Materials allowed to smolder for long periods in enclosed
spaces, may produce amounts of carbon monoxide which may reach the lower explosive limit for carbon
monoxide of 12.5% in air. Contact with strong oxidizers such as ozone or liquid oxygen may cause rapid
combustion.

The mix in question is 12% carbon 88% glass by weight, well mixed... the carbon is going to be very well buffered by the glass in this situation, so it is likely the glass will already be fusing before the carbon starts burning/giving off gas except what little carbon is close to the hot slumped plate... which is the intended result actually... I want glass with LOTS of tiny bubbles in it

As far as carbon monoxide is concerned, the kiln is going to be ventilated so the any carbon that is burning on a surface will have plenty of air, and any CO will be exhausted quickly since CO is heavier than air... I expect the bubbles to contain mostly CO since they are burning in an essentially airless environment, but that is fine because it is all contained in the molten glass.

I will of course be doing the initial attempts small scale.. no more than a few ounces of the glass/carbon mix against a preheated plate of glass, but given the burning characteristics I do not see a huge problem from the carbon. I was more worried about dumping a fairly large mass of cold glass on an already slumped plate of glass at slumping temps.

[Morganica]

"The casting part seems simple enough..."
That statement is kinda red flagging it for me; casting perfectly smooth, transparent glass in a precise compound curve may not be all that simple.

Wanna make sure I get what you're doing: You're using foamed glass (similar to the Corning Foamglas, maybe?) as an insulated backer/spacer for the meniscus mirror, presumably to ensure stability in weather extremes (?), and you're making a sandwich of foamed glass, then mold, then mirror? Are you putting the foamed glass in its own mold or will it sit flat on the kilnshelf/floor?

Are you planning to glue the backing to the mirror, post-firing?

I'm assuming that you'll be coldworking the mirror into its final shape? Or do you expect the cast/slumped form to be its final shape? How big is your mirror? Is it a solid thickness (or how much)?

Sorry for all the questions.

I'm thinking you could baffle the foamglass and slow down the heatwork...but I want to make sure I understand what you're doing first.

Thanks--

--cynthia

[Laurie Spray]

I am really unclear on what you are trying to accomplish but......could you not heat up the frit in a seconf kiln to around 800 where it is not hot enough to melt but hot enough to not cause shock when added? I have NO idea......just thinking out load.

[John Barrett]

OK -- the entire process... I'm using spin casting techniques on plaster to create a negative parabolic mold approximating the shape of the final mirror surface, to which I will add appropriate clay to create a positive slump mold, which is then fired to create a "bisque" (I think).. the result of this is a plate with a circular parabolic hump in the middle, and a 1-2" flat border to support a containment ring made from castable refractory... this assembly is placed in the kiln with a 1/4-1/2" thick disc of glass the correct diameter (purchased or cast in a separate step) balanced on the hump. The kiln is brought up to temp and when the plate is observed to have slumped to the mold, then the glass frit/carbon mixture will be added using the tube and funnel method described earlier, and generally smoothed into the mold with a stainless steel paddle. The kiln is then returned to temperature, and soaked for 30-60 minutes, then cooled in the typical annealing curve for the amount of glass being cooled... the result is a parabolic front surface with foam glass back ready for cold working to grind out any defects and polish the front surface to the exact parabolic shape in preparation for the application of the mirror coating.

@Morganica: the glass doesn't have to to be perfect when it comes out, it just needs to be close... nor does it need to be transparent.. foamed glass is FAR from being transparent !! I'm attempting to get the front surface plate and the foam glass created and fused in a single firing... to make it more clear. from bottom to top, parabolic hump mold, front surface plate, foam glass mix... so the final mirror is face down in the mold.

@Laurie: I don't have a second kiln or furnace for preheating... I would also need to be careful that I did not preheat to the point that the carbon begins burning... If this single firing idea doesn't work, then I'll have no choice but to slump and anneal the front surface, then add the foam glass mix either when the kiln is cold, or at least much cooler than the low end annealing temp, then run another firing cycle.

