The magnetic separator downstream of our grinder recovers about 175 tons of ferrous metal a year — nails, staples, banding and the occasional forklift tine.
It is 0.4% of everything that crosses our scale and by far the least interesting product we make. It is also, it turns out, a surprisingly good indicator of what is happening upstream.
The baseline
A 48×40 GMA holds 78 to 90 nails at roughly four grams each — about three-quarters of a pound of steel per pallet.
If everything that entered the yard were a standard GMA and everything were ground, ferrous recovery would be a predictable fraction of intake weight. It is not, because only about fourteen per cent of intake is ground, and the units that reach the grinder are not a random sample.
So the ferrous number is really telling you about the composition of the grind stream, which is telling you about the quality of what came in.
What the year showed
The Q4 jump is the interesting one. It coincided with two new industrial collection customers whose material carries more banding, more fasteners and more mixed hardware than distribution stock does.
We noticed it in the ferrous number before we noticed it anywhere else, which was not something we expected.
| Quarter | Intake (tons) | Ferrous recovered | As % of ground material | Note |
|---|---|---|---|---|
| Q1 | 9,800 | 38 t | 1.4% | Winter intake, wetter, more grinding |
| Q2 | 11,400 | 47 t | 1.5% | Spring peak, higher damage rate |
| Q3 | 10,900 | 42 t | 1.3% | Drier stock, more reuse |
| Q4 | 10,600 | 48 t | 1.7% | Two industrial contracts started |
Why it is worth counting
Partly because a diversion figure that ignores every stream too small to be interesting is not a measurement. Our published 99.4% includes the nails on both sides of the fraction, which is the only way that number means anything.
Partly because it is a cheap leading indicator. A rising ferrous fraction means the grind stream composition is changing, which means either intake quality is falling or the customer mix is shifting, and both are things worth knowing early.
And partly because 175 tons a year is 175 tons. It goes to a scrap processor, it comes back as steel, and it is not in a mulch pile on somebody's flower bed.
The general point
Every stream in a yard is telling you something, including the ones that are commercially trivial. We started weighing the ferrous output because the diversion arithmetic demanded it and kept doing it because it turned out to be useful.
That is roughly the argument for measuring anything: you find out what it means afterwards.