Problems with rolling 9ct gold

hi, I've been sandcasting gold rings for a while but thought I'd try making some bangles then maybe have a go at making a curb chain.
I've recently picked up a new rolling mill and Researched the hell out of which way to go about making 9ct yellow gold wire.
The problem I'm getting is the gold starts to get small cracks after rolling for a while (whilst annealing in between). And I can even snap the gold like a matchstick.
I've seen so many different ways online of the full process. Which is the best way? Or where am I going wrong?
I'll list equipment I'm using then process...
Small kiln with 4oz graphite crusible
Wire reversable ingot
Basic rolling mill (unknown make/cost £180)
9ct scrap yellow gold. (Already melted down 2 times)

Melt in crusible adding a small amount of borax toward the end
Heat ingot slightly
Pour
Run through the mill as I've read everyone does, closing down slightly each time
After around anneal until red, leave till red goes away then quench
Then mill again.

I have read that hammering down after pour can help but I haven't tried this yet.
Will try this today.

I'm definitely no expert, but do you need to quench?

When knife making, quenching hardens the metal and makes it more brittle and prone to snapping. And only done just before finishing stages, after shaping has been completed. It's prior to quench and post-heating that the metal is soft and best for reworking. I don't know if this all translates over to jewelry though, as the properties of gold are much different than steel.

I hope this helps but maybe best leave this to someone more experienced than me!

Thanks for the reply, just about every video I've watched it's quenched but I'll try still that.
It's a long process of trial and error but I'll get there

And while the wire is fat, reduce the gap more slowly and anneal after two passes.

One of the problems with annealing wire is that if the flame is too fierce and not moved around a lot some sections will become overheated and too brittle to work. Anneal in a darkened corner and never heat more than dull red, as evenly as possible.

If your annealing is successful, the wire will feel soft to slight manipulation by hand. Obviously making wire in a mill is not entirely successful and you next need to think about draw plates for the final passes.

Unfortunately, the heat treatment of ferrous metals is completely different and almost the direct opposite. Dennis.

Thanks Dennis. Yes I do have a draw plate set up in a vice ready.
Maybe it's down to the annealling then.
I'll try annealling it more often.
Just to check the way I'm doing it is right...
I heat the metal with the tip of the flame at one end until red then move along it making the rest red as I go. So not moving the flame side to side.
Then leave it till the red has gone then 5 seconds or so then quench and roll

As said above, the wire is easily overheated in patches and will then break. The point of the flame outside the blue cone is the hottest part, so pull the flame back a little and move it from side to side, to avoid lingering in one place.

A darkened area allows you to check that the metal only ever goes to dull red and that you have done the whole length evenly. Dennis.

It may be better to anneal the wire in the kiln to get a more even temperature across it?
Its definitely the annealing that is the problem.
This table may be useful although it doesn't give the time needed:
http://www.nancylthamilton.com/wp-co...for-Metals.pdf
Also this video shows 14k gold being annealed in a kiln throughout the drawing process:
https://www.youtube.com/watch?v=WlIuO0aKfqk

There seems to be a lot more to it than I thought.
I've been sandcasting gold rings with scrap for a couple of years now and never really come across many issues.
I've just read that it's advised to add new gold to the melt as scrap gold can crack etc when milling for a variety of reasons.
Which in turn will make my hobbie very expensive.
My kiln is very basic. It only holds a 4oz crusible and has no temp indicator.

Ah ok, just thought it might be easier but obviously won't work with that kiln.
In that case just follow Dennis' advice above and be careful to heat it evenly.
I did read that you should hold at the dull red glow for 30 seconds so I always do that with mine, any less and it doesn't always anneal properly.

Quote:

---

Originally Posted by Simmo

..
I've just read that it's advised to add new gold to the melt as scrap gold can crack etc when milling for a variety of reasons.
Which in turn will make my hobbie very expensive.

---

I am very far from an expert, but I believe this is only recommended if you don't know the source and/or quality of the scrap you are using, or, you are melting down scrap jewellery that has been soldered in places. The experts say that it is the impurities like solder in your scrap that causes the metal to become brittle and crack when cold working. This is why they advise adding pure gold or silver (depending on the alloy you are working with) to your scrap.

