Last updated 10 August 2026
Four times the per-side figure, but only half the coat is per-side. A 1/2-13 UNC-2B tapped hole anodized to a .0004 total coat has to be tapped to .4508 / .4573 on pitch diameter.
Two corrections land on the same feature and they do not cancel. A coating sits normal to the thread flank, so it moves pitch diameter about four times the per-side thickness rather than twice. And an anodize call-out states the TOTAL coat, of which only about half grows above the original surface, so the per-side figure is half the number on the print. Four times a half is two: an anodized thread moves about twice what the drawing's coat thickness suggests, not four times and not half.
The direction is the other half of it. This is a tapped hole, so the oxide grows inward off both flanks and closes pitch diameter, and the tap has to leave it open. On a stud the same call-out runs the other way.
This is the case where a class 3 fit and an anodize spec on the same drawing is worth a phone call. A small fine-pitch thread in class 3 has so little pitch diameter tolerance that a normal anodize range consumes the lot, and the honest answer is that it cannot be held: the calculator will say NO SIZE HOLDS rather than round it away. Shops mask small threads on anodized parts for exactly this reason.
| On the print | .4500 / .4565 in |
|---|---|
| Process | Anodize |
| The feature | thread, both surfaces in the process |
| Per side | 0.0004 in |
| The dimension moves | smaller at this operation |
| Shift per unit of per-side effect | 4 times |
| Hold before the process | .4508 / .4573 in |
Open this case in the calculator, and change the numbers.
Computed by ctrlPlanner. MIL-STD-870, Federal Screw Thread Specification H28.
This is one characteristic. A real part has forty of them through five operations, and every one needs the same walk when the drawing revises. That is what ctrlPlanner does from a ballooned print.
Open the demo · a print is already loaded, no sign up.