Thickness survey analyzer: corrosion rate and remaining life across many readings

Try one free question

Paste a column of CML or TML readings from two surveys and get a short-term corrosion rate and remaining life for every point, the governing point, and the half-life next inspection date. Sort the table, then download it as a CSV.

Work corrosion rate, remaining life and the half-life interval for every point in a survey at once, using the same math as the single-point calculators. A check on hand work; your code edition, procedures and the responsible inspector govern real work.

Analyze a thickness survey

An example is loaded. Replace it with your own readings, one point per line, or paste straight from a spreadsheet. Everything runs in your browser and nothing is saved.

The unit applies to every number you paste, so set it before you read the results. Switching it does not convert the text. The 10-year starting cap is the vessel maximum; for piping, enter the maximum for the piping class.

Commas, semicolons or tabs between values, periods for decimals. A header line is skipped. Up to 500 points.

Governing point

Point with the least life
CML-3
Short-term rate
8.75 mils/yr
Remaining life
4.57 years
Half-life interval and next inspection
2.29 years, by 2028-06-12

5 points over 4.0000 years: 3 with an estimate, 1 with no loss, 1 to review, 0 at or below t-required. The next inspection date uses 10 years as the cap and half the remaining life of the governing point.

One row per point. Sort by any underlined heading.
(mils/yr) (in)Previous → current (in)Status and notes
CML-180.11814.752033-07-150.5 → 0.468, t-req 0.35Estimate
CML-2120.1028.502030-05-310.5 → 0.452, t-req 0.35Estimate
CML-3 (governing)8.750.044.572028-06-120.375 → 0.34, t-req 0.3Estimate
CML-400.15No estimateNone0.5 → 0.5, t-req 0.35No loss
  • No loss between the two readings, so there is no finite life from this model. Zero loss over one interval does not mean the point cannot corrode.
CML-5-2.50.14No estimateNone0.48 → 0.49, t-req 0.35Review reading
  • Reading went up: the later thickness is greater than the earlier one. Check the gauge, the location and the surface before you trust either number.
  • No life estimate until the reading is resolved.

Rates are straight-line averages between your two surveys. The point with the least remaining life drives the interval for the survey. Never average points together.

Studying for API 510? Take the free 10-question check to see exactly which topics to fix first, then close the gaps with the Long Seam course.

How the analyzer works

You give it one line per point: the identifier, the thickness from the previous survey, the thickness from the current survey and the t-required for that point. The two survey dates and the maximum interval apply to the whole list. For each point it does the same steps the single-point calculators do:

  • Short-term rate = (previous − current) ÷ years between the two surveys. Years are calendar days divided by 365.25. For inches the rate is shown in mils per year (in/yr × 1,000), and for millimeters in mm/yr.
  • Remaining life = (current − t-required) ÷ rate. The margin is the corrosion allowance you have left at that point.
  • Interval = the smaller of half the remaining life and your maximum interval. The next inspection date is the current survey date plus that interval, rounded down to a whole day.

These are the rate, remaining life and interval steps in the API 510 Body of Knowledge, which points to section 7.1 for rates, 7.2 for remaining life and section 6 for the interval. Open your own edition for the wording; nothing from the code is reproduced here.

Why the worst point drives the survey

A vessel does not average its thickness. One thin spot reaching t-required is enough, so the governing point is the one with the least remaining life, and the interval comes from that point alone. The table labels that row (governing), and you can sort it by remaining life.

The tool also avoids three traps. A point that is already at or below t-required has no life left and is flagged for engineering evaluation instead of getting a date. A point with no measured loss gets no life estimate, since a rate of zero would divide by zero and prove nothing. A point whose reading went up gets flagged for review and is left out of the governing choice, because a negative rate is a measurement problem, not a gain.