But, if like me, you make your own alloy, and all your scrap comes from the same stock, and none of it has any impurities like solder in it, then I cannot see any logical reason to constantly add new silver or gold to it. There is nothing in the process that can alter the properties of the alloy, unless I am missing something(?).

This isn't just pure theory, I make my own sheet from my own stock of alloy, and my scrap has been melted down and rolled into sheet countless times without ever adding any new silver. No cracking or brittle problems yet.

So I'm still having the same problem.
I've been slowly heating it till just gone slightly red then hold it for around another 10/15 seconds...leave it for 19 seconds or so then quench.
It seems that by the time I get to the 3rd size of the round rollers the gold is quite curved/bent and just starting to show signs of cracking, annealing frequently inbetween.
I might contact the local jeweller and and ask if he could kindly show me or even take my gold there to see if there's any problem with it.

Is it actually annealing properly each time?

I can only assume so as I haven't done annealling before. I have read a lot of ways to do it and watched countless videos on it.
The only thing I'm not doing is pickling it.
I do all my work at my home so I wasent going to pickle until the end product (I haven't used pickle before by the way, when sandcasting rings I've just had to file sand and polish)

Quote:

---

Originally Posted by Simmo

I can only assume so as I haven't done annealling before. I have read a lot of ways to do it and watched countless videos on it.
The only thing I'm not doing is pickling it.
I do all my work at my home so I wasent going to pickle until the end product (I haven't used pickle before by the way, when sandcasting rings I've just had to file sand and polish)

---

What is the original source of your scrap 9 ct gold?

Sent from my SM-T520 using Tapatalk

If its annealed correctly it should be fairly malleable straight afterwards.

Not an expert here but from my limited metallurgy knowledge here are my thoughts:
Melting metal too hot will make an unstable brittle ingot that no amount of annealing will solve. Oxygen contamination will produce a brittle ingot. Hammering before rolling is important. Annealing temperture, length of time and type of flame.

Sent from my SM-T520 using Tapatalk

I get most of my gold from eBay. I've seen quite a few people disagree with getting it from there but I test everything before I use it and I've made items before and had them hallmarked in the uk just fine.
I haven't tried hammering the ingot so I may try this.
Over melting may be a cause too as I only check the kiln every 20 minutes to see if it's melted enough.
Oxygen contamination...I know nothing about.

I am a betting man and my money is on the impurities causing the problems. You buy scrap gold which probably contains a lot of solder joints. Zinc is a component of solder and it is the only metal generally used in 9ct gold described as being "hard and brittle at most temperatures."

I don't profess to be an expert, but from everything you say it makes the most sense to me and also explains why the suggestion is to add pure gold to your scrap. The purpose, I expect, is to dilute any zinc in the scrap in order to make it less susceptible to the very problems you are experiencing.

Zinc...my money is on the zinc!

Thanks for the advice everyone. I think I'll just cast another ring from this batch and have a go at milling with the next lot of gold I get.
...or look up how to extract the zinc from it.

So I've tried it again and...it's working for me better now.
I've been using a typical wire ingot but it's so hard to try and pour into the smallest wire hole, some of the gold goes in but most seems to fall over the side.
I only manage to get the wire ingot full half way around 1/10 attempts.
But it's rolling much better and annealling it has got to plan each time. It just seems to work better for me with thinner metal.
Only problem I've come across now is just when I'm getting to the thickness I would like (for a basic ring band) the metal seems to curve off to the side.
This may be fixable through hammering on the mandrel once I've shaped and soldered it together

1. You might do better just pouring into a groove in a block rather than bothering with a mould.
2. If the wire is crooked, anneal it again, clamp one end and pull with serrated pliers. It will reduce in diameter slightly, but become straight. Dennis.

You may find a mould like this more forgiving http://www.cooksongold.com/Jewellery...prcode-999-930
When rolling to shape if you turn the wire each time it will help keep it straight

Thanks but even that ingot is too big.
The size that I poured was a 3mm and a 4mm and I managed to mill them quite easily.
Maybe I'll start small like that then work my way up.

Is there anything I could use as a funnel for the small wire ingot? Like cut and shaping a coke can for example

You can reshape the access with a cylindrical burr (preferably a tungsten carbide one) to make it funnel like. However a neat groove in a block would be simpler. Dennis.