Hi Andrew,
I don't see how oxygen absorbed at the top of a button after
casting will affect a part way down inside the flask, particularly
if, as you say, the metal has solidified by then. Flux helps
protect the metal when it's being melted, but in this case, what's
probably happening is an insulation effect from the added material
piled on top. I usually cover castings with a piece of soft
firebrick immediately after casting; this helps keep the button
fluid and enhances its usefulness as a reservoir of semi-liquid
metal to feed the part.
Because oxygen atoms are much smaller than metal atoms they are very
mobile in the metal crystal lattice at high temperatures. Silver
especially has an affinity for oxygen at elevated temperatures and
can hold great volumes of it at elevated temperatures, this is the
mecain.sim that creates firestain in sterling, the deep oxidation of
copper by the traveling oxygen atoms. The glass of the flux keeps
the oxygen out, not completely as it can still enter through the the
investment but at a much lower rate. I have done experiments with
this and can go into detail off line if you would like.
As for the insulation if we are talking about jewelry sized items
here and not pours of pounds of metal then the models are solid
before you have even finished pouring the mold and trying to keep
the button /pouring cup molten via insulating material is not buying
you anything.
Doesn't matter if there is a tree or not, most jewelry sized
models are solid before the button starts to fill
"Solid" is a relative term, then.
No solid is definitely not a relative term it is a precisely defined
temperature at which the metal undergoes a phase change from solid
to liquid phase.
The metal inside the flask, which is typically cast hot, is in the
process of solidification, but it's still permeated with channels
of liquid metal, like a wet sponge. This will either draw metal
from the button or sprue as it shrinks, or it will draw it from the
part. I've seen it happen too many times to swallow this "no
button" theory whole.
You don’t have to swallow it just prepare several identical flasks
( there are so many variables in casting as you no doubt know that
basing statements of fact on one or two casts is not advisable) and
calculate your metal weights so that half are poured without the
button and half are cast with a button. statistically you will see
no difference between the two sets. Major manufacturers like Rio and
Stuller are not using buttons in their castings now because they
just don’t need them. It is not hocus pocus or something I just made
up to create controversy it as I said something you can easily test
in your own shop.
I've done sculpture casting as well as jewelry, and I have to
say that I'm unimpressed by this analysis, despite the hi-tech
trappings. When "most" of the metal is solid is when shrinkage
porosity does its damage, pulling metal from the surface as
capillary channels collapse. The effects are more easily
appreciated in larger-scale castings, but once one learns to
recognize it, it can easily be seen in smaller ones as well. I've
also had the misfortune to have castings disrupted by mishaps in a
centrifugal casting machine, and have seen that after quite a few
milliseconds had gone by, the castings were still fluid enough to
be partially emptied after they were full. When I was using an
overloaded centrifuge that slowed down after its first few spins,
I've had the button drip out and empty itself, with predictable bad
results to the supposedly solidified castings that remained.
Yes in large scale castings the process of solidification is
different in that is is slower. It is very much a matter of scale.
if you are talking about belt buckles or sculpture rather than rings
and earrings it is all about heat transfer, how large a mass of
molten metal is in the mold cavity gate and sprue. I am not trying
to make any assertions about larger castings as they have different
dynamics but I will stand by my statements as far as jewelry ( ie
rings, earrings small cross section items ) casting goes.
One of the things that is starting to be obvious in the casting
research being done on jewelry casting is that the reduction of
turbulence is very important to good castings (maybe the most
important factor?) . This is backed up by the simulation that was
presented in that if the metal is slowed down by turbulent flow it
will not be able to feed the model cavity enough metal in a timely
manner to produce a sound dense casting due to this rapid
solidification.
I do agree that a reduction in turbulence can help, but I'll
still be casting with enough reserve metal to accomodate shrinkage.
Shrinkage is not a myth, it's generally accepted as fact in the
casting trade, and I'm not the only one who believes it - see
http://kerrlab.com/JewelryProducts/technical/Glossary.htm or
http://www.sternleach.com/technical/articles/silvcast.php under
"shrinkage porosity".
Shrinkage is a fact of life in castings no mater what scale. Metal
volume changes significantly during cooling from the molten state.
Best practice is to provide sprues and gates that will allow for
rapid filling of the mold cavity so that a maximum amount of metal
will be present when the casting solidifies. The point that I am
trying to make is that trying to feed a casting that has already
solidified with a button is a waste of time and money. And the
research and practice is showing that a button in this scale
casting is not required.
Papers about this are available in the published proceedings of the
Santa Fe Symposium, for the most part this work is done by
metallurgists working in the jewelry industry. If you are interested
I can provide titles of some of the relevant papers for you to read.
Regards,
Jim Binnion
James Binnion Metal Arts
Phone (360) 756-6550
Toll Free (877) 408 7287
Fax (360) 756-2160
@James_Binnion
Member of the Better Business Bureau