A spool leaves the shop with a rainbow beside every weld, and someone asks the question that decides whether the job ships: is that colour acceptable? On carbon steel the answer is comfortable. On Hastelloy C276 (UNS N10276) it is not, because the tint is not a stain sitting on the metal. It is a different material, and the difference is chemical.

This article is written for the moment when the weld is finished and the line is about to enter chemical service. It explains what the tint actually is, where the familiar "straw is acceptable" rule comes from and why it does not transfer to C276, then sets out ten checks and the numbers that should appear on the release certificate.
- Discoloration on C276 is a thin oxide film, roughly 20–200 nanometres thick, formed where hot metal met residual oxygen. Beneath it sits an alloy layer depleted in chromium and molybdenum.
- Colour grades the thickness of the oxide, not the depth of the damage. Straw is thin; blue, violet and grey are thicker. None of them is a reliable measure of how deep the depletion runs.
- For wetted surfaces in chemical service, remove every visible tint. Light straw may remain only on non-wetted external surfaces, and only where the owner specification permits it.
- Ten checks decide release: tint grade, root purge, free iron, crevice geometry, heat input, filler identity, secondary phase, ASTM G28 Method A, finish and rinse, documentation.
- Write the criteria as numbers: no tint on wetted surfaces, no free iron, a stated corrosion rate, and a hardness limit where NACE MR0175 / ISO 15156 applies.
What Is Weld Discoloration on C276, and Why Is It Not Cosmetic?
The tint is a thin chromium-rich oxide sitting on top of a chromium- and molybdenum-poor layer of alloy. The oxide is what you see; the depleted alloy underneath is what fails.
Nickel alloys protect themselves the way stainless steels do. A passive film a few nanometres thick forms instantly on a clean surface and separates the metal from the environment. In C276 that film is built mainly from chromium oxide, with molybdenum and tungsten supporting it. The alloy's reputation in hydrochloric acid, wet chlorine and chloride-bearing process streams rests on that film re-forming after every scratch, cut and weld.
Welding interrupts the process. The arc takes the surface past the temperature at which the film breaks down, and even a good argon shield leaves oxygen in the parts-per-million range. Where that residual oxygen meets hot metal, chromium oxidises first, because it is the most reactive of the alloy's main elements. The oxide that grows is therefore richer in chromium than the alloy beneath it, and the metal left behind is poorer. Molybdenum is consumed at the surface too, in the hottest and most oxygen-rich areas.
The film you see is thin. Heat tints on welded nickel alloys and stainless steels typically run from about 20 to 200 nanometres, which is why they show interference colours instead of a visible coating. The depleted layer is not thin. It extends a few micrometres below the surface in normal welds, and deeper where the arc lingered, the purge was poor or the joint was repaired. The colour tells you how thick the oxide is; it does not tell you how deep the damage goes.

Figure 1. Schematic of a C276 butt joint in cross-section, with the tinted surface and the Cr- and Mo-depleted alloy below it (not to scale).
Electrochemically, that depleted layer is the weak link. It is less noble than the bulk alloy, so in a wet environment it becomes the anode and the bulk metal becomes the cathode. Corrosion starts at the base of the tint and spreads sideways under it, which is why the damage often appears as shallow, wide pits, or as a groove that follows the weld toe, rather than as a neat pinhole. The tint itself also traps moisture and deposits, creating the crevice that C276 is otherwise bought to resist.
There are two defence lines, and welding attacks both. Chromium carries the oxidizing duty: wet chlorine, hypochlorite, chlorine dioxide, the oxidizing zones of a scrubber. Molybdenum carries the reducing duty: hydrochloric and sulfuric acids, phosphoric acid, formic and acetic acid, and resistance to chloride pitting. A tinted surface is short of both.
One clarification prevents a great deal of confusion: discoloration is not sensitization. Tint is a surface phenomenon, a few micrometres deep at most. Sensitization is a through-thickness, grain-boundary phenomenon caused by time spent in the 650–1100 °C band, where intermetallic phases such as mu phase precipitate. Pickling removes the first. Nothing except a full solution anneal at 1120–1180 °C followed by a rapid quench removes the second. A joint can be beautifully bright and still be metallurgically compromised, which is why the checks below cover the procedure as well as the surface.
