Guide
Threads too tight after anodizing or plating: why does it happen and how do you fix it?
Threads too tight after anodizing or plating are a dimensional problem, not a coating defect. The finish adds material to every flank of the thread, and because of the thread angle, that thin layer moves the pitch diameter by several times its own thickness. A screw that ran freely in the bare part then binds, or a GO gauge that entered before finishing stops a turn in.
By Saola Manufacturing engineering team · Published · Updated
This guide is for mechanical engineers and buyers who have received anodized aluminum or zinc-plated steel parts whose threads no longer gauge, or who want to stop it happening on the next order. The rules here come from the governing documents: MIL-A-8625F for anodic coatings on aluminum, ASTM B633 for electroplated zinc on steel, and ASTM F1941/F1941M for plated fasteners. The table converts those published coating thicknesses into pitch-diameter change with the four-times rule, so you can see how much room a given finish needs before you pick a fix.
On this page
Why do threads get tight after anodizing or plating?
Threads get tight because the coating grows outward on an external thread and inward on an internal thread, and the thread angle multiplies that growth at the pitch diameter. Unless the thread was cut with room for the finish, the coated parts no longer fit the class they were machined to.
Picture one flank of a 60-degree thread. A uniform layer on that flank pushes it out along its normal, but the pitch diameter is measured across the part, at right angles to the thread axis. The flank is inclined at half the included angle, so an outward shift of the surface becomes a larger shift of the pitch line. Doing the geometry for both flanks on both sides of the diameter gives the familiar result: for a 60-degree thread, the pitch diameter changes by four times the growth of each surface.
That rule is not shop folklore. Per MIL-A-8625F paragraph 6.10.2, anodic coatings increase the major and minor diameters of a thread by two times the surface growth and the pitch diameter of a 60-degree thread by four times the growth. For threads with another included angle, the same paragraph gives the general form: two times the growth divided by the sine of half the included angle. Plating obeys the same geometry, because a plated layer also sits on each flank.
Two details make it worse than the arithmetic suggests. First, most drawings say nothing about whether the thread applies before or after finishing, and per MIL-A-8625F paragraph 3.9 an anodized part has to meet the drawing dimensions after coating. Second, when both the screw and the tapped hole are coated, their changes add: the external thread grows and the internal thread closes by the same kind of amount, and the fit loses both.
How much does a coating change the pitch diameter?
Pitch-diameter change is four times the surface growth, so it depends on how much of the coating ends up above the original surface. Anodizing grows each surface by about half its film thickness, while electroplated zinc adds its full thickness.
Anodizing converts aluminum into oxide, so the film grows partly into and partly out of the metal. Per MIL-A-8625F paragraph 6.10.1, the increase to expect on each coated surface is about one half of the film thickness. Type II films range from 0.00007 to 0.0010 in and Type III from 0.0005 to 0.0045 in per MIL-A-8625F Table IV, with a Type III nominal of 0.002 in when the drawing does not say otherwise. That is why hard coat is the usual culprit: a nominal hard coat alone moves an external pitch diameter by about 0.004 in per MIL-A-8625F.
Zinc plating is a different mechanism. It is built up on top of the steel, so the whole thickness counts as growth. ASTM B633 sets four minimum-thickness classes, Fe/Zn 5, 8, 12 and 25, where the number is the minimum zinc thickness in micrometers. Because B633 fixes a minimum and not a maximum, the real layer on your part can be thicker, and plating also tends to build more on crests and edges than in the root of a thread. Treat the table below as the least growth to plan for, not the most.
| Finish | Film thickness | Growth per surface | Pitch-diameter change |
|---|---|---|---|
| Type II anodize, thin film | 0.0002 in | 0.0001 in (half the film) | 0.0004 in (about 10 µm) |
| Type II anodize, top of Table IV range | 0.0010 in | 0.0005 in | 0.0020 in (about 51 µm) |
| Type III hard anodize, nominal | 0.002 in | 0.001 in | 0.004 in (about 102 µm) |
| Type III hard anodize, heavy (the 6.10.1 example) | 0.004 in | 0.002 in | 0.008 in (about 203 µm) |
| Zinc plate, Fe/Zn 5 (SC1) | 5 µm minimum | 5 µm (full layer) | 20 µm minimum (about 0.0008 in) |
| Zinc plate, Fe/Zn 8 (SC2) | 8 µm minimum | 8 µm | 32 µm minimum (about 0.0013 in) |
| Zinc plate, Fe/Zn 12 (SC3) | 12 µm minimum | 12 µm | 48 µm minimum (about 0.0019 in) |
| Zinc plate, Fe/Zn 25 (SC4) | 25 µm minimum | 25 µm | 100 µm minimum (about 0.0039 in) |
Which threads are most likely to bind after finishing?
