Project
Corrosion Coupon Test Plan and Coating Selection Project
Corrosion coupon project for exposure basis, coupon matrix, mass-loss rate, coating comparison, uncertainty, acceptance criteria, and release evidence.
This project produces a corrosion-control decision package for a carbon-steel equipment frame exposed to coastal industrial weather. The deliverable is not a generic materials note. It is a test plan and coating-selection package that an engineer could take to a design review, maintenance review, or supplier qualification meeting.
The project shows how to define the exposure basis, choose coupon and coating-panel tests, calculate corrosion rate from mass loss, compare coating systems, handle uncertainty, and state release criteria. The goal is to avoid the common mistake of selecting a coating from a datasheet while leaving the real failure mode, inspection method, and acceptance evidence undefined.
Project Objective
Select and validate a corrosion-protection system for a carbon-steel utility skid installed outdoors at a coastal process site.
The final package must contain:
- exposure and consequence basis;
- material and geometry assumptions;
- coupon and coating-panel test matrix;
- mass-loss corrosion-rate calculation;
- dry-film-thickness and holiday-inspection plan;
- coating-system comparison;
- uncertainty and error-budget notes;
- acceptance criteria;
- final recommendation and release evidence.
Engineering Context
The skid supports pumps, small-bore piping, cable trays, and an access platform. It is fabricated from carbon-steel sections with welded brackets, bolted cover plates, drain holes, lifting lugs, and field-cut edges. The owner wants a 15-year design intent with planned inspection every 3 years.
The main corrosion risks are:
- chloride wet-dry cycling from marine aerosol;
- trapped water at horizontal ledges and bolted details;
- damaged coating at lifting points and field-cut edges;
- crevice conditions under bolted cover plates;
- galvanic contact with stainless instrumentation brackets;
- poor coating repair quality after installation.
The design team is choosing among three protection systems:
| Candidate | Description | Engineering concern |
|---|---|---|
| A | single-pack industrial coating | low cost but weak edge and chemical resistance |
| B | zinc-rich epoxy primer, epoxy intermediate, polyurethane topcoat | stronger barrier system but requires controlled surface preparation |
| C | hot-dip galvanizing plus topcoat on exposed faces | robust zinc reserve but harder to repair after field welding |
The test plan should not pretend that an accelerated exposure is a perfect lifetime predictor. Its function is to screen mechanisms, compare candidates under the same exposure basis, and define what evidence is required before release.
Requirements and Acceptance Basis
Use these project requirements:
| Requirement | Acceptance basis |
|---|---|
| intended service life | 15 years with maintenance |
| inspection interval | 3 years |
| maximum uniform steel loss before repair trigger | 0.30\ \text{mm} |
| damaged-area screening corrosion rate | \le 0.015\ \text{mm/year} mean |
| maximum individual damaged-area rate | \le 0.025\ \text{mm/year} |
| coating underfilm creep from scribe | \le 2.0\ \text{mm} |
| dry film thickness | within specified minimum and maximum |
| holiday density after repair | zero holidays on critical edges and weld toes |
| field release | documented surface preparation, DFT, repair, and inspection record |
The corrosion-rate requirement applies to a controlled damaged-area screening coupon, not to every possible field defect. Pitting, crevice corrosion, coating disbondment, mechanical impact, and poor repair can still govern the real structure.
Exposure Basis
Define the exposure before defining the test.
| Exposure item | Project basis |
|---|---|
| atmosphere | coastal industrial outdoor atmosphere |
| temperature | 5 to 40\ \text{deg C} |
| wetting | chloride wet-dry cycling, rain, condensation |
| chemicals | occasional mild alkaline washdown |
| geometry | edges, weld toes, bolted covers, horizontal ledges |
| inspection access | good on outer faces, limited under cover plates |
| consequence | structural support degradation, downtime, repair access cost |
This basis leads to two different test needs:
- uncoated or intentionally damaged coupons to estimate relative corrosion severity;
- coated panels and representative details to check surface preparation, DFT, scribe creep, holiday behavior, edge coverage, and repairability.
Test Matrix
Use replicate specimens so that one coupon does not become the entire evidence base.
| Specimen group | Quantity | Purpose |
|---|---|---|
| bare carbon-steel witness coupons | 3 | exposure severity reference |
| candidate A coated panels with scribe | 3 | low-cost coating benchmark |
| candidate B coated panels with scribe | 3 | barrier and zinc-rich system evaluation |
| candidate C duplex panels with scribe | 3 | galvanized-plus-topcoat evaluation |
| edge-detail panels | 2 per candidate | edge coverage and repair inspection |
| bolted lap-joint details | 2 per candidate | crevice and fastener-detail risk |
| field-repair panels | 2 per candidate | repair adhesion and holiday retest |
Record for each specimen:
- base material heat or batch;
- surface preparation method and profile;
- coating batch, mix ratio, induction time, and cure condition;
- dry film thickness readings;
- artificial scribe or defect geometry;
- exposure start and end date;
- mass before and after cleaning;
- visual defects, blistering, rust staining, creepage, and holidays;
- photographs before exposure, after exposure, and after cleaning.
The test is valid only if the records can connect the result to the process used in production. A coating panel with perfect lab preparation is weak evidence if the field structure will be abrasive-blasted outdoors, repaired after welding, and inspected around awkward brackets.
Mass-Loss Corrosion-Rate Calculation
Use the standard screening form:
where:
- r is corrosion rate in \text{mm/year};
- K=87.6 for W in mg, \rho in \text{g/cm}^3, A in \text{cm}^2, and t in h;
- W is mass loss;
- \rho is material density;
- A is exposed area;
- t is exposure time.
