Something that bites us every few months and I have never seen a straight answer on: how much the residual elements in scrap-based tool steel move the heat treat result.
We buy our blocks through a distributor, send them out for hardening, and every so often a batch comes back off the number. Same recipe, same furnace, same guys running it. Hardness high on one block, low on the next, and occasionally a part that moves more in the quench than it should. Before I go and blame the heat treater I want to know how much of that is already in the steel when it lands on my floor.
The way I understand it, most of this steel starts as an electric arc furnace charge that is basically all scrap, and a whole list of elements comes along with it that nobody can take out at a price: copper, tin, arsenic, antimony, nickel, chromium, molybdenum. Low levels of some of these are harmless. Others are not, and the reasons are pretty specific:
- copper and tin are what give you hot shortness and surface cracking during rolling
- arsenic, antimony and tin are the ones pushing temper embrittlement
- nickel, chromium and molybdenum that are not held in a tight band shift the hardenability itself, which is exactly why one heat over-hardens on a recipe that under-hardens the next
I know the mills have levers here. Pickier scrap buying and sorting, diluting the charge with virgin metallics like DRI/HBI or pig iron, tighter control in the melt shop, and then secondary refining plus ESR on top for the critical grades. What I cannot tell is how much of that actually reaches a block of tool steel I can buy at a normal price.
So, four things I would like real numbers on:
- For a given grade, how much heat-to-heat spread in hardenability do you consider normal, and where does it stop being normal and start being a problem?
- Do you write residual limits (Cu, Sn and the rest) into your purchase orders, or do you just take whatever the mill’s standard practice gives you?
- Has anybody actually measured what residuals do to dimensional stability after quenching, meaning distortion and growth, not just hardness?
- Tool steel specifically: is there a residual level where you would refuse the heat outright?
What I am chasing is a way to tell steel-caused variation apart from furnace-caused and operator-caused variation, because right now every bad batch looks the same from where I sit.