Small, fine-pitch and close-class threads bind first, because their whole pitch-diameter tolerance is small compared with the growth of a heavy finish. Internal threads in blind holes and threads under hard anodize are the next most common trouble spots.
Thread tolerances scale with size and pitch, but coating growth does not: a hard coat grows the same amount on a small screw as on a large one. On small and fine-pitch threads, the pitch-diameter change from a nominal hard coat can be larger than the entire tolerance band of a general-purpose fit, so no amount of care on the lathe rescues a thread that was cut to the standard size.
Class matters too. ASME B1.1 builds an allowance, a deliberate clearance below basic size, into Class 1A and 2A external threads, while Class 3A externals and all B-class internal threads have none. The allowance on a Class 2A thread is small and is there for more than coating, so it can absorb a thin zinc layer on some sizes but not a hard coat. The NASA fastener design manual describes the same split in plain terms: classes differ in tolerance and allowance, and Class 3 is the close-toleranced fit (NASA RP-1228). The ISO metric system does the same job with tolerance positions: an external thread in position g has a clearance, and one in position h does not.
Tapped holes behave unpredictably. Per MIL-A-8625F paragraph 6.10.1.1, the coating inside small holes and tapped holes can vary from no film at all to a full normal coating, and holes over 1/4 inch should be anodized. Plating inside deep blind holes is similarly uneven. The result is a thread that is tight near the mouth and bare deeper down, which is hard to predict from the thickness callout alone.
Should you cut threads oversize, mask them, or chase them after coating?
Mask the threads when they do not need the finish, cut them oversize when they do, and chase them after coating only as a repair. Each option trades corrosion protection against fit and cost in a different way.
Masking. Plugs or caps keep the finish out of the thread, so the thread stays at its machined size. It is the simplest fix when the thread only needs to hold a screw and will be sealed or covered in the assembly. The cost is a bare thread, which matters on aluminum in wet service and on steel anywhere. MIL-A-8625F also cautions designers to require adequate thread and hole sealing in later assembly where corrosion resistance is needed.
Pre-coat oversize. The machinist cuts the external thread smaller, or taps the internal thread larger, by the expected pitch-diameter change, so the part lands inside the class limits once finished. This keeps the finish in the thread and is the right answer for hard-anodized wear threads and heavily plated parts. It needs two numbers agreed in advance: the finish thickness range and which way the thread limits apply. It also needs a stable finishing process, because a thickness range turns straight into a pitch-diameter range.
Chasing after coating. Running a tap or die through the finished thread restores the fit but strips the coating from the flanks, often unevenly. On anodized aluminum the chased thread is bare metal, and a common practice is to touch up the chased flanks with a chromate or trivalent conversion coating so they are not left unprotected. (MIL-A-8625F paragraph 6.10.1.1 makes a narrower version of this point: Type II parts with a total tolerance of 0.0004 in or less that are lapped, honed or stoned to size after anodizing should then be treated with QPL-81706 materials.) On plated steel, chasing leaves bare steel next to zinc. Chasing is a sound way to save a lot of parts already made; it is a poor thing to write on a drawing as the plan.
A fourth route is design: specify a thread insert that goes in after finishing, or put the thread in a part that is not coated. That moves the problem rather than solving it, but on thin anodized sections it is often the cleanest choice.
How should the drawing call out a thread that will be anodized or plated?
State the thread class, the finish with its standard and thickness, and whether the thread limits apply before or after finishing. Then mark every thread that must be masked.
MIL-A-8625F paragraph 6.10.1 recommends putting both a machining dimension and a coated dimension on the drawing. For threads, that usually means one of two notes. The first keeps the class after finishing and hands the compensation to the supplier: 1/4-20 UNC-2B AFTER ANODIZE. HARD ANODIZE PER MIL-A-8625F, TYPE III, CLASS 1. The second masks the thread and says so: MASK ALL TAPPED HOLES. THREADS GAUGE BEFORE AND AFTER FINISH.
For zinc-plated steel, name the ASTM B633 class and passivate, and say whether the plated thread must still accept the class GO gauge. Threaded fasteners are a separate case: ASTM B633 points buyers of fasteners to ASTM F1941/F1941M, which covers application, performance and dimensional requirements for plated fasteners with inch and metric threads and also compares barrel and rack plating (ASTM F1941/F1941M-16, scope). If your part is a fastener, cite that standard rather than B633 alone.
The rest of the finish note, from masking to Cr(VI)-free passivates, is covered in the anodizing vs powder coating vs zinc plating guide. For general tolerance blocks and how they interact with coated features, see the ISO 2768 and NADCA tolerances guide.
How do you check coated threads before the parts ship?
Gauge the threads twice: once after machining against the pre-coat size, and again after finishing against the class limits on the drawing. Only the second check tells you whether the part will assemble.
Thread plug and ring gauges are the practical tool for this, and the GO member is the one that catches coating build-up. A tight GO gauge after finishing, on a thread that gauged cleanly before, points to the finish and not the machining. Recording both results for the first lot tells you which thread sizes need compensation and which can stay at standard size. If the first lot uses a new finisher, new tank chemistry or a new thickness, repeat the two-stage check rather than assuming the last result carries over.
This matters more when the parts are made far from where they are assembled. With parts machined and finished overseas and assembled in the US or Europe, a thread that fails the GO gauge is found at incoming inspection, after freight, when the only fix left is chasing. Asking the supplier to confirm in the quote whether thread limits apply before or after finishing, and how the pre-coat size is set, moves that discovery to the drawing review. The CNC turning and surface finishing pages describe the two steps involved: turning at our lead CNC plant and finishing at partner plants we manage, with the quote naming where each step is done. The RFQ preparation guide lists what else to put in the package, and you can send a drawing for a quote with the thread and finish notes as they stand.
Frequently asked
Why is it four times and not two times the coating thickness?
Because each flank is inclined and the pitch diameter spans both sides of the part. For a 60-degree thread, one over the sine of half the angle is two, and counting both sides of the diameter doubles it again, which gives the factor in MIL-A-8625F paragraph 6.10.2.
Is anodizing growth always exactly half the film?
No. The one-half figure in MIL-A-8625F is design guidance for what to expect. The real split between penetration and build-up depends on the alloy and the process, so confirm it on first-lot parts before relying on it for tight threads.
Can I just switch to a Class 2A thread to make room for plating?
Sometimes, for thin zinc on an external thread. The Class 2A allowance is small and only exists on the external thread, so it will not cover hard anodize, and it does nothing for a coated internal thread.
Does chasing a thread after plating cause hydrogen embrittlement?
Chasing itself does not; the risk comes from the plating process on hardened steel. ASTM B633 advises against zinc electroplating steels with very high tensile strength to that specification, and plated high-strength parts usually need a bake after plating, as the NASA fastener design manual explains.
Should threads in die-cast aluminum be anodized?
Rarely. High-silicon die-casting alloys anodize poorly, so castings are more often powder coated or plated with the threads masked. The ADC12 vs A380 guide covers why.
Sources
- MIL-A-8625F, Anodic Coatings for Aluminum and Aluminum Alloys, US Department of Defense, 10 September 1993, with Amendment 1 of 15 September 2003 (paragraphs 3.9, 6.10.1, 6.10.1.1, 6.10.2 and Table IV)
- ASTM B633-23, Standard Specification for Electrodeposited Coatings of Zinc on Iron and Steel, ASTM International, 2023
- ASTM F1941/F1941M-16, Standard Specification for Electrodeposited Coatings on Mechanical Fasteners, Inch and Metric, ASTM International, 2016
- NASA Reference Publication 1228, Fastener Design Manual, R. T. Barrett, NASA Lewis Research Center, March 1990