My First Gold Smelting Attempt: What Went Wrong and What I Learned

Yesterday I ran my very first gold smelting attempt.

I would love to say it ended with a neat little button of gold, sitting there like proof that I knew exactly what I was doing.

That is not what happened.

But that is also the point of this post – and probably this whole gold experiment series.

I am not documenting this because I have already mastered the process. I am documenting it because I am learning it from the ground up. That means the failures, the ugly results, the underwhelming first attempts, and the parts where I have to stop and admit that I clearly need to improve something before trying again.

This first smelting attempt was not what I would happily call a success.

But it was a start.

And sometimes the start is messy.

Where This Attempt Fits Into the Bigger Gold Experiment

This post is part of my ongoing project:

Learning Gold the Hard Way: Fossicking, Smelting, and Small-Scale Experiments

The broader goal is to learn the full process properly, from finding and collecting material through to processing, separating, testing, and eventually attempting small-scale smelting.

I am not approaching this as an expert.

I am approaching it as someone who is interested, curious, and willing to learn by doing – even when doing leads to less-than-glamorous results.

That is exactly what happened here.

This was my first real smelting operation, and it has already shown me that there is a large difference between understanding the idea of smelting and actually getting a good result from it.

The Goal of the First Smelt

The goal of this first attempt was fairly simple:

I wanted to see whether I could take gold-bearing material or concentrates and produce some kind of visible result from a small smelting process.

Not necessarily a perfect result.

Not even necessarily a clean result.

Just something that would help me understand the process better.

I wanted to test the basic workflow:

Prepare the material
Add flux
Heat it properly
Allow the material to melt and separate
Cool it down
Inspect the result
Work out what went wrong or what needs improving

That was the plan.

The reality was not quite as straightforward.

The Material I Started With

For this first attempt, I used material that I had collected and processed from my fossicking efforts.

At this early stage, I am still learning how well I am actually preparing the material before it gets anywhere near a crucible.

That is already one of the biggest lessons from this attempt.

Smelting is not magic.

If the material going in is not properly prepared, concentrated, cleaned, or understood, then the result coming out is probably going to be confusing.

For this attempt, the material included:

Source material: panned concentrates collected from recent fossicking trips
Approximate amount: less than 100g
Visible gold before smelting: uncertain
Amount of black sand or heavy material: moderate
Preparation before smelting: concentrates were re-panned at home to further separate undesirable materials, magnet used in wet material to collect the magnetic materials, panned again and then placed into stainless steel pot to heat until completely dry.

Looking back, I suspect this is one area I need to improve before the next attempt.

I may have been too eager to get to the smelting stage before I had properly refined and cleaned the material.

That is a very easy mistake to make, because smelting is the exciting part.

But the exciting part probably depends heavily on the boring preparation part.

The Setup

My setup for this first smelting attempt was basic.

That is not a complaint. It is just the truth.

I am not using professional refining equipment or a commercial setup. This is a small-scale learning process, and the equipment reflects that.

For this attempt, I used:

Crucible: 3kg graphite crucible
Flux: homemade anhydrous borax, soda ash (sodium carbonate); 25% of raw material each
Heat source: ToAuto 3kg electric furnace
Safety gear: fire-safe area, metal-working table, long-sleeved leather gloves, leather apron, safety glasses, steel-cap workboots, breathing mask, outdoor environment
Cooling method: steel bucket containing cold water
Mould or receiving surface: graphite moulds

The setup was enough to run the attempt, but I am not yet convinced it was enough to run the attempt well.

That distinction matters.

It is one thing to get material hot.

It is another thing entirely to get the right material hot enough, for long enough, in the right conditions, with the right preparation and flux balance.

That is where I still have a lot to learn.

What Happened During the Smelt

Once everything was set up, I added the prepared material and flux to the crucible and began heating.

At first, things looked promising enough.

The material responded to the heat, and there were visible changes as the process continued. But as the attempt went on, it became clear that I was not heading toward the clean result I had imagined.

The first observations I noticed as the temperature was rising, was the foul-smelling brownish-yellow smoke coming from the top of the furnace. It was very much like a rotten egg smell – an indication of sulfides.
I am extremely glad that I was outside performing this smelt, and that I had a breathing apparatus handy. The gas emitted, was Hydrogen Sulphide, which is quite toxic.

As the smelt progressed, the smoke coming from the top vent on the furnace lessened, which I took to be a good sign that it had burnt off.

I commenced preheating the graphite moulds with a butane torch, along with a graphite stirring rod.

Once it had been sitting on 1100 degrees Celcius for around 10 minutes, I opened the top and stirred it with the graphite rod, and then closed it back over.

I then proceeded to pour the contents of the crucible into the mould.

Instead of a clear separation and an obvious metallic button, the result was more uncertain.

The material appeared to very quickly turn from an orange-red colour to a dark red and grey colour, allowing no time at all to perform a hot separation process.
I flipped the experiment onto a steel plate and picked it up with steel tongs, quickly submerging it into the cold water.
There was no satisfying pop, sizzle, or anything remotely like that.

Instead, the mass simply crumbled and sunk to the bottom of the bucket, turning the water a black colour as it did so.
And the smell – it was back seemingly worse than before!
The respirator went back on, and I then recovered as much of the crumbled glass material as possible, washing and panning and rinsing the material until the water was much clearer than ‘black’.
The washed material is now set aside, drying so that I can crumble the glass back down into a powder to be smelted again – with higher flux ratios to account for the sulfides within the concentrates.

