debugging a silver cast?

All,
I've made a couple of attempts doing some silver casting, and I would appreciate some advice in debugging my process.
I am using a Kerr Electro-melt to melt silver. I am using molds carved from Tufa (a compressed volcanic ash) for which I hope the silver to flow into.
Attempt #1: Flat relief cutting of the tufa. I melted the silver and poured it onto the flat. The silver immediately cooled into balls and did not take an impressions.
Attempt #2: Read that the mold needed to be of a temperature similar to the metal. So what I did was to try to do the casting INSIDE the Electro-melt crucible. I lathed a pair of tufa cylinders, 1 inch long and 1 1/2 inches long. I carved a funnel into the 1 inch long cylinder. I split the 1 1/2 inch long cylinder lengthwise. On one side I carved a pattern complete with sprue holes and air holes. The other side just simply had notches reamed for the sprue entrance and silver exit.
I aligned the cylinders inside the crucible, placed clean scrap silver and a pinch of borax on top of funnel, and ran the Electromelt at 1950 degrees at 10 minutes before cooling it down. I observed that the silver had fallen into the funnel as I had hoped.
After cool-down I found that the silver had completely split the cylinder and that while the flux had entered the pattern, the silver simply wasn't in the pattern. All of the silver that had gone through the funnel had puddled at the bottom of the split cylinder. Further, all the tufa had turned into a glassy kind of stone.
Synopsis: Attempt #2 demonstrated that while simple gravity would cause fluxed silver to move through the funnel, there are other factors requiring correction.
1) Simple compression of the split cylinder inside the tufa was not tight enough, because as the silver expanded the halves of the mold were forced apart. Wrap the halves with iron wire next time?
2) Tufa will vitrify. Maybe fire the mold halves and the cylinder first inside the crucible, let cool, and then melt the silver with pre-fired tufa?
My wife tells me: "Back to the drawing board." I think it is a simple matter of debugging, but I wanted to pick the brains of the community for ideas. Also, I don't have much money, so right now I am trying to bootstrap a process with what I have.
Suggestions?
Thanks in advance.
The Eternal Squire
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snip---

Could be that the use of that particular stuff is beyond the ability to act as a mold for the temperature involved. You might explore the art of proper investment casting. It's a tried and proven way to go-------
If not, it's not necessary to have the mold at the same temperature as the molten metal---just preheat it to around 1,200 degrees or so-----so it is slower at absorbing the heat from the molten metal.
Also, don't get any flux involved in the casting process. Flux the melting vessel well enough for the silver to slide easily, but any excess is going to end up in your casting, time and again.
Harold
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
message

Don't think you will get what you want with your process. I suggest you try centrifugal casting or vacumm casting instead. The surface tension of the silver will always be too high to allow it to flow out into a mold and leave details. If you have to pour, I suggest you cut a small sprue all the way through the mold and provide a slight vac on the bottom. Should not take much but be sure you add a seperator downstream of your mold to catch and silver that may pass through the mold. If you were in my area, I'd loan you my wind up centrifugal caster that I have not used in years
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Do not heat the tufa. It is meant to be poured at room temp.
Attempt #1. Molten silver has too high a surface tension to pour into an open mold like that. In an open iron ingot mold heated to 400, the silver will still look like a loaf of white bread.
Attempt #2. You tried to solve too many problems in one shot.
Take two pieces of tufa and rub them together to get a good seal.
Carve a mold and sprues into one side of the tufa. Make the sprues bigger and longer than you would guess. You are using hydrostatic pressure to fill the mold.
Cut several vents from the cavity to the edge of the mold. These should be only big enough to let air out but not metal.
Carve your funnel on the piece with the mold. This will save on tufa.
Bind the two pieces of tufa together with binding wire. Let the uncarved piece stick up over the funnel opening like the backboard on a basketball hoop. It will serve the same purpose.
Stand them up in a pan of sand and arrange it so it will be comfortable to pour.
Pour as in attempt #1 and get back to us with the results.
Paul K. Dickman
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.