[Lauri Levanto]

I think ( with no further experience) that two phase firing is the fastest way to go.
First slump the front plate. 1/2" does not take too long to anneal and cool.
Then add the foam glass mixture for a second firing. That may be a little tricky.
You have a thin biscue dome under the sheet. It takes some time to heat the glass through it beyond the shock zone. Say up to 550 C. It is then safe to add the bubble mix, which I gues you want 2-4 inch thick depending the mirror size.
The disk may have cooled near to 475 C shock zone. The second heating must begin fairly slowly, the bubble mix is good thermal insulator. No doubt the top surface will overcook to quite bubbleless solid.

The annealing is critical. There you spend most of the process days. Forget the 2-4" tables. The bubbles inside still resist thermal flow out of the mass. Can you bury a sacrificial thermocouple in the bubble mix. You need to know that the thermal gradient does not exceed 5 C during annealing. That figure is for soda lime glass. Use of borosilicate could save essential amount of time. I do not know properties of borosilicate.

One final notion: You are sure your bubble mix and front blank are compatible glass.

Tha is for my 4/5" (=2 cents).
Disclaimer: I do not give any warranty, but wish good luck that goes with good design
-lauri

edited: To understand the thermal insulation capacity of bubble mix, refer to some glasswool insulation data sheet.

[John Barrett]

I don't know about faster given that some of the schedules I have seen for doing telescope mirrors take DAYS to run, which I can understand on larger mirrors where I might be casting as much as 20-40 lbs of glass, and in some cases MORE !!!

All the glass I'm using is recycled, and the best I will ever be able to say is the front plate will always be one type of glass, and the foam glass will all be one type, but no assurance that the front and foam will be the same type. If you think that will cause problems getting the foam fused to the front, then I need to know now because I really have no idea what specific formulations of glass I am getting... all I know is that it is clear float glass, usually from discarded table tops. Some may be slightly colored but I avoid anything with heavy coloring. A bit of explanation about what can go wrong mixing glass formulations would really help about now

I think I need a better explanation of shock zone... but in case I got it.. you are saying that between cold and 500, that putting cold on hot will cause problems, but that once the front plate is past 500 that adding the foam mix is O.K. ?? If that is the case then why don't you like the idea of adding the foam mix at 1500 ?? Its not like I'm worried about the foam glass shattering before it heats up

If I'm going to add a thermocouple, it'll be in the bisque just under the surface at the peak of the mold.. I can't afford sacrificial probes

OK -- lets talk schedule for a minute.. my starting point schedule looks something like:

5h to 950 soak 0h
1h to 1250 soak 1h
2h to 1750 soak 1h
FAST to 1050 soak 8h (kiln open, vent running, close for soak)
30h to 950 soak 0h
15h to 850 soak 0h
20h to 90 off

Just over 3 days total run time, and I better start the run early in the morning so I'm into the annealing before I go to bed This schedule is for a honeycomb backed mirror that is cast face up with plaster hexagons washed to the kiln shelf.. something of a different animal than what I'm doing, but at least its a starting point for the huge amounts of glass I'm casting compared to most art glass designs.

Based on what you just said, I'm probably going to stretch that 8h annealing soak to 12h, and add more time on the last 3 steps... maybe 35, 20, 25 to give the glass time to change temps more or less evenly. I'm also going to dump the peak temp down to around 1400-1500 since I'm not trying to get the glass in between a bunch of hex inserts spaced a fraction of an inch apart.. I just need whatever it takes to get the front plate to slump, and 1400-1500 is the right range for getting the foam glass to to gas up and fuse.

Still, with those changes I'm looking at a 4 day firing cycle, so you can imagine why I don't want to do it twice for a single mirror !! Even if the front plate were annealed and cooled before adding the foam mix, I still can't see it taking less than 2 days to slump and cool the front plate (slap me if I'm wrong here... I haven't looked into schedules for doing the front plate separately at all), which would push the total time up to 6 or 7 days... that is a LOT of time to run the kiln even if most of it is slow ramp down time which doesn't take that much power.

You mentioned that the foam glass is a good insulator, and its not going to be helping that the containment ring and "kiln shelf" are either castable refractory or fireplace insulating plate.. might as well think of both being very good insulating fire brick. I can get the insulation numbers from Corning for their foam glass products as a rough approximation.

