Phoenix Ascent

Turn a pile of discarded metal into a weighed ingot, and know 💡 the numbers 🔢 before you light the furnace.

Phoenix Ascent is a workbench 🛠️ for anyone who melts scrap to recover the metal in it. Drink cans, window extrusion, brass fittings, a box of broken chain, inherited silverware, a bag of gold-plated connector pins — bring the pile, and the app 📱 answers the three questions it actually raises.


WHAT IS THIS?

Weigh a piece in air, then weigh it again submerged in water. Phoenix Ascent back-solves its specific gravity and ranks every alloy it could plausibly be, each with a confidence figure and the measurement error ❌ behind it.

The uncertainty is the point. Your scale's resolution sets the width of the band, so a small piece on a coarse scale produces a wide answer — and the app 📱 shows you that instead of hiding it. When two candidates sit too close to separate, it tells you which scale resolution would separate them, or says plainly that specific gravity cannot settle it and names the tests that can. Gold and tungsten differ by 0.07, and the app refuses to pretend otherwise.

Stamped and hallmarked pieces resolve straight from their mark, flagged as asserted rather than measured — because stamps are forged.


WHAT DO I ADD, AND WHAT WILL I LOSE?

Sort 🔄 your pieces into a lot and pick a goal 🎯.

For casting alloys, choose a target 🎯 specification and the app 📱 solves the mass balance ⚖️: exactly what to add, in grams, to land every element inside its window. When the target cannot be reached it says so and diagnoses why ❓ — naming the element that binds, how much clean stock you would need to dilute it, and what you get if you simply pour it anyway, matched against the nearest standard alloy.

For gold and silver, it computes the refining route 🧭 from the lot itself, including how much silver or copper to add to bring gold below the parting threshold. Acid quantities come from the reactions and the concentrations you keep on your bench, shown as both the stoichiometric minimum and the practical figure. Evolved gas volume is calculated for your specific lot and drives a ventilation status, because a small bag of low-karat scrap is a bigger job than it looks.

Loss is driven by form, not alloy. Thin sheet oxidises away; a solid billet barely does. Zinc boils out of hot brass and magnesium burns off aluminium, so the app 📱 shows the composition that actually survives the melt — along with fuel, cost, elapsed time ⏱️, and a temperature schedule 🗓️.


WHAT DID I ACTUALLY GET?

Log 📝 the real recovered mass. Phoenix Ascent compares it against the prediction, attributes the gap to specific loss stages rather than shrugging at a shortfall, and adjusts its model of your furnace and your technique. Every later forecast uses what it learned, and shows how confident it is in that learning.


ALSO IN THE APP 📱

• Photograph each piece and its hallmark, so you know 💡 which is which later
• A stock ledger of recovered metal, valued at prices you set
• Save 💾 an ingot card straight to your gallery 🖼️ — composition, mass, fine content, purity, date
• Return an ingot to the bench as feedstock for a future melt, closing the cycle
• Charts 📊 covering recovered metal over time ⏱️, where your losses actually go, cost per recovered gram, and how closely predictions are converging on reality


HOW IT BEHAVES

Every figure carries its origin: the drivers that produced it, the confidence around it, and the stage that would change it. Ambiguity is reported as a result, not smoothed over — one unidentified piece makes every downstream number 🔢 a range, and verdicts are stated against the pessimistic bound. The same inputs always produce the same answer.

Phoenix Ascent computes; it does not supervise. Molten metal and strong acids demand proper equipment, ventilation and training, and nothing here replaces them.