Alloy is splitting and cracking when I roll it

Was: Red spots in homemade gold alloy

I’ve a related question - I’ve alloyed gold before with success, but
currently I’m having a terrible time - the alloy is splitting and
cracking when I roll it. I’m a low budget operation - I melt in a
well made in a charcoal block. Still, this has worked in the past,
so I don’t understand why the 18K I’m making now is - amazingly -
black on the inside when I cut it open.

Any hints on how I can correct this?

try to use at least 50% new casting grain in the scrap… the problem
sounds like too much scrap not enough new gold, as for the black on
the inside- if you are pouring into a charcoal block mould, the pour
is probably bringing to the surface about half way through powdered
charcoal from not blowing out the mould first, (canned air) and
trapping it in the rest of the pour… that’s all i can figure… also
make sure all solder is removed from the scrap too as the zinc in it
can burn up leaving sometimes, black impurities that were not removed
from the molten “ball” while rolling in the crucible

Hello Laura,

Perhaps if you share what your alloy was there may be some
indication of what has gone wrong? It seems to me that you may have
mixed something in here that may not have been what you thought it
was - zinc instead of silver perhaps - or a low grade copper???

please explain the entire process you used to alloy?


try to use at least 50% new casting grain in the scrap.. 

I’ve being asking myself -for a long time now- why this 50 percent
of fine material is needed to be mixed in an older batch?

What is it doing do the alloy? Is there a sort of de-integration of
the metals used in a previous melt?

I have no doubt that something is going on in the structure of the
alloy, With all means, I don’t like to beheave like a monkey just
because of an unspoken rule says so.

It might be a stupid question but not asking an die stupid is not
smart eighter. I’m sure that the answer has to do with metalurgie
but still…I like to know. Anyone out there who has the time to
explain this or knows about a site where I can find the answer for
irritating question which keeps me busy all the time when iI alloy

Enjoy and have fun

Hoover and Strong’s catalogue has a good article on reclaiming
scrap- though consider they are in the business of selling casting
grain and raw materials, the article is worth the few seconds to read
on master alloys, prepared alloys and scrap refining with new grain
and your scrap gold.

You will always get better results if you add at least 25% new gold
(or whatever the metal is that you are working with) to the pour.
With gold hitting 1000+ an ounce though the least possible to give
you good results in my opinion is 33. 3% new 24Kt gold with whatever
you are alloying to get whatever karat or colour you desire. Adding
no new gold is not recommended, neither is just adding new master
alloy to “clean up” and deoxidize the scrap you are using if your
pour is all scrap, sweeps, etc.

It is critical though that you remove all excess, or remaining
solder from your reclamations; just a little can cause splitting or
incomplete recrystallization in the pour. If you are making, say, a
blue- green gold alloy you would mix together:

1 gram of 24 kt yellow
0.532 grams of .999 silver
0.058 grams of Copper
0.1120 grams of Nickel

then multiply the resultant weight (1.710 grams) in this receipt
times the quantity you want to end up with for example you want 1
ounce of blu-green gold, then you would multiply the metals by about
15-16 parts of each (the troy ounce result you want being more or
less equal to 31.103. or 1 troy oz. )

If you use all scrap for the 15 grams of fine gold your result would
potentially split, or more likely pit when you rolled it out. Always
put the mass of scrap in a plastic box and run a magnet through it
to pull out any bits of ferrous containing scrap- that alone helps
eliminate many problems, then if you wish, add any master alloys you
like according to the manufacturers instructions as to karat and
colour desired and remember to account for the addition in any
formula you are using.

But most importantly to avoid crumbling, splitting or otherwise
spotted (from not mixing thoroughly) and streaked pours is to use as
clean ingredients as possible, and pre-heat the mould before the
pour. Failure to do any of these steps can give you streaks,
crumbling during rolling, pits on the ingot and/or in sheet rolled
from the ingot and holes in the ingot.

The mould should be neither too hot nor too cold- you should be able
to touch the mold without getting a burn. The best indicator though
is to coast the inside of your mould with wax or oil, when it smokes
it is at pouring temperature, the excess simply lubricates the mold
for easy release. I hope this helps.


Firstly we need to know what carat you were using or trying to make
and where did the copper come from?Were you using pure silver, were
you adding any zinc?

I do not know if you can still get a purifying flux which contains a
large proportion of potassium nitrate and borax and other chemicals
which is added to the molten metal also use lots of borax powder
before and just before you pour.

Potassium nitrate is dangerous it will flare up on application. A
carbon crucible in an electronic melting furnace can help but the
metal must be left molten in the furnace for some time. Stirring the
gold with a carbon rod can help. or even a piece of dry wooden
dowel. Never use iron or steel or other metals to stir, and make sure
there is no iron, say springs from bolt rings or catches. Minute
percentages can ruin an alloy. Make sure scrap is clean and filings
absolutely clean, rake with a magnet, you will be surprised at what
can be in the filings, tiny bits of saw blades iron filings from

If you start off with pure gold and fine silver buy pure copper,
scrap electrical wire should be pure copper, or buy that alloy
specially prepared for what ever carat you want. You add the fine
gold. Do not use copper sheet or pipe they usually have zinc or
other impurities to harden them. Unless you know it is pure copper.
You may have to trade your bad gold in when you next sell your sweeps
and lemel. Try looking in the Ganoksin Archives for answers.

Best of luck

try to use at least 50% new casting grain in the scrap.. 

Actually, from my experience, when making an ingot for rolling, using
all new metal is preferred. Making sheet you might get away with old
and new metal, making wire is more difficult.

Using 50% new metal is the rule for casting, not for rolling as far
as I know. Using old metal and new metal for casting is much more
forgiving that using old metal for ingots for rolling in my

When casting an ingot for rolling, I have had much more success using
an ingot mold and making the ingot as if you were making wire. I have
far less problems with cracking this way.

Using old and new metal for rolling white gold has always resulted
in unusable product.

There actually are alloys specially made for rolling.

Richard Hart G.G.
Jewelers Gallery

Thanks, R.E. & Gwen.

This alloy comes from melting scrap - again I have done this before
with success. There must have been something in there I didn’t know
about, although I did try to look for stampings on everything I
melted. (OK, there was one thing…) I took this melted 14k gold
scrap and added 24k to make 18K.

I was taught to make gold in a well in a charcoal block - so there is
no pouring involved. I am going to try to mix well with a charcoal
stick, add a pinch of borax, see if this improve things. If not, my
only recourse is to send it out to be professionally refined,



I find using Jay Whaley’s WHIP and a crucible far easier to melt any
noble metals, whether new or scrap. There is much more control, and
easier to gently roll the liquified metal in the crucible and add a
pinch of Borax to purify, and then scrape away the remaining gunk. I
then pour it into a reversible ingot mold to get it to where I want
to go from, sheet or wire.

If you are not removing solder before melting, how are you
determining how much 24K to add to reach 14K?

There is a video here on Orchid BenchTube, “Pouring an ingot,” by
Jay Whaley, that shows the WHIP in use. The WHIP is available for
purchase from most catalog companies. It has made melting metal so
very much easier than the old style long handles heavy melting

Terrie, for me.

Further to my post yesterday, and reading the message about adding
50% ‘casting’ grain to scrap metal. Perhaps ‘Casting’ grain could be
the problem as sometimes casting gold alloys and silver are alloyed
with silicone which lower the melting point and aids the flow of the
gold. The alloys so made are not generally suitable for rolling or
forging, they should be used only for casting.

I melt scrap all the time and never add new metal. Dont add filings
melt them separately after careful cleaning. And test roll the
billet before adding it to the other scrap. See yesterdays post. I
will try to photograph and describe my simple melting and pouring
equipment and put up on Ganoksin.

David Cruickshank

It would be my guess that it’s a problem with the scrap. It may help
if you know the karat, whether or not the scrap has been melted
before, and how many times it has been melted before. It could also
be that the scrap could be laden with solder or the alloy is broken

Jon Head

I was taught to make gold in a well in a charcoal block - so there
is no pouring involved. I am going to try to mix well with a
charcoal stick, add a pinch of borax, see if this improve things.
If not, my only recourse is to send it out to be professionally
refined, correct? 