Checking your data before you trust the result

  • Same spot, same method. A rate is only as good as the two readings being comparable. Gauge, probe, couplant, surface preparation and temperature all move a UT reading.
  • Rising readings. Read them as scatter or a relocated point, not as growth. A loss of a few thousandths over a short gap can sit inside the gauge variation, so be careful with rates from very small differences.
  • Units. Every number you paste is read in the unit you choose, and switching the unit does not convert the text. Mixed units in one list will give wrong answers with no error.
  • t-required. It is your input and it can differ by location. Use the value from your own calculation. The MAWP and thickness calculator shows how a required thickness is worked for a shell or head.

Worked example 1: two points in inches

Surveys on 1 March 2022 and 1 March 2026, which is exactly 4 years. CML-2 went from 0.500 to 0.452 in with t-required 0.350 in. CML-3 went from 0.375 to 0.340 in with t-required 0.300 in.

CML-2 lost 0.048 in, a rate of 0.012 in/yr = 12 mils/yr. Margin 0.102 in, so remaining life = 0.102 ÷ 0.012 = 8.5 years and half of that is 4.25 years.

CML-3 lost 0.035 in, a rate of 8.75 mils/yr. Margin is only 0.040 in, so remaining life = 0.040 ÷ 0.00875 = 4.57 years and half of that is 2.29 years. The slower point has the thinner margin, so CML-3 governs: next inspection 12 June 2028, even though CML-2 is corroding faster. Sorting by rate alone would have picked the wrong point.

Worked example 2: a point where the 10-year maximum wins

In millimeters, surveys on 1 June 2022 and 1 June 2026 (4 years). Point A went from 12.0 to 11.2 mm with t-required 9.0 mm. Point B went from 12.0 to 11.8 mm with t-required 9.0 mm.

Point A lost 0.8 mm, a rate of 0.2 mm/yr. Margin 2.2 mm gives 11 years of remaining life, and half is 5.5 years, so its date is 30 November 2031. Point B lost 0.2 mm, a rate of 0.05 mm/yr. Margin 2.8 mm gives 56 years, and half of that is 28 years, so the 10-year maximum holds the interval at 10 years. Point A governs. Point B tells you the vessel is healthy there, but not that you can wait longer than the maximum.

These are original arithmetic examples with invented numbers, not exam questions or real survey data.

Thickness survey questions

Which point sets the inspection date for the whole survey?

The one with the least remaining life, not the average. A vessel is only as good as its thinnest margin, so this tool finds the point with the shortest life and takes half of it, capped at the maximum interval you enter. Points with long lives do not buy time for the weak one.

What do I do with a point whose reading went up?

Treat it as a bad reading until proven otherwise, because metal does not grow back. Check that the gauge was calibrated, that you were on the same spot, and that the surface, temperature and coating were handled the same way both times. The tool lists these points as review and leaves them out of the governing choice until you fix the number.

A point shows no loss between surveys. Is its life unlimited?

No. Zero loss over one interval only tells you the rate was too small to see with your gauge in that time. The tool gives no life estimate for that point rather than a made-up number, and you decide what rate the point deserves, using the history, the service and the next survey.

The single-point tools show every step for one location: the corrosion rate calculator adds a long-term rate from an initial reading, the remaining life calculator lets you edit the interval rule, and the inspection interval calculator shows which limit governed. For the full list of calculation topics, see the API 510 calculations guide.

Try one question

Question 1 of the free 10-question check on API 510. No signup.

Question 1 of 10 · free checkUntimed

IV. Practical knowledge - general

Which hydrotest situation carries the highest brittle fracture risk?

Choose an answer to begin

Free API 510 study plan

A week-by-week plan from today to your exam, weighted to the published exam outline. We show it here and email you a copy.

Ask Patch before you buy AI your Long Seam study buddy

See what's on the API 510 exam, take the free 10-question check, or find the plan that fits you.

Published references

Preparing for API 510?

You just took a whole survey from raw readings to a governing point and a next date. The API 510 exam asks for the same chain by hand, one point at a time. The free 10-question check shows which API 510 topics to fix first, no signup. The Long Seam course drills them and explains every answer.

Take the free 10-question check

By the Long Seam editorial team. Report an error.

Add this analyzer to your site

Copy this HTML with its attribution link. The frame scrolls to include the analyzer and explanations; no external script is needed.