Totally naive about a possible process.

[Wendell Brown]

Let's see if I can ask this intelligently. I am brain-storming about making telescope mirror blanks. The process I want to do involves slumping glass to a mold plus providing a curve to the mirror face to save grinding time. I guess my question becomes can one do both a slump and a press fit process of a convex mold to the top of the glass to preform a concave glass surface?
Thinking (careful here) the slumping would be easy but arraigning the top surface of the glass to take on a shape (paraboloid if possible, circular otherwise) would involve a mold that as the glass slumps that it presses down on the glass (mold would halt the depressing at a point) and form the concave glass top surface. Overflow troughs could accommodate excess material.
Anyone ever seen something like done before?

Thanks,
Wendell Brown

"The Impossible Just Takes a Little Longer"

[Tony Smith]

Not sure what you mean by "press fit", but you can certainly slump a flat parallel borosilicate blank over a refractory mold to get a meniscus close to near net shape. It saves a lot of grinding and is how many large telescope mirrors are made. The problem arises in supporting the finished mirror. If you have a diameter/thickness ratio of 6/1 or less, the shape will be self supporting. Greater than that and you'll have to come up with some variant of a whiffletree mount to support the mirror and keep it from distorting throughout its range of elevation angles.

Tony

[Don Burt]

Variants of the Whiffletree mount. Some stuff you have to go with just because the name is cool.

[Wendell Brown]

Thanks for the reply Tony.
This process would involve having the glass conform to the mold for the bottom and top shapes. http://www.mdpub.com/scopeworks/hexagon ... exback.jpg shows what the glass would slump to the mold. Now on top it would have a mold to shape the concave sagitta. The slumping is a well understood process it seems. What I am researching is the process to form the top surface. So while the glass is being slumped is it possible to apply a mold that would conform the top shape onto the glass below? I envision a weighted mold that as the glass softened would impress itself into the glass. This top mold would function as a "stamping" process stamping the sagitta into the softened glass.
So at the end of the process, the honeycombed mirror would have the sagitta needed for focal length and saving grinding efforts.

Thanks to all,
Wendell Brown

[Marty]

The process sounds reasonable- honeycomb on the bottom, a collared mold with stops around the perimeter and a heavy convex press mold to sit on top and shape the top surface to concavity (with room for excess glass to escape).
This isn't slumping, it's casting and probably better controlled outside the kiln by pouring molten glass and then pressing the convex form on the surface. Graphite mold? Make sure those hexagons are tapered for glass release.
Would the time and expense involved in the mold save you enough in coldworking?
Don't forget to anneal the bejeezus out of it- I'd anneal for far more than double the overall thickness due to those hexes.

[Wendell Brown]

@Marty - All points understood. I was thinking of maybe using Castalot for your mentioned graphite mold. Envisioned having the cap (top) mold sit above the glass surface off the glass until the slumping process was complete, then lower the mold cap onto the soft glass (a twist guide onto alignment pins). Then take the temp down to the point the glass would hold its' shape and remove the top cap. Then anneal the glass slowly to room temp. The top concave shape just has to be close as the mirror grinding will remove any discrepancies.
Seems reasonable huh?
Another consideration will be trying to determine the volume of the glass in the mold. The top compression mold once pressed on the glass mass will cause the glass material to rise within the bottom mold. Minimal overflow of excess will avoid messy flash removal. Humm. So maybe some way to use volume displacement of the material to tell me volume?

Thanks for sharing your thoughts,
Wendell Brown

[Morganica]

I'm not familiar enough with the process for making telescope mirrors to know if this is a stupid comment or not, but...you can definitely use a multipart mold to achieve both shape and compression into that shape.

However, it may not be a completely CLEAN process, i.e., the casting may have some internal defects, and it will pick up texture from the mold that will need to be coldworked and polished. When I'm grinding and polishing a casting it can be difficult to do it perfectly evenly to preserve the original profile--depending on the precision you're expecting the pre-shape might not buy you much.

But again, I've never messed with mirrors and lenses much, so I don't know...

[Wendell Brown]

@Morganica - Yeah. It just needs to be close. The mirror grinding process can get better than 1/4 wavelength of light to the intended shape so no worries there. I have a CNC mill that can shape anything to form a mold mold. Just got to keep thinking. Nothing like generating a solution to keep me engaged. Using Alibre Design (CAD program) to design the electic kiln dimensions. Gotta know how many IFB I need.

[Morganica]

Ahhh. CNC would be the answer, certainly. And...hmmm. Are you by any chance the Wendell Brown who once published an article in my old magazine, BYTE? And has had more than a little to do with, oh, say, VoIP?