And FINALLY... I'm not too worried about the top of the foam glass going solid on top, though it wouldn't bother me if it did. Since the glass is gonna spend at best an hour or two at fusing temps, that doesn't give much time for bubbles to migrate out, and I'm sure they will be replaced by bubbles from deeper in the mix.. I would actually expect the front plate to get thicker rather than the top of the foam glass glazing over solid.

[Lauri Levanto]

I don't know about faster given that some of the schedules I have seen for doing telescope mirrors take DAYS to run, which I can understand on larger mirrors where I might be casting as much as 20-40 lbs of glass, and in some cases MORE !!!

It is not the weight that counts but the distance the heat from core must travel to the surface.

If you think that will cause problems getting the foam fused to the front, then I need to know now because I really have no idea what specific formulations of glass I am getting... all I know is that it is clear float glass, usually from discarded table tops

There you have a huge risk for incompatibility. Both in the front plate and between the front and foam. Compatibility here refers to the thermal expansion rate. On my site I have some examples. Unfortunately the text is in Finnish.
http://lauri.lsd.dk/lasi/mater/compatib2.php
The first picture tells what happened when an apple of 3 mm float glass was fused on a base of 5 mm float. The different expansion/contraction caused a couple of cracks. Within a year the apple was completely separated from the base. Click the picture to see it large. For a telescope project the front piece MUST be of a single sheet.

Sorry it is late night here. Let us continue tomorrow.

[Morganica]

OK, I think I get it (or at least get more of it). John, have you seen successful examples of foamglas/mirror glass fusing? I've heard of it done, but the systems I've seen attach foamglas to mirror with a low-expansion adhesive instead of fusing them as a single unit.

I agree with Lauri. My concern would be less about excessive heatwork to the foamglas and more around differing expansion/contraction rates of the two materials, particularly if this is recycled glass. I think fusing them together is gonna cause problems. Glass contracts, sometimes dramatically, as it cools. If it doesn't contract at the same rate as its substrate/mold/anything else connected to it, it develops stress.

That stress can cause the glass to come apart in the kiln, sometimes violently, or cause the piece to come apart later, during grinding and polishing...or worse, cause the unit to crack later on if it encounters strong temperature changes.

I don't know what the implications are for the carbon powder in the mix--it might provide a measure of flexibility to help prevent stress introduction...or not. I suspect not. And having the (foamglas backing?) cast over plaster hexagons is another concern--if the foamglas doesn't flex, and/or the plaster contracts at different rates inside those cells...you've got another potential stress point.

As Lauri suggested, I think you're safest with dual firings...assuming you don't have compatibility issues. More heatwork in the first firing, for the mirror forming, then a second firing to cook and fuse the foamglas. And I'd agree that the ramp and anneal are going to be a bit tricky with the foamglas insulating one side of the mirror--if you can set up a witness block with a sacrifice thermocouple you'll probably have better luck.

[John Barrett]

@Lauri: Whenever is convenient for you... I monitor email all day and have this thread subscribed so I'm likely replying to any post within 10 minutes of you hitting submit

Incompatibility may be what kills me since I'm not buying glass of a known formula... I guess I'm going to have to find a way to test the expansion rates... Google FTW or find a way to blend the glass from the various sources and get a consistent "product"

[John Barrett]

OK -- measuring thermal expansion is a bit of a pain, but not impossible given the tools I have on hand.. so I can test the glass I get and make sure it is at least close to compatible... common techniques involve securing a sample say 1" x 4" in a metal tube with a digital indicator inserted at one end in contact with the sample and a temperature probe through the side of the tube in contact with the sample.. hot water or steam is then run through the tube, and the change in temperature and size measured, giving the CTE of the sample A bit of brass pipe and some fittings from the hardware store should handle that setup easily enough.

Blending glasses is probably too much trouble.. I'd have to grind the incompatible glasses up very fine and hope everything mixes well under heat.. too much hassle with glass powder given what I have read about bubble formation during firing... nothing like what foam glass is intended to create, but still requiring a long soak, possibly under negative pressure to get most or all of the bubbles out of a piece meant to be a mirror front surface.