Is Any Tint Acceptable, or Must It All Be Removed?
For any surface that will be wetted by the process, remove every visible tint. Light straw may be left on non-wetted external surfaces, and only where the owner specification says so in writing.
The widely quoted rule that light straw and pale blue tints are acceptable comes from austenitic stainless steel practice. It was developed for 304 and 316 in atmospheric, mildly corrosive or sanitary service, where a thin tint costs some pitting resistance but rarely decides the life of the equipment. Carried over to C276, it loses its logic for three reasons.
- The environment is different. C276 is normally specified because the medium is aggressive. In hydrochloric acid, wet chlorine, hypochlorite or a flue-gas desulfurization slurry, the margin between passive and pitting is narrow, and a surface that has lost chromium and molybdenum has already spent part of that margin before commissioning.
- The alloy's balance is different. C276 runs 14.5–16.5% chromium, lower than 316L, and depends on 15–17% molybdenum and 3–4.5% tungsten for its performance. Depleting the surface hits the elements the alloy leans on hardest.
- The acceptance charts are different. Stainless tint cards were built for stainless steels. A card that says "straw equals acceptable" is a statement about 304 or 316 in a defined environment, not a universal pass.
The practical consequence is visible in specifications. Chemical plant owners increasingly write "no heat tint on any wetted surface", with no colour chart attached. Where the specification is silent, the defensible default is removal. A supplier who waives tint because the customer did not mention it is taking a risk that will be discovered at the first shutdown inspection.
Table 1. Heat tint grades on C276 and the default action in chemical service
|
Tint grade |
What it looks like |
Oxide film |
Default action on a wetted surface |
|
No tint |
Uniform metallic grey, no iridescence |
None, or an invisible passive film |
Accept, subject to the free-iron test |
|
Light straw |
Faint yellow sheen, iridescent at an angle |
Thinnest visible film, tens of nm |
Remove; may remain only on non-wetted surfaces where the specification allows |
|
Brown or bronze |
Even brown, matt |
Thicker, hundreds of nm |
Remove; do not accept on wetted surfaces |
|
Purple or violet |
Purple-to-blue rainbow |
Thick |
Remove; review the heat-input and shielding records |
|
Dark blue or blue-grey |
Deep blue, low sheen |
Thick |
Remove; suspect a loss of shielding or purge |
|
Grey, black or "sugar" |
Sooty, granular, sometimes dendritic |
Heaviest |
Reject the joint; grind out and re-weld rather than pickle |
Read the table with one caveat. Colour is a proxy for oxide thickness, not for depth of depletion, and lighting, surface finish and camera white balance all shift what the eye sees. A pale tint on a heavily ground surface can hide a thicker oxide than its colour suggests, and a blue spot left by a brief tack weld can be thinner than the straw left by a long, slow pass. That is why the checks below do not stop at "looks acceptable".
Which Ten Checks Should Be Completed Before Chemical Service?
The ten checks fall into four groups: what you can see, what you can test on the surface, what the records prove, and what a corrosion coupon proves. Work through them in order; each answers a question the others cannot.
The first four can be carried out with a torch, a borescope and a test kit. The next four need documents and, in two cases, a laboratory. The last two turn the findings into a decision.
Grade the tint by location and extent, not by impression
Record the colour grade, the band width and the exact surface, weld cap, toe, heat-affected zone, root or outside diameter, for every joint, and photograph it with a scale.
Discoloration is a process fingerprint. A narrow tint confined to one tack weld is a local defect that can be ground out and re-welded. A continuous straw band running the length of a spool is evidence that the shielding, travel speed or purge practice was wrong across the whole run, and grinding every joint will not fix the cause. Grading also creates the baseline against which the post-cleaning inspection is judged, because "clean" has to mean something specific on this job.