If so, that oughta be a mighty interesting kiln design...

[Bert Weiss]

Wendell, IFB are a very poor material for walls and roof of a kiln that fires glass. They are high in mass, and require money to heat up and wasted time to cool down. The temperature profile can be accomplished just fine with a controller. Floors are a different matter, due the their relative strength and non-warping characteristic.

The honeycomb design was created to add strength and shorten anneal time, relative to solid glass. However it is correct that you must take extra time to equalize the temperature of the thin and thick parts of the glass.

[Wendell Brown]

@Morganica - Sorry that I am not him but I did read Byte magazine back in the day. Seems the WBs in the world are numerous. I have a funny story about how I got married in my previous life but I digress.

@Bert - IFB are not the best choice then for kiln design? Seems most electric kilns use them and my design was meant to minimize the cubic volume needed. I am open to suggestions and incorporating a PID is not beyond my expertise since my degree is in electrical engineering with extensive embedded/Java programming.

[Marty]

I like brick walls and floors; fiber board and blanket roofs.

[Bert Weiss]

I like brick walls and floors; fiber board and blanket roofs.

I don't know a single advantage to a brick wall.

[Marty]

long slow anneals.

[Morganica]

cheaper
harder to shred or glop up IFB
easier to repair IFB if your glass explodes
contributes to structural integrity of the kiln
you can multitask the spare brick for lots of little glass-related chores

and, as Marty said, the insulation value can be an advantage in annealing some projects

I think overall a pure fiber kiln is more controllable (assuming you haven't done what most do and learned your own kiln so well that control isn't an issue), and it's more efficient when you need to spill heat rapidly. But it's also less forgiving of annealing errors. If I'd had one when I first started fusing (before I realized that my kiln was cooling more slowly than the actual schedule), I would have lost a lot more projects.

[Bert Weiss]

long slow anneals.

The higher mass of bricks simply cost you more to heat up and take longer to cool at the end. The purpose of your controller is to create long slow anneals. My fiber board lined kilns were built in 1982 and 1990, and are both in fine shape today.

Yes spare bricks are useful.

I can't remember the last time I had an annealing issue.

Bricks suck up heat so, during the cool to anneal cycle, they are busy giving off that heat as the power is off. This simply creates more uneven heating, as you get closer to the walls. I can see no good that could possibly come from this. It requires you to spend more time equalizing the glass temperature. I believe that more often than people realize, brick sided kilns make annealing impossible if you soak too long. Once you have introduced a temperature gradient in to the glass because the edges are too hot, it is very difficult to lose, especially sitting close to hot walls. This is avoidable by using just the right amount of time, but often we don't really know what that is.

I was taught about all this by an industrial heating engineer. His recommendation is to only use fiber insulation and only use solid state relays. The relays I use are on or off 120 times a second. Today's controller capabilities allow this fast control even with simple inexpensive SSR relays. The reason you don't see these relays being used in the manufactured kilns is that they can not function on the side of a hot kiln, and manufacturers are too cheap and lazy to move them 18" away. The other catch is that the Bartlett and Orton controllers most fusers use are so old fashioned, that they do not have the capability to cycle fast (but they can be used with solid state relays, if mounted away from the heat).

You always have to learn to program your own kiln. What I know by reading this board is that my kilns work better than the store bought ones. I heat faster without breaking and I have no anneal issues.

[Brock]

Apples versus oranges, yet again. Monolithic float versus multiple piece art glass. There is no definitive answer as anyone could deduce by the number of types of kilns successfully fusing glass. Repetition does not make one right . . .

[Bert Weiss]

Brick walls on a kiln are nothing but a hinderance to annealing. This is true regardless of how many companies make brick walled kilns. It may not be readily intuitive, but it is true. And yes if you have a fiber walled kiln, you have to be aware of how to program it, that may be different from other kilns.

[Bert Weiss]

The goal of annealing is to get 100% of the glass within 5ºC, inside the annealing range. Having a brick floor creates a relatively even heat sink. Having brick walls creates an uneven heat sink, especially as you get closer to those walls. The longer you hold glass in an uneven temperature gradient, the more the temperature of the glass will resemble that gradient. Annealing (stress relief) can not take place if the temperature across the entire piece of glass is more than 5º C difference, while in the annealing range.

[Brock]

You said,"I don't know a single advantage to a brick wall. Well . . . several of us do, long slow anneals. That heat that has been put into those brick walls is released slowly and assists in the annealing cycle. I tried to give you a hint, "Monolithic float versus multiple piece art glass" but you never get them . . .

For the particular purpose described, a thick casting, I would go with a brick kiln.