Better yet will be to make my mirrors and foam out of a glass from a single source That means a bit more waste and keeping track of which table top a given plate or powder was made from ... I'll probably have to ignore smaller pieces of glass in my scrounging efforts.. a single table top should yield enough glass for 2 or 3 mirrors without risking mixing glasses from different sources. Only the larger mirrors will probably end up forcing me to test glass compatibility before casting... just depends on the size of table top I start from

While I don't think this conversation is 100% over, I would like to take a moment to thank everyone who has participated so far... I've learned a lot and have a much clearer idea of where to go from here

[Lauri Levanto]

You may have tools for measuring thermal expansion.
The point here is avoiding permanent stress. It can be seen with a pair of polarized filters. Search this forum about polarization.

I strongly recommend that you make a small test.
It can be a 4"square you process according your plans. See if it comes out
from kiln intact. If so make the polarization test. If Ok, then proceed.

I have not seen any reference to the size of your project. The only hint was that you considered 20 lbs as MUCH of glass.

A solid block 8" diameter and 70 mm thick weighs a bit over 5 kg.
For a 8" disk you need maybe only 1½ " thickness if solid.
This willl eliminate much of the hassle with foam glass.
1½" block is quite simple to fuse from 8 mm disks.
Does the foam backing have other merits than mere saving of weight.

-lauri

[Kevin Midgley]

I have no idea of its availability in the sizes you require but you might be better off using sheet borosilicate or quartz glass and a torch to bend it.
You'd have less issue with distortions due to temperature changes in the finished piece

[John Barrett]

The largest mirror I have done the numbers on so far is a 24" honeycomb backed mirror that takes 70lbs of glass, and I figure that's going to be about the limit for anything I'm going to try to do... between the glass, base plate, and mold elements, I'm going to be looking at 100lbs if not more that I'll have to get in and out of the kiln ... my poor aching back !!!

There are a couple of other people doing designs that want to use my kiln and they are talking about 20" at 40lbs

I'm setting up for a max mirror thickness of 4" for 16-24" mirrors, and 3" for anything smaller... doesn't matter if the mirror is done as honeycomb or foam backed.. that's the height of the containment rings I'm making for each diameter

@Lauri: yer driving me insane mixing english and metric measurements

@Kevin: half the reason I'm doing this is to avoid the $2000-3000usd price for a 24" borosilicate plate 3 or 4 inches thick !! The other half is to bypass some of the work required to grind out the parabola on a large sized mirror... On that 70lb mirror I mentioned, grinding out the shape means removing about 12lbs of glass and that takes a lot of effort making the grinding tools, and a crapload abrasives... cold working sucks !!

[Marian]

Forty pounds of glass and 20 inch diameter? Sounds like you should do some smaller tests of the plan first. Opening a kiln at 1600 degrees with pounds of glass is asking for burns, too, not to mention searing your arms and eyes.

[John Barrett]

I've put a lot of thought into the setup here... I've got a gantry system with an electric winch for getting things in and out of the kiln... welding gloves to protect from the heat... a wire rope and pully system for opening the cover without having to actually reach over the kiln, a full face plexiglass face mask, and a kiln vent to draw heated air down the kiln to where the work is happening and hopefully reduce some of the searing heat at the top when the kiln is open

Also.. many smaller tests are planned, but the final goal is what counts I'm still working on setting up the molds for the ring walls, only just got my glass cutter in for doing the big circles a few days ago, got the 26" round "kiln shelf" made from a solid sheet of fireplace insulation made up.. it is much thicker than your typical kiln shelf so it won't break under the weight of that much glass

I figure I got 2-3 more weeks of work just getting set up for the first full run fusing some of the thin plate glass I've got that I need to turn into "normal" mirror blanks... I need 2 x 12" discs 1.5" thick... that will give me a good check of my temperature program and annealing cycle. Then I can move on to some of the more complicated things I have planned... honeycomb backed mirrors, meniscus mirrors, and foam glass backed mirrors

[Rich Canby]

Wondering if you had any success?

I am doing similar stuff. Successfully fuzed multiple layers of glass to make a 24" circle 1 1/4" thick. I have been rough grinding it for a couple of weeks. Sagitta is at .090" but I need about .28". This is how deep the central depression is.

I have also unsuccessfully fuzed glass into a honeycomb backed shape. Looks like it broke during the annealing process. I went too fast.

I like the idea of making a glass front with a foam glass back. I am planning to do a test piece, but would love to hear about any watch-outs.
Rich