Use consistent lighting, a tint comparator card if the project has one, a borescope for the inside diameter, and a joint map keyed to the isometric so that every record points to a weld number rather than to "the second spool".
Inspect the inside diameter before anything else
If the root is oxidised, the joint fails regardless of how good the outside looks: the process wets the root first, and it is the one surface nobody can reach with a brush.
Root oxidation, the sooty granular deposit welders call sugaring, forms when the purge gas contains more oxygen than the alloy can tolerate. For nickel alloys in critical wetted service, purge oxygen is commonly held at or below 100 ppm, and often at 50 ppm, with a low dew point and a flow that maintains positive pressure until the joint has cooled.
Demand the evidence rather than the assurance: the purge gas certificate or analyser log, the plug or dam design, the flow rate, the point at which the purge was released, and borescope images of the root. "We always purge" is not evidence. A low-oxygen start that drifted when the plug was disturbed halfway through the pass is exactly the case the record is meant to catch.
In small bore and long runs an oxidised root cannot be pickled to acceptance. It is ground out, or cut out and re-welded, which is why the root check belongs before the finishing work, not after it.
Prove there is no free iron
Run a ferroxyl test on the cleaned weld and heat-affected zone; a blue stain means embedded iron, and the joint goes back for re-cleaning.
Free iron arrives quietly: a carbon steel wire brush borrowed for one pass, a grinding disc that previously cut mild steel, a wire wheel on a shared bench, chains and slings, a linisher, or spatter from work nearby. The particles embed in the relatively soft nickel surface, rust in moist air, and their corrosion products acidify the local environment and start attack on the alloy they sit on.
The ferroxyl test, potassium ferricyanide in a dilute acid solution, is the practice described for stainless parts in ASTM A380 and is applied to nickel alloys in the same way. It is fast, inexpensive and unambiguous when the surface has been cleaned first. Test after cleaning, again after any rework, and again after the hydrostatic test if the water was not chloride-free.
Look for the crevices the tint hides in
Crevice geometry decides where the first pit forms, so reject undercut, incomplete fusion, excessive convexity, spatter and any backing ring or lap left in place.
C276's molybdenum is what allows it to survive chlorides inside a crevice, where oxygen is consumed, the pH falls and chloride concentrates. Remove molybdenum and chromium from the crevice wall and the alloy loses exactly the margin that the crevice demands. A tinted crevice is therefore worse than either defect on its own.
Inspection points: toe angle and undercut depth against the project's visual acceptance criteria, usually ASME B31.3 for process piping or a stricter owner specification; stop-start craters; arc strikes outside the joint; spatter that will sit under insulation and hold moisture; weld ripple depth on the inside diameter; and any temporary attachment that was welded on and cut off, leaving a crater or a lap.
A backing ring left in place is a permanent crevice. It should be removed, or the joint redesigned to avoid it.
Verify the heat input and interpass records
Ask for the actual numbers: heat input within roughly 0.5–1.5 kJ/mm, interpass temperature at or below 150 °C and preferably 93–120 °C for critical acid service, stringer beads, and no intermediate stress relief.
Heat input and interpass temperature control the width of the heat-affected zone, the thickness of the tint, and the time the alloy spends in the 650–1100 °C window where mu and other topologically close-packed phases precipitate. A log that shows a comfortable interpass temperature alongside a very slow travel speed on thick plate is telling you something different from what the headline number suggests.
Records worth seeing: the WPS, the welder's parameter log with current, voltage and travel speed, temperature readings at the start of each pass, and the gas type and flow. Where a data logger exists, its output is stronger evidence than a hand-written sheet.
Also confirm there was no "helpful" stress relief. A cycle in the 600–700 °C band, routine for carbon steel and harmless there, is actively damaging to C276 because it sits inside the precipitation window.
Confirm the filler is ERNiCrMo-4, and that the certificate is real
Check the filler certificate against AWS A5.14 / ASME SFA-5.14 for ERNiCrMo-4 wire or AWS A5.11 / ASME SFA-5.11 for ENiCrMo-4 electrodes, then PMI the finished weld. A spotless surface on the wrong filler is still a failing joint.