These were the moments where the difference between “watching smelting videos” and “actually doing it yourself” became very obvious.

In a successful-looking smelt, you expect some kind of confidence in the outcome.

In this attempt, I mostly ended up with questions:
Was there not enough gold in the material?
Was the material not clean enough?
Was the heat insufficient?
Was the flux wrong?
Was the ratio wrong?
Did I rush the process?
Did I misunderstand what the input material actually contained?

At this stage, I do not have a perfect answer.

But I do have a result to learn from.

As a plus note, there does appear to be at least some gold in the glassy black mess. I can see some flecks, but it does not seem like they got to a molten state.

The Result

The final result was not a clean little gold button.

That would have been nice.

Instead, I ended up with sludge and glass mixture that possibly contains small metallic-looking specks.

It was not useless, though.

A failed result still gives information.

It told me that my process needs work. It also reminded me that the quality and preparation of the material before smelting is probably more important than I wanted to admit going in.

The result may not have looked impressive, but it has given me a baseline.

This is attempt number one.

From here, I can compare future attempts against it.

If the next result is cleaner, I will know I improved something.

If the next result is just as bad, I will know I still have a deeper issue to solve.

Either way, the process now has a starting point.

What I Think Went Wrong

I do not want to pretend I know exactly what went wrong yet.

But I do have a few likely suspects.

1. The material may not have been prepared well enough

This is probably the biggest one.

I may need to spend more time separating, cleaning, drying, and concentrating the material before attempting another smelt.

If too much unwanted material is going into the crucible, then I am making the smelting stage harder than it needs to be.

2. There may not have been enough gold in the sample

This is the very unromantic possibility.

Maybe the material simply did not contain enough gold to produce a visible result.

That is not failure by itself. It just means I need to test better samples and not assume that heavy material automatically equals a worthwhile smelt.

3. The heat may not have been right

Smelting is not just about applying heat.

The material needs to reach the correct temperature and stay there long enough for the process to work properly.

If the heat was too low, uneven, or not sustained for long enough, that could explain part of the result.

4. The flux mix may need improvement

Flux is one of those areas where I clearly need to learn more.

The wrong amount, wrong type, or wrong balance could easily affect how well the material melts, separates, and forms slag.

For this first attempt, I was mainly trying to get the process moving.

For the next attempt, I need to be more deliberate.

5. I may have rushed into smelting too early

This might be the most honest answer.

I wanted to try the smelting stage.

That is understandable.

But I may have jumped ahead before the material was ready.

The next attempt should probably start much earlier in the process, with better preparation and better observation before anything goes into the crucible.

What I Learned From This First Attempt

Even though this first smelting attempt was not a success in the way I had hoped, it was still useful.

The main lessons so far are:

Smelting is not a shortcut around poor preparation
Concentrates need to be properly cleaned and understood
A visible result depends heavily on what is actually in the sample
Heat, flux, and timing all matter
A failed first attempt is still a useful reference point
I need to slow down and improve the steps before the smelt

The biggest lesson is probably this:

The smelting stage is only as good as everything that happens before it.

That feels obvious now.

It did not feel quite as obvious before I tried it.

What I Will Change Next Time

Before I run another smelting attempt, I want to improve the preparation stage.

The next attempt should include:

Better classification of the material
More careful panning and separation
Less unwanted material going into the crucible
A clearer idea of whether visible gold is present
Better notes on sample size and source
More deliberate flux use
More careful observation of heat and melt behaviour
Photos or video at each stage, if practical

I also want to keep better records.

For the next smelt, I should be able to write down:

Where the material came from
How it was processed
How much material was used
What flux was used
How long it was heated
What the material looked like during the process
What the final result looked like
What changed compared to this first attempt

That way, this becomes more than just “try again and hope”.

It becomes a proper learning process.

Was This First Smelt a Failure?

Yes and no.

If the goal was to produce a clean gold result, then yes, this attempt failed.

If the goal was to begin learning the process, then no, it did exactly what a first attempt often does.

It showed me that I do not yet know enough.

It exposed weak points in my setup and preparation.

It gave me questions to answer before the next attempt.

And it gave this whole series a very honest beginning.

That is probably better than pretending everything went perfectly.

The Honest Starting Point

There is something fitting about starting this gold experiment with a failed smelt.

Gold has a reputation for being shiny, valuable, and exciting.

But the process of getting it, cleaning it, separating it, and trying to turn it into something useful is not always shiny at all.

Sometimes it is dirt.

Sometimes it is black sand.

Sometimes it is slag.

Sometimes it is a disappointing lump that makes you question what you actually did.

But that is the process.

And this is where mine begins.

Next Step

The next step is not simply to smelt another batch and hope for a better result.

The next step is to go backwards.

Back to the material.

Back to the concentrates.

Back to the preparation.

Before I attempt another smelt, I need to make sure the sample going into the crucible is actually worth smelting and has been prepared properly.

That will likely be the focus of the next update.

For now, this first attempt stands as the beginning of the record.

Not a success.

Not a disaster.

Just the first real step in learning gold the hard way.

Part of the Gold Experiment Series

This post is part of the jaysndees ongoing gold fossicking and smelting experiment series: