Guide
How do you check 6061-T6 hardness and temper on a finished aluminum part?
6061-T6 hardness is typically about 95 on the Brinell scale (500 kgf load on a 10 mm ball), and together with electrical conductivity it is the quickest physical evidence that a part made from 6061 is really in the solution-treated and artificially aged T6 condition, not annealed (O), only naturally aged (T4) or softened by heat after it left the mill.
By Saola Manufacturing engineering team · Published · Updated
This guide is for design and quality engineers who receive machined aluminum parts and want a fast, defensible answer to one question: is this actually T6? Paperwork answers what the mill shipped. The checks below answer what is in your hand. They cover which test to use on a finished part, how to read hardness and conductivity together, what quietly changes a temper after the bar is cut, and the drawing and RFQ wording that makes the check part of the contract instead of an argument at receiving.
On this page
Why can’t you see a wrong temper on a machined 6061 part?
Because 6061 bar in the O, T4 and T6 conditions machines to the same dimensions, takes the same anodize and looks identical on the bench. A wrong temper only shows up as a property: lower yield strength, threads that strip early, or a bracket that takes a set in service.
The temper letters describe heat treatment, not geometry. T6 means solution heat treated, quenched and artificially aged; T4 stops after natural aging at room temperature; O is fully annealed. The difference in strength is large. MIL-HDBK-5J lists a 35 ksi minimum tensile yield for 6061-T6 bar, and annealed 6061 is typically only a fraction of that. A dimensional report will pass all three conditions without comment.
The mill test report and the certificate of conformance are the first line of defense, and they are covered in detail in how to read a mill test report. What paperwork cannot prove is that the bar on the certificate is the bar your parts were cut from, or that nothing heated the parts afterward. For the meaning of T6 versus T651 and T6511, see 6061-T6 vs T651; this guide assumes you already know which temper you specified and want to confirm you got it.
What is the typical 6061-T6 hardness, and how do other tempers and alloys compare?
Typical 6061-T6 hardness is about 95 HB, with an electrical conductivity of about 43% IACS. Annealed 6061-O is typically near 30 HB and T4 near 65 HB, and none of these are specification minimums, because the ASTM bar and extrusion specifications accept 6061 on tensile results.
That last point matters when you write an acceptance limit. Published hardness and conductivity figures for aluminum are typical values from handbook and producer data, so a reading a little below the typical number is not, on its own, a nonconformance. Aerospace practice handles this with SAE AMS 2658, which sets hardness and conductivity acceptance criteria for wrought aluminum parts; outside aerospace, the drawing or purchase order has to state the limits, or the reading has nothing to be judged against.
Rockwell numbers cause the most confusion. A typical 6061-T6 part reads around 60 on the Rockwell B scale, but labs also test aluminum on the E or F scale, or on a superficial scale for thin walls, and the same part gives a different number on each. Conversions between Brinell, Rockwell and Webster readings on aluminum are approximate, so name the scale you will accept instead of relying on a converted value.
| Alloy and temper | Brinell hardness, HB | Conductivity, % IACS at 20 °C |
|---|---|---|
| 6061-O (annealed) | 30 | 47 |
| 6061-T4 (naturally aged) | 65 | 40 |
| 6061-T6 / T651 | 95 | 43 |
| 6063-T6 | 73 | 53 |
| 2011-T3 | 95 | 39 |
| 2024-T3 | 120 | 30 |
| 7075-T6 / T651 | 150 | 33 |
Why does a temper check need both hardness and conductivity?
Because neither reading identifies the condition by itself: 6061-T6 and 2011-T3 share about the same typical hardness, and 6061-T6 sits between T4 and O on conductivity, so one number can look right on the wrong material. Read as a pair, they separate annealed, under-aged, over-aged and wrong-alloy parts, which is the logic behind SAE AMS 2658.
The physics explains the pattern. Conductivity in a heat-treatable aluminum alloy depends mostly on how much of the alloying content is still dissolved in the aluminum; in 6061 that is magnesium and silicon. Right after solution treatment and quench, nearly all of it is in solution, so a T4 part has the lowest conductivity of the three. Artificial aging pulls those elements out as fine precipitates, which raises both strength and conductivity, and T6 sits at the peak of strength. Keep heating past that peak and the precipitates coarsen: conductivity keeps climbing while hardness falls. Full annealing is the extreme of that path, so an O-temper part reads soft and high.
The practical consequence is that a conductivity meter alone cannot tell you which side of T6 a part has fallen to, and a hardness tester alone cannot tell 6061 from another alloy of similar hardness. The table below shows how to read the two together on a part that was sold as 6061-T6.
| Hardness | Conductivity | Most likely condition | Next step |
|---|---|---|---|
| At or near typical | At or near typical | T6 as specified | Accept and record readings |
| Well below typical | Above typical | Annealed (O), heavily over-aged by heat exposure, or 6063 bar | Quarantine; check for welding, coating cure or wrong bar |
| Below typical | Below typical | T4: solution treated but not artificially aged | Quarantine; check the cert and any re-heat-treat history |
| Slightly below typical | Slightly above typical | Mild over-aging, often from a coating cure or rework | Retest at valid locations; compare with untreated parts of the lot |
| Near typical | Outside the 6061-T6 range | Different alloy (for example 2011-T3, which reads lower) | Confirm alloy by lab chemistry before anything else |
Which hardness test should you use on a finished part?
Use Rockwell to ASTM E18 or Brinell to ASTM E10 on a non-functional surface when the section is thick enough, a Webster gage to ASTM B647 for thin walls and extrusions, and microhardness on a sectioned sample when every surface is functional or the part is too small.
Every indentation test leaves a mark, so the first decision is where the mark is allowed. Mark a test location on the drawing, or agree that one part per lot is sacrificed, or accept a coupon cut from the same bar and processed with the parts. Any of those is better than an inspector choosing a sealing face because it happens to be flat.
Validity depends on geometry more than on the instrument. ASTM E18 requires the piece to be thick enough that the test leaves no mark on the underside, sets minimum spacing from edges and from earlier indents, and gives correction factors for convex cylindrical surfaces, which is where small turned parts usually go wrong. The part must sit solidly on the anvil; a part rocking on a burr or supported only at its ends reads low. ASTM E10 Brinell testing with a large ball averages over more grain and is the reference method for many handbook values, but its indentation is too large for most small machined parts.
The ASTM B647 Webster gage is a hand-held clamp tester built for aluminum sheet, tube and extruded sections. It is fast and portable, which makes it a good receiving-dock screen, but its range of section thickness is limited and its scale is its own, so record Webster readings as Webster readings. For a sectioned sample, a Vickers or Knoop microhardness test on a polished cross-section avoids marking a delivered part at all and can also show whether a surface layer differs from the core.
How do you get a conductivity reading you can trust?
Use an eddy-current conductivity meter calibrated per ASTM E1004 against reference standards that bracket the expected value, and take the reading on a flat area larger than the probe, away from edges, with part, probe and standards at the same temperature.
Conductivity testing is non-destructive and fast, which is why it is the better tool for screening a whole lot or a mixed bin. The catch is that the meter measures a small volume under the probe, so the conditions of the reading decide whether it means anything. ASTM E1004 covers the method, including the use of reference standards, temperature effects and lift-off. Conductivity changes with temperature, so a part carried in from a cold truck or straight off a machine reads differently from one that has sat beside the standards.
Thickness and coatings are the next traps. At the frequencies portable meters typically use, the eddy currents in 6061 reach only a fraction of a millimeter deep, so a thin wall can read low because the field runs out of metal. Anodize and paint are non-conductive and act as lift-off; meters compensate for lift-off only up to a stated coating thickness, so check that limit against the anodize or paint thickness on the drawing. Metallic plating is worse, because the meter partly reads the plating instead of the aluminum. On a plated or heavily coated part, take conductivity readings before finishing, at first article, or on an unplated witness piece.
Small turned parts add curvature. A standard flat probe on a small diameter reads low, so use a small-diameter probe, a flat machined into a test location, or a coupon from the same bar. This is also the point to decide on finishing sequence: if your part is turned from bar and then anodized, readings before and after anodizing should agree closely once lift-off is compensated, because anodizing is an electrochemical process that does not heat the part to aging temperatures; a meaningful shift points to a measurement problem, not a temper change.
For context on parts made here: the CNC turning page lists 6061-T6 and 2011-T3 among the bar grades Saola turns at its lead CNC plant; anodizing on aluminum is applied after turning by a partner finishing plant we manage, and the quote names where each step is done. Heat treatment does not appear on any of the capability pages, so the temper of a turned 6061 part here is the temper of the bar it was cut from, and material certification for that bar is available on request per the quality page.
What can change the temper after the bar leaves the mill?
Machining does not change it, but heat can: welding, hot straightening, a re-heat-treatment with a slow quench, and coating cure cycles near or above the aging temperature can all leave a 6061 part softer than T6. The most common cause is simpler still: the wrong bar was cut.
Bar mix-ups. 6061-T6, 6061-T4, 6063-T6 and annealed stock look the same on a rack once the end marking is cut off. 6063 is the classic substitute, because it is common, cheaper to extrude and close in name; it is typically far more conductive than 6061, which makes conductivity the fastest way to catch it. An alloy question that the pair of readings cannot settle goes to a lab for chemistry.
Welding. The heat-affected zone next to a weld in 6061-T6 loses most of the aging strength, and the base-metal T6 values no longer apply there. If a welded assembly must meet T6 properties, the drawing has to call for post-weld heat treatment and say how it is verified.
Coating cures and rework. Powder coat and baked paint cure at temperatures above the 6061 aging temperature, for minutes rather than hours. A single cure usually costs little, but strip-and-recoat cycles on thin sections can add up, and the signature is a conductivity rise with a small hardness drop. Hot-water sealing after anodizing runs well below aging temperature and is not a concern; see Type II vs Type III anodize for what the anodizing process does change.