The charcoal block reduces oxidation, which is good, And borax flux
will remove existing oxides as well as also helping prevent more. but
oxidation is likely not the root of your problem. Rather, it’s likely
contamination with base metals such as traces of iron, zinc, tin,
cadmium, solder, etc. To help with these, you need a “refining” flux.
Principal among these in common use is ammonium chloride. Add a pinch
of the stuff to the already molten pool of metal and lightly stir.
You’ll also want a little bit of the borax. Ammonium Chloride
disassociates to yeild free chloride ions, (or chlorine gas), which
tend to scavenge those base metals forming metal chlorides. These, as
it happens, are no longer soluable in the molten metal, and rise to
the top as a slag, which mixes with the borax flux and is thus
eliminated. You may have to repeat this more than once if the metal
is badly contaminated. But it will help a lot. If you cannot find a
small quantity of ammonium chloride, you can use ordinary table salt.
It doesn’t work as well, but it does some of the same thing. In both
cases, do this with GOOD ventillation. Ammonium chloride releases
some nasty fumes.

By the way, just for the record, raising the karat of commercial 14K
up to 18K may not always be such a good idea. Most 14K commercial
alloys have some zinc or silicon in there as deoxidizers for casting.
These are not desirable in an 18K yellow gold alloy, so while you
have raised the karat to 18K, you might find the result is not quite
as nice a metal as you’d expect, either to work it, or for color.

Hope that helps.
Peter Rowe

if you are going to continue in the well method, try mixing a pinch
of sal ammoniac into the refinning lot in the well. Ordinarily if I
were going to use clean scrap (all solder removed, or rather, bench
sweeps, etc.) I would use my standard refining flux that is 3:1
powdered charcoal to sal ammoniac. I keep a small amount mixed up
and very tightly sealed as it is an humectant- it draws ambient
moisture from humid environs. I always add a generous pinch to the
crucible ( well glazed) when refining scrap or making a coloured
alloy. Since you are starting in a charcoal environment, I suppose
you could just add the ammonium sulfate to the block and let it
purify the scrap and new grain you are using.

Do stir though with a graphite or carbon rod (with a pot holder as it
heats up quite fast) - it will pick up debris you can see sticking to
the molten ball in the block, I melt over a mixture of silicon
carbide grain and powdered charcoal in my annealing pan, and just
give a snap of the wrist to the bits on the rod into the pan and then
remove them once cold to a steel ashcan (even grey heat is sometimes
still warm enough to ignite oily rags in a trash pail…so i use a
safety can for those types of “hazardous materials!” - nonetheless,
get them out of your annealing pan for later sifting and subsequent
reclamation of any retrievable metals that you send off to the
refiners as your own schedule dictates.

The ammonium sulfate (available in blocks called ‘sal ammoniac’ at
any stained glass supplier) yields a bright tough ingot great for
rolling out in a mill, or annealing and printing after processed
into sheet or sizing stock, etc… You must however store the block in
a tightly closed container as it will become damp rapidly in humid
weather, or a non-dehumidified studio. I know from years of using
this that it works well. as good as any deoxidizing grain, and as
for melting the scrap you described, there is a solder disclosing
powder you can get that will show any joins on a fabricated piece so
you can clip out the soldered areas and not contaminate your lot-
just throw the soldered clippings in the refinery container and leave
that part to your major refining at the end of whatever your cycle


The ammonium sulfate (available in blocks called 'sal ammoniac' 

“sal ammoniac” (an archaic name for the stuff) is ammonium chloride,
NOT ammonium sulphate. It acts to semi refine precious metals
because it preferentially forms metallic chlorides with the more
reactive metals, which are generally the contaminants causing
problems, like tin, zinc (not always a problem, but the ammonium
chloride will pull it out too), iron, lead, etc. These chlorides,
once formed, are not soluable in the molten metal, so in essence, the
contaminants are seperated into a slag floating on top.

Peter Rowe

I saw your advice regarding cracking of annealed gold. I am having
the same problem, but your explanations do not seem to fit my case.
My alloy is very close to 75% gold, 25% silver, and 0.3% copper on
analysis, so there is no iron, lead. Etc. Three goldsmiths have
annealed this alloy (obtained from natural placer gold) without
success. Do you have any further suggestions?

Ben Wilkinson


I am curious about you saying that it is obtained from placer gold.
How was it refined? I assume you are pouring an ingot and rolling
sheet. When you say what the percentages of metals by analysis, how
is that done? Have you used this alloy before?