A 316L or ERNiCr-3 (Inconel 82) filler drops molybdenum from 15–17% to almost nothing. The weld becomes the least noble part of the assembly and corrodes first, usually at the weld toe where dilution from the parent metal is highest. The corrosion test in check 3.8 will often catch this, but only if it is run on a coupon made with the production filler.
Positive material identification by X-ray fluorescence confirms chromium, molybdenum, tungsten and iron within seconds, but it cannot see carbon, which is the element that governs sensitization. Carbon comes from the mill certificate, 0.010% maximum in the base metal and 0.020% in ERNiCrMo-4, or from a combustion analysis if the certificate is disputed.
Traceability matters as much as chemistry: heat number on the pipe or fitting, filler lot number on the weld map, and an EN 10204 type 3.1 certificate for both. A certificate that cannot be tied to the joint in front of the inspector is decoration.
On heavy sections and repairs, check for secondary phase
Where the wall is thicker than about 25 mm, or after any repair, require metallographic or hardness evidence that no mu or sigma phase formed. Removing the tint cannot undo it.
The 650–1100 °C precipitation window is reached in the heat-affected zone of thick sections, in multipass welds with poor interpass control, during flame straightening, and in hot forming that was not followed by a solution anneal. C276's low carbon and silicon make it resistant to this, not immune, and repairs are the usual place where the margin is lost.
Evidence options, in increasing order of cost: a hardness survey across the heat-affected zone, where a rise above the parent-metal value is a flag rather than a proof; metallographic replication; a coupon cut from the same procedure and tested to ASTM G28 Method A; or destructive sectioning of a witness piece. A magnet is not a phase test and should not be presented as one.
If the phase is present, the only remedy is a full solution anneal at 1120–1180 °C with a rapid quench. An intermediate stress relief will make the microstructure worse, not better.
Run the corrosion qualification, and read it correctly
Require an ASTM G28 Method A result for the production procedure, with the weld and the parent metal reported separately.
ASTM G28 Method A is the standard screening test for intergranular susceptibility in nickel-rich, chromium-bearing alloys: a boiling ferric sulfate–50% sulfuric acid exposure. C276 is exposed for 24 hours rather than the 120 hours used for slower-corroding grades, because its corrosion rate in that solution is high enough to show a difference within a day.
Acceptance is set by the owner specification, not by the test method. A common criterion for critical chemical service is a corrosion rate at or below 0.5 mm/year, with the weld no worse than the parent metal. Quoting a rate without saying which surface it came from, and how that surface was cleaned, is the most common way the test gets misused.
Two limits are worth keeping in mind. G28 measures metallurgical susceptibility through the section; it does not measure the effect of a surface tint, so a bright, fully pickled coupon can pass while production welds are left discoloured. And for chloride pitting or crevice duty, G28 alone is not enough: add an ASTM G48 ferric chloride test at the service temperature when the specification calls for it.
Measure the finish and the rinse water
Specify and measure roughness on wetted surfaces, and rinse with low-chloride water. A perfect pickle followed by a dirty rinse is a wasted pickle.
Roughness governs how long a corrosive liquid stays in contact and how easily a crevice forms. Many owner specifications call for Ra of 3.2 µm or better on wetted surfaces, and 6.3 µm or better elsewhere, with the weld cap dressed flush where flow, cleaning pigs or inspection demand it.
Chloride is the other residue. Pickling pastes, tap water, hydrotest water, marker pens, adhesive labels and chloride-bearing cutting fluids all leave chloride behind. Final rinses are commonly specified with demineralised or low-chloride water, frequently below 50 mg/L chloride, followed by drying with oil-free air.
Measure rather than assume: a profilometer or replica tape for roughness, chloride strips or a titration kit for the rinse water, and a written record of when the surface was dried and protected.
Close the documentation trail
A weld is released when the paper matches the metal: WPS and PQR to ASME IX, welder qualifications, EN 10204 3.1 certificates, NDT reports, cleaning and test records, and a signature.
