The count is the easy half
Stack nine bars, weld them, and you have nine layers. Cut that in half and weld it back on itself and you have eighteen. Do it again and you have thirty-six. The count is the starting stack multiplied by the multiplier raised to the number of operations, and that is genuinely all there is to it.
The multiplier does not have to be two. Cutting a billet into three and restacking gives a factor of three per operation, which climbs much faster: nine pieces and four operations gives 144 layers by folding and 729 by cutting into three. Neither is harder to arrive at, and the difference in the finished pattern is a matter of how thin the layers end up rather than of how many operations you did.
The number that matters is the layer thickness
Layer count on its own says nothing, because it is a count across whatever thickness the billet happens to be at the time. What you can actually see is the thickness of an individual layer at the point the blade is finished, and that is simply the finished stock thickness divided by the count.
Grinding does not change the count. Take a 144 layer billet down from half an inch to a tenth and there are still 144 layers across it — they are just each a fifth as thick as they were. That is why the pattern gets finer as you grind, and why the same billet can look coarse on a thick chopper and almost silky on a thin kitchen blade.
It is also where the diminishing returns live. Past some point the individual layers are thinner than the etch and the finish can resolve, and adding another weld buys nothing you can see while costing exactly as much steel as the first one did. The page prints the layer thickness at every stage so you can find where that sits for your own etching and finishing.
The losses compound, and the first weld counts
| Welding heats | Steel left at 8 percent a heat |
|---|---|
| 1 — the stack weld only | 92.0 percent |
| 3 — stack plus two folds | 77.9 percent |
| 5 — stack plus four folds | 65.9 percent |
| 7 — stack plus six folds | 55.8 percent |
The first weld is the one people forget to count. It does not multiply anything — nine bars welded together are still nine layers — but it takes exactly the same bite out of the steel as every fold after it, because it involves the same scale and the same grinding back of the edges. A schedule described as four folds is five welding heats.
Eight percent a heat is used in that table purely to show the shape of the compounding; the real figure is yours to weigh, and it moves a great deal between a set of welds that took first time and a set where you ground back a bad edge twice.
What the arithmetic does not know
It counts laminations. It has nothing to say about what those laminations look like on the surface, because that depends entirely on what you do to the billet after the last weld — drawn flat they run along the blade, and twisted, ladder cut, raindrop punched or ground into they produce completely different things from the same count.
It also has no view on which steels to lay against each other, on whether a weld took, or on what any of it does to the finished blade. Those are questions about specific steels and about the pieces in front of you, and neither belongs on a page that only knows how to multiply.
Questions people ask
How many layers do I get from nine bars and four folds?
One hundred and forty-four. Nine in the stack, and each fold doubles it, so nine times two to the fourth. If you cut and restack into three instead of folding in half, the same four operations give 729. The count is the only genuinely simple part of pattern welded work, which is probably why it is the part that gets quoted.
Does grinding the blade thinner reduce the layer count?
No. The layers run through the thickness, so grinding makes each one thinner without removing any of them. A 144 layer billet is still 144 layers at a tenth of an inch as it was at half an inch, with each layer a fifth the thickness. That is exactly why the pattern gets finer as the blade gets thinner, and it is the calculation this page is built around.
Should the first weld count towards the material loss?
Yes, and it is the one that gets left out. Setting the stack does not multiply the layer count, but it costs the same scale and the same grinding back of edges as every fold that follows. So what people describe as four folds is five welding heats, and at eight percent a heat that difference alone is another eight percent of the billet. The page counts operations plus one.
Is there a point where more layers stop being worth it?
There is, and it is set by your finish rather than by the steel. Once an individual layer is thinner than the etch and the polish can resolve, another weld doubles a count nobody can see while removing exactly as much steel as the first one did. Where that lands depends on how you etch and how fine you finish, which is why the page prints the layer thickness at every stage instead of naming a number.
What loss percentage should I use?
One you have weighed. Put the stack on a scale before the first weld and put the billet on it after the last one, and the difference is scale plus everything ground off the edges and cut off the ends. It is not a property of the steels involved — it is a property of how cleanly the welds went, so a set that took first time and a set where you chased a bad edge produce very different numbers on the same materials.