Anodizing is a conversion coating, so roughly half of it grows above the original surface and half of it eats into the metal. That means an outside dimension gets bigger, a bore gets smaller, and a tapped hole loses four times the coating thickness on pitch diameter. This calculator gives you the pre-anodize size to machine and the tolerance the coating will eat.
Last updated July 2026
No masking required. Machine at 0.9997 in and the feature lands inside the print after coating with 0.00160 in of tolerance left over.
The benchmark. Standard 50/50 conversion, clean uniform color, best all round anodize response of any structural plate alloy.
General purpose sulfuric anodize. Minimal dimensional effect on most features.
A 1.0000" bore getting Type III hard coat at 0.0020" on 6061 grows 0.0010" per surface, closes 0.0020" on diameter, and must be machined at 1.0020" before anodize. A 1/4-20 tapped hole in that same part loses 0.0040" of pitch diameter, which is 82 percent of the entire class 2B tolerance band, so plug it or tap it after anodize.
| Coating | Thickness Range (in) | Default Target (in) | Thickness Tol (in) | Growth Per Surface (in) | Notes |
|---|---|---|---|---|---|
| Type II Class 1 clear, sulfuric | 0.0001 to 0.0009 | 0.0003 | +/- 0.0002 | 0.00015 | General purpose, minimal dimensional effect |
| Type II Class 2 dyed, sulfuric | 0.0004 to 0.0010 | 0.0007 | +/- 0.0002 | 0.00035 | Black dye runs the thick end of the range |
| Type III Class 1 hard coat, undyed | 0.0005 to 0.0045 | 0.0020 | +/- 0.0005 | 0.00100 | MIL-A-8625 default is 0.002 in unless the print says otherwise |
| Type III Class 2 hard coat, dyed | 0.0010 to 0.0040 | 0.0020 | +/- 0.0005 | 0.00100 | Same growth math as undyed hard coat |
| Chem film per MIL-DTL-5541 | about 0.00003 | 0.00003 | negligible | 0.000015 | No meaningful dimensional change on any feature |
Growth per surface is half the target thickness at a 0.50 growth fraction. Chem film per MIL-DTL-5541 is listed for comparison, it is a chromate conversion coating rather than an anodic film and adds no usable dimension.
| Base Alloy | Growth Fraction FG | Growth Per Surface at 0.0020 in (in) | Coating Behavior |
|---|---|---|---|
| 6061-T6 / T651 | 0.50 | 0.00100 | Standard 50/50 conversion, best all round anodize response |
| 7075-T651 | 0.50 | 0.00100 | Good, comes out darker than 6061, thick hard coat is harder to build |
| 7075-T7351 | 0.50 | 0.00100 | Same coating behavior as T651 |
| 7050-T7451 | 0.50 | 0.00100 | Same family behavior as 7075, expect a darker result |
| 5052-H32 | 0.50 | 0.00100 | Excellent bright clear anodize |
| 5083-H321 | 0.50 | 0.00100 | Good response, slightly grey cast |
| 2024-T351 | 0.40 | 0.00080 | High copper, more penetration than growth, poor hard coat quality and limited thickness |
| 5000 series cast tool and jig plate | 0.50 nominal | 0.00100 | Unpredictable. Blotchy, lot to lot color variation, porosity can trap acid. Prefer ALCA-5 class over Mic-6 class, or move to wrought 6061-T651 |
Growth fraction is the share of total coating thickness that sits above the original surface. MIL-A-8625 does not publish a per-alloy split, so treat the 0.40 on 2024 as a shop rule of thumb driven by its copper content, not a specification value.