For sour service the trail gains another layer. NACE MR0175 / ISO 15156 governs material selection in H2S-containing environments. C276 is listed for the solution-annealed condition with a maximum hardness of 35 HRC, and the environmental limits for UNS N10276 must be checked against the standard's tables for the actual temperature, H2S partial pressure and chloride level.
The minimum package: material certificates tied to heat numbers; filler certificates tied to lots; WPS and PQR; welder and operator qualifications; heat-input and interpass logs; NDT reports; cleaning records; the corrosion test report; hardness results where applicable; dimensional records; and a release certificate signed by someone whose name appears on it.
Table 2. The ten checks, the evidence to look for and the usual acceptance point
|
Check |
Evidence to see |
Typical acceptance |
|
1. Tint grade and location |
Joint map, photographs with a scale, borescope video |
No visible tint on wetted surfaces; band width recorded |
|
2. Root condition |
Purge log (oxygen, dew point, flow), borescope images |
Bright root, no sugar; purge oxygen ≤100 ppm, ≤50 ppm for critical work |
|
3. Free iron |
Ferroxyl test on the cleaned weld and HAZ |
No blue stain at any tested location |
|
4. Crevice geometry |
Visual at 10×, penetrant testing where specified, weld profile |
No undercut beyond the project limit, no incomplete fusion, no backing ring left in place |
|
5. Heat input and interpass |
WPS, parameter log, temperature readings |
0.5–1.5 kJ/mm; interpass ≤150 °C, 93–120 °C preferred; no 650–1100 °C exposure |
|
6. Filler identity |
Lot certificate, PMI of the weld, weld map |
ERNiCrMo-4 or ENiCrMo-4; Cr, Mo and W within specification; carbon per certificate |
|
7. Secondary phase |
Hardness survey, replication, coupon test |
No hardness excursion; metallography free of continuous grain-boundary phase |
|
8. Corrosion qualification |
ASTM G28 Method A report and the coupon cleaning record |
Owner criterion, commonly ≤0.5 mm/y, weld no worse than parent metal |
|
9. Finish and rinse |
Profilometer or replica, chloride test, rinse record |
Ra per specification, commonly ≤3.2 µm wetted; low-chloride rinse, then dried |
|
10. Documentation |
Complete package and a signed release |
WPS/PQR, welder qualifications, EN 10204 3.1, NDT, cleaning and test records; NACE evidence where sour |
How Should the Tint Be Removed Without Damaging the Alloy?
Remove it chemically with a nitric–hydrofluoric pickle or mechanically to bright metal. Do not passivate over it, do not pickle with hydrochloric acid, and do not let carbon steel tooling touch the joint.

Chemical removal. A nitric–hydrofluoric pickle, commonly around 20% HNO3 with 5% HF, dissolves the oxide and the depleted layer together, typically in 30–60 minutes depending on temperature and how heavy the tint is. It is applied as a paste on site or as a bath in the shop, then neutralised and rinsed thoroughly. HF is hazardous: the process needs proper personal protective equipment, ventilation, training and waste handling, and the rinse water needs a chloride check before it is accepted.
Mechanical removal. Where pickling is not permitted, dedicated stainless or nickel-alloy brushes, carbide burrs and iron-free abrasives take the surface back to bright metal. Dedicated means dedicated: a brush that has touched carbon steel is a contamination source. Grinding should remove metal rather than smear it; a disc polished over a tint can fold the depleted layer into the surface and leave a surface that looks bright and tests badly.
Two treatments that do not work. Nitric-acid passivation alone is a stainless-steel treatment that rebuilds a passive film on an already clean surface; on a high-molybdenum nickel alloy it does not remove the depleted layer, it seals the defect under a film. Hydrochloric-acid pickling is worse: HCl is the medium C276 is bought to resist, and using it as a cleaning acid deposits chloride into exactly the crevices the alloy cannot tolerate.