Re-heat-treatment. A part that is solution treated and aged again after delivery can take the T62 designation under the ANSI H35.1 temper system, and its results depend heavily on quench speed. Furnace times and temperatures for aluminum heat treatment are set out in SAE AMS 2770. On finished parts the quench usually distorts features that were already machined, which is why a temper failure on a finished lot normally ends in a remake rather than a rescue.
How should the drawing, purchase order and RFQ call out temper verification?
Name the alloy, temper and product specification in the material block, then add a separate verification note that states the test, the location on the part, the sample size and the result that accepts the lot.
A material callout alone, such as “6061-T6”, makes the supplier responsible for the temper but gives your receiving inspector nothing to measure against. A workable drawing note looks like this:
MATERIAL: ALUMINUM 6061-T6 BAR PER ASTM B211 (T651 ACCEPTABLE).
TEMPER VERIFICATION: HARDNESS AND ELECTRICAL CONDUCTIVITY PER SAE AMS 2658, SAMPLE SIZE PER PO, ON SURFACE MARKED “HT”. RECORD READINGS AND SCALE ON THE INSPECTION REPORT.
INDENTATION NOT PERMITTED ON SURFACES MARKED A AND B.
If you are not working to AMS 2658, replace that reference with your own limits: a minimum hardness on a named scale and a conductivity range, both taken from your design margin rather than copied from a typical-value table. Put the sample size on the purchase order, using the same plan you use for dimensional sampling; AQL sampling for machined parts covers how to choose it.
On the RFQ, answer five questions up front: which alloy, temper and product specification; whether readings are required on every lot or only at first article; who takes them (the supplier, a third-party lab, or your own receiving inspection); where on the part a mark is allowed; and whether a coupon from the same bar is acceptable for a part with no free surface. If readings will be reported, ask for them on the first article inspection report alongside the dimensional results, and tie the bar to the parts through the heat or lot number on the certificate of conformance.
Saola’s quality page lists material certification, first-article inspection reports and dimensional records among the documents it supplies, and does not list a hardness or conductivity test. If your drawing requires temper verification readings from the supplier, say so in the RFQ, the same way the surface finishing page asks buyers to flag any test standard outside its menu before tooling, so the answer comes before the quote rather than at first article. You can send the drawing and note through the quote form.
Frequently asked
Is 6061-T651 hardness different from 6061-T6?
No meaningful difference is expected. T651 is T6 material that the mill also stretched to relieve quench stress before aging, and handbooks list the same typical hardness and the same strength minimums for both. Stretching changes how much the part moves when you machine it, not what a hardness tester reads. The details are in 6061-T6 vs T651.
Can a file test or scratch test tell T6 from annealed 6061?
Not reliably. A file may skate less on T6 than on fully annealed stock, but it cannot separate T4 from T6, it depends on the person holding it, and it produces no number for a report. If the temper matters enough to question, it matters enough for an instrument reading on a named scale.
Does anodizing change the temper or the readings?
Anodizing does not change the temper of 6061. It can change the readings: the oxide layer adds lift-off for a conductivity probe and, on a thick hard-anodize layer, can affect superficial Rockwell results. Take readings before anodizing at first article, or on an unanodized witness piece, so later readings have a baseline.
Should one low hardness reading reject the lot?
Not on its own. First verify the tester on a reference block, then retest at valid locations away from edges, curvature and earlier indents, and take a conductivity reading. If the pair of readings still points to the wrong temper, quarantine the lot and settle it with a tensile test on a sample or coupon from the same bar, because the product specification accepts the material on tensile properties.
Can parts in the wrong temper be heat treated back to T6?
Sometimes, but rarely on finished parts. Solution treatment needs a fast quench that tends to distort machined features, the result can be designated T62 rather than T6, and the parts then need full dimensional and temper re-inspection. For most machined parts, remaking them from verified bar is the cleaner path.
Sources
- ASTM E1004-23, Standard Test Method for Determining Electrical Conductivity Using the Electromagnetic (Eddy Current) Method, ASTM International, 2023
- ASTM E18-25, Standard Test Methods for Rockwell Hardness of Metallic Materials, ASTM International, 2025
- ASTM E10-23, Standard Test Method for Brinell Hardness of Metallic Materials, ASTM International, 2023
- ASTM B647-23, Standard Test Method for Indentation Hardness of Aluminum Alloys by Means of a Webster Hardness Gage, ASTM International, 2023
- SAE AMS2658E, Hardness and Conductivity Inspection of Wrought Aluminum Alloy Parts, SAE International, 2026
- SAE AMS2770R, Heat Treatment of Wrought Aluminum Alloy Parts, SAE International, 2020
- MIL-HDBK-5J, Metallic Materials and Elements for Aerospace Vehicle Structures, Department of Defense, 2003
- ASTM B211/B211M-23, Aluminum and Aluminum-Alloy Rolled or Cold-Finished Bar, Rod, and Wire, ASTM International, 2023