| Feature | Change | Pre-Anodize Size | Change at 0.0020 in Hard Coat (in) |
|---|---|---|---|
| Outside dimension across two coated faces | +2 x g | target - 2 x g | +0.0020 |
| Bore or hole diameter | -2 x g | target + 2 x g | -0.0020 |
| Slot width | -2 x g | target + 2 x g | -0.0020 |
| Single face from an uncoated datum | +1 x g | target - g | +0.0010 |
| Internal thread pitch diameter | -4 x g | target PD + 4 x g | -0.0040 |
| External thread pitch diameter | +4 x g | target PD - 4 x g | +0.0040 |
| Thread | 2B PD Min (in) | 2B PD Max (in) | Tolerance Band (in) | Band Eaten by 0.0040 in PD Loss |
|---|---|---|---|---|
| 6-32 UNC | 0.1177 | 0.1218 | 0.0041 | 98% |
| 8-32 UNC | 0.1437 | 0.1475 | 0.0038 | 105% |
| 10-24 UNC | 0.1629 | 0.1672 | 0.0043 | 93% |
| 10-32 UNF | 0.1661 | 0.1697 | 0.0036 | 111% |
| 1/4-20 UNC | 0.2175 | 0.2224 | 0.0049 | 82% |
| 1/4-28 UNF | 0.2268 | 0.2311 | 0.0043 | 93% |
| 5/16-18 UNC | 0.2764 | 0.2817 | 0.0053 | 75% |
| 3/8-16 UNC | 0.3344 | 0.3401 | 0.0057 | 70% |
| 1/2-13 UNC | 0.4500 | 0.4565 | 0.0065 | 62% |
The last column is the pitch diameter loss from a 0.0020" Type III hard coat, which is four times the 0.0010" growth per surface, divided by the full class 2B band. Anything over 100 percent cannot be tapped oversize and still pass a go gauge after coating. Plug the hole for anodize or tap it after the parts come back.
Let t be the target coating thickness in inches and FG be the growth fraction for the base alloy, meaning the share of the coating that sits above the original surface.
The factor of four on threads is the standard thread rule. Pitch diameter is measured across the thread flanks, so the coating on both flanks of both sides of the thread stacks four times along that one measurement. This is why a tapped hole that measured perfect before coating will not take a screw after a hard coat.
The verdict is a tolerance budget. When the remaining machining tolerance is more than half the print tolerance the feature is machinable to print. Between zero and half it is tight but achievable, and you should hold the pre-anodize size hard and confirm the anodize house thickness. At or below zero the coating variation alone is wider than the print tolerance, so mask or plug the feature and finish it after coating.
Chem film per MIL-DTL-5541 is a special case. The film runs about 0.00003 inch, so the dimensional change is negligible on every feature and no pre-anodize compensation is needed.
About 0.001 inch per surface for a standard 0.002 inch Type III hard coat. Anodizing is a conversion coating, so roughly half the coating thickness grows above the original surface and half of it consumes the aluminum below. That means an outside dimension gains about 0.002 inch total, a bore loses about 0.002 inch total, and a single face from an uncoated datum gains about 0.001 inch.
A bore closes by twice the growth per surface, because the coating builds inward from both walls. With a 0.002 inch Type III hard coat the diameter drops about 0.002 inch. With a 0.0003 inch Type II clear it drops about 0.0003 inch. Machine the bore oversize by that amount before the part goes out. For anything held tighter than a thousandth, plugging the bore and reaming it after anodize is cheaper than chasing the fit.
Yes, and by four times the coating growth per surface, not two. Pitch diameter is measured across the thread flanks, so the coating stacks four times along that measurement. A 1/4-20 with a 0.002 inch hard coat loses about 0.004 inch of pitch diameter, which eats roughly 82 percent of the entire class 2B tolerance band. Below about 1/4 inch, plug the hole for anodize or tap it after coating.
Take the final print diameter and add twice the growth per surface. For a 1.0000 inch bore with a 0.002 inch Type III hard coat on 6061, machine it at 1.0020 inch. For a 0.0007 inch Type II dyed coating, machine it at 1.0007 inch. Always work from the actual coating thickness the shop is running, not the range on the spec sheet, because Type III can legally be anywhere from 0.0005 to 0.0045 inch.
Because cast plate has a non-uniform grain structure and inconsistent chemistry between heats, and it can carry porosity that traps acid. The coating grows at different rates across the surface, so the color comes out mottled and it varies lot to lot. If a cast plate part has to be anodized, ALCA-5 class material generally looks better than Mic-6 class, but the reliable answer is to machine appearance parts from wrought 6061-T651 instead.
Yes, but expect a darker result and a harder time building thick hard coat. 7075 and 7050 anodize with roughly the same 50/50 growth-to-penetration split as 6061, so the dimensional math is the same, but the zinc and copper content shifts the color toward grey or bronze on clear coatings and limits how thick a Type III layer will build before it starts to burn. Budget for a color match sample if appearance is on the print.
Anodize compensation only works if the blank starts where you think it does. Get an instant quote on 6061, 7075 and 7050 plate cut to size with mill certs on every order, so the stock you compensate from is the stock you actually get.