Table 3. Cleaning methods for C276 weld tint, and what each can and cannot do
|
Method |
Best used for |
Limits and precautions |
|
Nitric–hydrofluoric pickle paste |
Field welds, local tint, shop touch-up |
Around 20% HNO3 with 5% HF for 30–60 min; HF hazard; neutralise and rinse thoroughly |
|
Immersion pickle |
Workshop spools and fittings needing uniform coverage |
Bath strength and temperature must be controlled; over-pickling etches the toe and grain boundaries |
|
Mechanical removal (brush, burr, iron-free abrasive) |
Where pickling is not permitted; heavy scale |
Dedicated tools only; remove metal rather than smear it; avoid grinding burns |
|
Electrolytic cleaning |
Reproducible shop finish on simple geometry |
Electrolyte control required; chloride-free rinse afterwards |
|
Nitric or citric passivation alone |
Rebuilding the film on an already clean surface |
Does not remove tint or the depleted layer; not a substitute for pickling |
|
Hydrochloric acid |
Nothing on C276 |
Deposits chloride, attacks the alloy, and creates the crevice conditions to be avoided |
After cleaning, rinse with low-chloride water, dry, and protect the surface. Then re-run the ferroxyl test, because cleaning is also the step at which contamination is most easily reintroduced.
How Do You Prove the Surface Is Clean Enough to Release?
Release on three independent evidence sets, the surface, the procedure and the records, and never on a single visual pass.

Figure 2. Five gates between a finished weld and a released C276 joint. A "no" at any gate returns the joint to the previous step.
Gates 1 and 2 are production steps. Gate 3 is the surface evidence: a tint-free, uniform, ferroxyl-negative surface with a measured finish and a rinse record. Gate 4 is the procedure evidence: a coupon that passed ASTM G28 Method A under the same procedure, the WPS and PQR, the parameter logs, and hardness where NACE applies. Gate 5 is the signature, and it is worth nothing unless the first four were closed with evidence rather than opinion.
The most common gap in real projects is not a failed test. It is a test that was never asked for. Fabricators routinely ship with dimensional and NDT records and no cleaning record at all, because the purchase order specified the alloy and the dimensions but not the surface condition. If the order does not say "no heat tint on wetted surfaces", the shop will interpret a visible tint as acceptable, and it will be right about the contract and wrong about the service.
Re-inspect after transport and after site welding. Field welds are usually made under worse conditions than shop welds, with wind, moisture, restricted access and no trailing shield, and they are the joints most likely to be tinted and least likely to be borescoped.
What Errors Cause Most Early Failures in Chemical Service?
Six mistakes account for most premature C276 failures, and five of them leave a surface that still looks acceptable at handover.
- Blending the tint instead of removing it. A flap disc that polishes the colour away leaves the depleted metal in place and often smears it across a wider area. The failure appears at the toe, usually within the first year or two of wet service.
- Importing the stainless rule. "Straw is fine" is a statement about 304 in a benign environment. Applied to a hypochlorite line, it removes the margin the alloy was chosen to provide.
- Trusting colour instead of records. A pale tint after a slow, hot pass can conceal deeper depletion than a blue tint after a fast one. Without the heat-input and purge records there is no way to tell the two apart.
- Ignoring the root. Outside-diameter inspection is easy, and the inside is where the flow is. An oxidised root produces the first leak, and it produces it from the inside, where the repair is most expensive.
- Re-contaminating a clean surface. Carbon steel slings, chloride marker pens, tap water rinses and hydrotest water left standing all undo a correct pickle within minutes.
- Applying "helpful" stress relief. A 650–700 °C cycle for dimensional stability puts the whole assembly into the precipitation window and can ruin a joint that was welded perfectly.
What Should the Release Certificate Say?
Write the criteria as numbers, by service severity, and sign against them. A certificate that says "welds cleaned and inspected" protects nobody.