When there are three people getting the same results, is this the
same metal used three times?

I have had more problems when making sheet using a plate mold to
pour an ingot than when I use a mold that produces a rod. Some people
forge sheet or rod ingots after pouring and then anneal before
rolling to change the crystaline structure to reduce the possibility
of cracks.

I do not know where the silver and copper come from. There have been
posts about what the source of copper is for using in alloying as
the copper can have impurities that make it cause problems for

Have you tried using a gram of the problem gold and mix 3/4 gram of
24kt pure refined gold and add the alloy to make 18kt and try rolling
that? If you tried that and still have problems, something is wrong
with the metal you are having a problem with and you need to refine
it. James Binnion is the most experienced person on this forum to
give suggestions and advice about this.

I have a lot of experience in trying to use customers old gold to
melt and roll out to use in fabricating pieces for them. If I do not
get success the first melt, usually melting more times will not
result in success without adding new metal. If you have any other
questions or info let me know, and let me know if you have success,
what did you do to achieve it.

Richard Hart G.G.
Jewelers Gallery
Denver, Co.

Thanks. This helps a lot! I might try salt first, but I will search
out the sal ammoniac. Just being careful to fully heat on my charcoal
block, use borax, and stir with a charcoal stick has given much
better - if not perfect - results. I really appreciate this help &


Hey Ben, I dredge as a hobby and I know the placer gold around here
is about 22k. Did you refine your placer prior to alloying? There is
no telling what trace contaminants it may contain that could be
causing the problem.

The river doesn’t care, it grinds up everything.

Good luck
Jim Doherty

every stained glass supply catalogue lists blocks of Sal Ammoniac to
date - it is a commonly used consumable in the stained glass trade
and is not called anything other than sal ammoniac- while it is an
archaic term for ammonium chloride- that’s how it is sold to this
day cheaply from any stained glass supplier. Mr. Rowe seems to differ
in his opinion on this but if anyone looks up up stained glass
suppliers you can order the stuff for generally under 3 bucks a
block- enough to last years. It is far superior to table salt- in
fact, table salt that is iodized is as useless as sugar in this
application- sal ammoniac as Carles Codina points out in many of his
books as well, is a great refining flux when combined with charcoal
to remove impurities in reclaimed gold scrap.

it is a commonly used consumable in the stained glass trade and is
not called anything other than sal ammoniac- while it is an archaic
term for ammonium chloride- that's how it is sold to this day
cheaply from any stained glass supplier. Mr. Rowe seems to differ
in his opinion on this 

No, we’re in agreement. I said sal ammoniac is the archaic term for
ammonium chloride. It did not say the term is no longer used. If you
buy it from a laboratory or scientific chemical supply company, it
likely will be labeled ammonium chloride, perhaps referencing the old
term for clarity. It’s proper modern chemical name is ammonium
chloride. But the popular old name of sal ammoniac is certainly still
in use, especially in those trades, like stained glass and no doubt
others, where correct modern chemical nomenclature is not required.

The same can be said for any number of other chemicals we might use.
Liver of sulphur, for example, is not precisely correct current
scientific chemical nomenclature, but to jewelers, that certainly is
the recognized name.

The only thing we are not in agreement about was the post in which
you referred to sal ammoniac as ammonium sulphate rather than
ammonium chloride. No doubt a simple typo, but it nevertheless could
have been misleading in the unlikely case some reader went shopping
for sal ammoniac and instead bought ammonium sulphate, so I pointed
out the error. I trust you didn’t mind.

...It is far superior to table salt- in fact, table salt that is
iodized is as useless as sugar in this application 

Sodium chloride (salt) is certainly less effective as a refining
flux. But it IS cited in a number of old turn of the century (George
Gee’s books, for one) references on refining fluxes, which refers to
salt as an agent which can help in “toughening” cracky gold alloys.
I also believe I recall the note that it tends to make the fluxes
thicker, more viscous, perhaps making pouring an ingot without flux
inclusions, easier (?). I doubt it’s in common use today, but
certainly at one time, was used often enough to warrant mention in
such books. I was not aware that the tiny amount of iodine in
iodized salt would affect it’s ability to draw base metals as
metallic chlorides out of the melt, but I’ll take your word for it.

Peter Rowe