For carbon steel:
Use coupon area:
Use exposure time:
Bare Witness Coupon
One bare witness coupon loses:
Calculate:
The bare witness indicates a moderately aggressive exposure in this screening test. This value is not a direct 15-year prediction for the coated structure. It is a severity reference that helps interpret the coated specimens.
Candidate B Damaged-Area Coupon
Candidate B has three damaged-area coupons with mass losses:
Mean mass loss:
Mean corrosion rate:
Individual high value:
Candidate B passes the damaged-area screening criteria:
- mean rate 0.010\ \text{mm/year}<0.015\ \text{mm/year};
- maximum individual rate 0.0107\ \text{mm/year}<0.025\ \text{mm/year}.
The engineering interpretation is that the damaged-area corrosion rate is low enough for the project screen, assuming surface preparation and edge coverage are controlled in production.
Compare the Coating Candidates
Use the same calculation for all candidates. The following table summarizes the screening data.
| Candidate | Mean mass loss | Mean rate | Max individual rate | Scribe creep | Screening result |
|---|---|---|---|---|---|
| A | 138\ \text{mg} | 0.0308\ \text{mm/year} | 0.036\ \text{mm/year} | 4.5\ \text{mm} | fail |
| B | 45\ \text{mg} | 0.0100\ \text{mm/year} | 0.0107\ \text{mm/year} | 1.2\ \text{mm} | pass |
| C | 61\ \text{mg} | 0.0136\ \text{mm/year} | 0.018\ \text{mm/year} | 1.8\ \text{mm} | pass |
Candidate A fails both the damaged-area corrosion-rate criterion and the scribe-creep criterion. Candidate B and candidate C pass the simplified exposure screen.
The decision should not stop at the mass-loss numbers. Candidate C has a zinc reserve and good field durability, but field welding after galvanizing would require careful repair. Candidate B is easier to specify for fabricated steel if surface preparation, DFT, stripe coating at edges, and cure conditions are controlled.
Coating Quality Checks
The test plan must include coating quality measurements because a nominal coating system can fail if it is applied poorly.
| Check | Acceptance logic |
|---|---|
| surface profile | must match coating supplier range for adhesion |
| soluble salt contamination | must be below project limit before coating |
| dry film thickness | each layer and total thickness within specified range |
| edge coverage | stripe coat applied to welds, edges, holes, and corners |
| holiday testing | no holidays on critical edges after repair |
| adhesion test | no unacceptable loss of adhesion after exposure |
| cure condition | temperature and humidity within application window |
| repair procedure | repaired areas retested after cure |
These checks protect the project from a false conclusion. If candidate B passes in the lab but production skips surface preparation or edge striping, the test result no longer represents the installed structure.
Selection Matrix
Use a weighted engineering decision matrix after removing candidate A for failing the acceptance screen.
| Criterion | Weight | Candidate B score | Candidate C score |
|---|---|---|---|
| corrosion-control margin | 0.30 | 5 | 4 |
| edge and detail performance | 0.20 | 4 | 4 |
| field repairability | 0.15 | 4 | 3 |
| inspection evidence | 0.15 | 4 | 4 |
| cost and schedule | 0.10 | 3 | 3 |
| compatibility with field welding | 0.10 | 4 | 2 |
Weighted score for candidate B:
Weighted score for candidate C:
Candidate B is the recommended system for this project because it passes the corrosion screen and is easier to control around fabricated details and field repairs. Candidate C remains a valid alternative for modules with no post-galvanizing welding and good dimensional compatibility.
Uncertainty and Error Budget
The decision package should state the main uncertainty sources instead of presenting the calculated rate as exact.
| Source | Effect | Control |
|---|---|---|
| mass measurement | affects W directly | calibrated balance, repeat weighing |
| cleaning method | overcleaning removes base metal, undercleaning leaves corrosion product | written cleaning procedure and witness coupons |
| exposed area | affects denominator | masked edges and measured exposed area |
| exposure severity | accelerated test may not match field wet-dry cycle | bare witness coupon and field coupon rack |
| coating process variation | lab panels may outperform field work | production process records and field repair panels |
| localized corrosion | mass loss can hide deep pits | pit-depth measurement and visual grading |
If the coating decision is safety-critical or expensive to reverse, increase replication, add field exposure coupons, inspect representative welded details, and include first-year field validation before accepting the maintenance interval.
Release Package
The final deliverable should include:
- exposure basis and service assumptions;
- selected coating specification with surface preparation and DFT requirements;
- coupon and panel test matrix;
- raw mass-loss data and corrosion-rate calculations;
- scribe-creep, blistering, rust, adhesion, and holiday records;
- uncertainty and limitations;
- selection matrix and decision rationale;
- production inspection plan;
- field repair procedure;
- first-year and 3-year inspection triggers.
The recommended release statement is:
Release candidate B for the utility skid only if production surface preparation, stripe coating, DFT, cure records, holiday repair, and field-repair procedure match the tested process; install a field coupon rack or inspection witness area for the first year of service; and review any coating damage at edges, weld toes, or bolted covers before extending the 3-year inspection interval.
Common Mistakes
Avoid these errors:
- selecting a coating without defining the exposure and geometry;
- testing flat panels while the real failures occur at edges and weld toes;
- using accelerated exposure as a direct service-life prediction;
- ignoring surface preparation and soluble-salt contamination;
- reporting average corrosion rate while ignoring pitting and scribe creep;
- accepting a coating system without a field repair procedure;
- treating DFT as the only coating quality metric;
- forgetting galvanic compatibility around stainless brackets and zinc-rich coatings;
- failing to connect coupon results to inspection intervals and release criteria.
The engineering value of the project is traceability. A reviewer should be able to see why the exposure was chosen, what was tested, what passed, what failed, what uncertainty remains, and what field evidence will confirm that the coating system is working.