Table 4. Acceptance criteria by service severity (typical owner-specification values; the project specification governs)
|
Service tier |
Tint on wetted surfaces |
Free iron |
Corrosion evidence |
Finish and rinse |
|
A. Non-wetted or atmospheric |
Light straw may remain if the specification allows; loose scale removed |
Negative at transitions and coated areas |
ASTM G28 Method A on the procedure |
Not normally specified; dry and protected |
|
B. Mild wetted (cooling water, dilute neutral streams) |
None visible |
Negative |
ASTM G28 Method A to the owner criterion |
Ra ≤6.3 µm; low-chloride rinse and dry |
|
C. Severe chemical (HCl, H2SO4, wet chlorine, hypochlorite, FGD slurry) |
None on any wetted surface, including the root |
Negative after the final clean |
G28 A at or below 0.5 mm/y, weld no worse than parent; G48 where pitting duty applies |
Ra ≤3.2 µm; cap dressed; final rinse below 50 mg/L chloride, then dried |
|
D. Sour service (H2S) |
None visible |
Negative |
G28 A plus a hardness survey |
As tier C; NACE MR0175 / ISO 15156 evidence, solution annealed, ≤35 HRC |
Two notes on using the table. It is a template for a project specification, not a substitute for one: the owner's corrosion engineer sets the final numbers, and the values here are the industry-common ones that most specifications converge on. And the tiers are defined by the wetted environment, not by the alloy. A C276 spool in a dry, insulated location and a C276 spool in a hypochlorite line can come off the same drawing, but they should not carry the same acceptance criteria.
Frequently Asked Questions
Does weld discoloration really reduce C276's corrosion resistance?
Yes. The tinted zone is a chromium- and molybdenum-depleted layer, not a surface stain. In aggressive chemical service that zone corrodes first and often several times faster than the parent metal; the exact multiple depends on the medium, the temperature and the chloride level. It is for this reason that owner specifications for wetted C276 surfaces normally require the tint to be removed rather than judged.
What colour of heat tint is acceptable on C276?
On wetted surfaces, none. On non-wetted external surfaces, light straw may be accepted where the specification allows it and loose scale has been removed. Blue, violet, grey and black tints indicate thicker oxide and should be removed wherever the surface can be wetted by the process.
Can I pickle the weld and skip the other checks?
No. Pickling addresses the surface layer only. It does not correct an oxidised root, a wrong filler, a crevice at the toe, a sensitized heat-affected zone or embedded iron. Pickling is check 4 of the cleaning step, not the whole release.
Is pickling required after every C276 weld?
For wetted surfaces in chemical, marine or high-chloride service, yes, or an equivalent mechanical removal to bright metal. For non-wetted structural welds, the specification may allow a light tint to remain. The decision belongs in the specification, not in the shop.
Can nitric acid passivation replace pickling?
No. Nitric passivation is a stainless-steel treatment that rebuilds a passive film on a clean surface. On C276 it does not remove the depleted layer, and nitric acid is not the environment the alloy is selected for. It is acceptable as a final step after pickle removal, not instead of it.
How deep does the damage go?
The oxide itself is tens to a couple of hundred nanometres thick. The depleted alloy beneath it typically extends a few micrometres, and deeper where heat input or oxygen access was high. The colour grades the film, not the depth, which is why records and tests carry more weight than appearance.
Does post-weld heat treatment fix a discoloured weld?
No. Heat treatment does not remove surface oxide, and the intermediate stress relief often requested for dimensional stability, in the 600–700 °C band, makes the metallurgy worse because it sits inside the 650–1100 °C precipitation window. C276 is normally used as-welded; if heat treatment is genuinely required, only a full solution anneal at 1120–1180 °C with a rapid quench is valid.
What single test best predicts service performance?
No single test does. For the surface, a tint-free visual inspection plus the ferroxyl test; for the procedure, ASTM G28 Method A on a coupon cleaned the same way as production. A project that relies on one and skips the other is guessing.
How quickly can a tinted weld fail?
There is no fixed timeline. In wet chlorine, hypochlorite or hot chloride-bearing streams, pitting can begin within weeks of commissioning; in dilute, cooler or less oxidizing streams the same defect may take years to appear. The absence of an early leak is not evidence that the surface was acceptable.

