Merchant Shear Angle from Chip Thickness Ratio

In 1945 Eugene Merchant published two papers in the Journal of Applied Physics that turned metal cutting from a craft with tables into a mechanics problem. The model is deliberately simple: a straight edge wider than the cut, taking a chip of uniform thickness, with all the deformation happening on a single flat plane running from the tool tip up to the free surface. That is orthogonal cutting, and the plane's angle $\phi$ is the quantity everything else depends on.

What makes the model usable is that $\phi$ can be measured without instrumenting anything. Because the volume of metal is conserved, a chip that comes off thicker than the depth of cut must come off correspondingly shorter, and the ratio $r = t_0/t_c$ of uncut to chip thickness pins the geometry down completely. Stop the cut, collect a chip, and either measure its thickness with a ball-ended micrometer or weigh a measured length and back the thickness out of the density. The ratio is always less than one — the chip is always thickened — and a ratio quoted above one is a ratio taken upside down.

The geometry then gives $\tan\phi = r\cos\alpha/(1 - r\sin\alpha)$, and the message is that a high shear angle is good. The shear plane runs across the cut at $\phi$, so a small angle makes a long plane, a thick chip, and a large volume of metal deformed for every unit removed. Since nearly all the cutting energy goes into that deformation, a low shear angle is directly a high specific cutting energy and a high cutting force.

Merchant's own contribution was a second relation to close the model. He assumed the shear angle settles wherever the total work is a minimum, and derived $2\phi + \beta - \alpha = 90°$, with $\beta$ the friction angle at the chip-tool interface — that is, $\beta = \arctan\mu$. Two consequences fall straight out. Every degree of positive rake buys a degree of shear angle, which is why positive-rake tooling cuts more freely and needs less power. And reducing friction on the rake face raises $\phi$ directly, which is the mechanism by which cutting fluid and low-friction coatings work — not primarily by cooling, but by changing $\beta$.

Treat that relation as a guide rather than a prediction. It assumes a single shear plane, a perfectly sharp edge, and a friction coefficient independent of the enormous normal pressure at the interface, and none of those is true. Real cutting has a thick shear zone, an edge with a radius, and a built-up edge that appears and disappears with speed. Measured shear angles come out well below what Merchant's relation predicts, and Lee and Shaffer's competing slip-line-field solution gives a different answer again. The model earns its place not by being numerically right but by explaining, in one line of geometry, why rake angle, friction and cutting force are the same subject.

One last practical note on rake. Negative rake looks like a mistake in this model — it lowers the shear angle, thickens the chip and costs power — and it is nonetheless what most carbide turning of steel uses, because a negative-rake insert has a stronger edge, can be indexed to use both faces, and puts the cutting force into compression where carbide is strongest. Positive rake is what soft or gummy materials and light machines need; negative rake is what hard materials and rigid machines can afford.