PT reference test blocks check

December 2011

We have received a very interesting question from a good ol’ friend:

"My question is quite simple: how do we check the modifications (wear) of the test panels used for fluorescent PT?
Do you know of standards or papers (and if so, where can I get them from?), which provide (or allow for the establishing of) a procedure to perform this check? Indeed, according to the PRI (NADCAP) questionnaire, any change in the dimensions of artificial discontinuities shall be less than 30% (further, measured on their longest dimension.) This is for the Tam panel/PSM® 5. However, how should we do for the Type 1 reference test blocks of the ISO 3452-3 standard?

How can one guarantee such an accuracy for the PSM-5® smallest defects? By using a binocular magnifier or a microscope? Shall one proceed under UV-A radiation?
Do you think small PT users can process it? Would it be better to require the PT panels’ suppliers to check panels? Alternatively, is it even better to replace the panels regularly, even if their price is a major obstacle? Once a year?
Thank you in advance for your help."

In the best of our knowledge, no standard or specification deals with such a check.

The question is which length is to be measured. The crack’s? Or the indication’s? A binocular magnifier with a reticule is the right instrument. NADCAP agrees that the reticule be calibrated once and for all - one should hope so!!!
Nothing prevents to perform it using a Level 4 fluorescent penetrant and to measure the indications’ length as time goes by and the panel is used. If an equipment such as the ones that some (not all) PT manufacturers/suppliers have, a file may be opened, in which the data collected every year are recorded. This makes it possible to see the trend and to warn on an incoming problem.
As far as the Type 1 reference test blocks of the ISO 3452-3 standard are considered (these panels are rarely used in the USA), it is not that easy: there are far too many indications.

To get more pieces of information on the reference test blocks cleaning, you may read a paper we have already released on our Website(1) on this topic.
However, our experience is that, regardless of the precautions taken, with the Type 1 reference test blocks of the ISO 3452-3 standard, discontinuities tend to clog. This cannot be countered, even with suitable cleaning means. The Type 1 reference test blocks of the ISO 3452-3 standard shall be changed, at least every five years.

Meanwhile, and regardless of the reference test block, a yearly reference photograph MUST be taken: there is a slow performance drift along the years, and, by using a picture of the test panel made four or five years previously, it is likely that, at a time, the difference between the picture and the current panel will be such that it will lead to problems. Therefore, a new photograph shall be done every year at least.
Want to buy brand new panels every year? A dream for the suppliers!!! However, not that realistic!

A user without the suitable laboratory means should better rely upon the test panel supplier to perform the yearly check. The reliable suppliers (most of them are) do not "play the ‘buy a new test panel’ card," except if really needed. Moreover, a certificate/advice from a third party is more "valuable" vis-à-vis an Auditor.

However, it is one of our duties to instruct users how to treat test panels with respect. They are fragile and expensive equipment. We have seen two major users who, after nine months, had nearly destroyed their reference test blocks, and who required a free substitute. We went to the PT process lines; we investigated and found that all the test panels were often cleaned in an alkaline detergent ... a process that the Instructions for use formally advise against. We also gave the operators a piece of information about the cost of each panel. We never faced the same problem due to these users!

1- In-house recalibrations

1.1 PSM-5®/TAM 14640 test panel

This test panel is manufactured according to PRATT & WHITNEY AIRCRAFT (P&W A) specifications. (P&W A) also issues the document governing recalibration, FPM-MASTER Appendix C. However, this document only states the results to get. There is no detailed step-by-step procedure. The actual processing steps are described in P&W A’s FPM manual(2).

Following the guidelines described in the FPM manual, a recalibration procedure could be as follows:

1) Clean the panel thoroughly before processing by soaking it overnight in isopropanol or another appropriate quick evaporating solvent or by running it in an ultrasonic bath using a solvent cleaner.

2) Apply unused penetrant, the same one, which is or, which will be regularly used on the panel, with a rag or a brush after the full evaporation of the solvent. Allow for the penetrant dwelling for at least 10 minutes, but no more than an hour.
3) Rinse the panel for 90 seconds with sprayed water at a pressure not greater than 210 kPa (30 psi) and in the 10°C-30°C (50-90°F) range.

4) When using a post-emulsified penetrant, immerse the panel in unused hydrophilic emulsifier, diluted at the strength to be set in the procedure, and then take it out of the emulsifier. Allow for the emulsifier draining for 90 seconds at most.

5) Wash the reference block the same way as in step 3, no longer than 120 seconds.

6) Dry the panel either in an oven set at 70°C (160°F) maximum, or with a hot gun without exceeding 70°C (160°F) on the part surface, for no longer than 30 minutes.

7) Immediately, after drying the panel, apply unused dry developer by immersing the panel chromated-face down in the developer, so that all the five discontinuities are exposed to the developer. Let in the developer for ca a second (yes, a second), then take the part off the developer.

8) Allow for a developing time of at least 10 minutes.

9) Using a gage (also called "fluorescent indication comparator" or "transparent comparator") like TAM 135273, measure the five indications. The minimum sizes listed in Appendix C in the FPM-MASTER are E-0.180" D-0-125" C-0.075" B-0.045". A-no minimum. Additionally, all indications must be in proper proportion to each other. In other words, E>D>C>B>A.

Figure 1: Transparent token measuring 20 x 100 mm compliant with the requirements of the P&W A’s TAM 135273, with two sets of nine indications, one of linear indications, the other one of rounded indications. Marks in both imperial and metric units. Lengths of the linear indications and diameters of the circles range from 0.01 to 0.18 inch; from 0.2 to 4.6 mm.

10) For a Level 4 fluorescent post-emulsified penetrant, all five indications must be seeable. For a Level 3 fluorescent post-emulsified penetrant, the four largest indications must be visible. For a Level 2 fluorescent water-washable penetrant, the three largest indications must be visible.

11) If, after processing, the indications are not visible or are undersized due to clogging, the panel must be cleaned. This can be done effectively by an ultrasonic cleaner and/or by spraying a non-aqueous wet developer onto the panel several times and wiping it off. Repeat this cleaning process as many times as necessary so that no trace of penetrant appears under UV-A, under viewing conditions in inspection booth. In the most difficult cases, one way to unclog cracks is to place a drop of a mild acidic material, such as a household mineral remover (generally made of an aqueous solution of 23% orthophosphoric acid) directly on the crack for about three minutes. Then rinse thoroughly with water. Use de-ionized or distilled water if the next step is to dry the panel.

12) If, after several cleanings, the indications are still undersized, out of proportion, or not visible using the appropriate penetrant, the panel must be deemed "out of specification".

This procedure should be used on both "daily" and "master" reference test blocks. The user must process the "master" panel through all the penetrant systems that the panel will be used with (using unused materials). However, it is recommended to use a separate (Master) reference test block for each PT process. The result should be noted and referred to whenever the panel is run to ensure the consistency of results. This procedure must be comprehensive enough to provide accurate results.

In most cases, we only talk about the under sizing of the artificial cracks (due to either impurities, or limestone left by the water used for the repeated rinsing/washings). However, there may be also over sizing of these cracks (by chromium coating deterioration). This shall be taken into account.

Additional information on fluorescent indication comparators

We think that there is more than one manufacturer of these comparators. This assumption comes because one of us was always supplied with good-quality items, while the other got sometimes-defective comparators.

In all cases, these comparators, which do not have any serial number, are not calibrated...; this would make them far too expensive.
They must be qualified as "indicators", very useful tools, indeed, that can send a warning signal to use more accurate means if a drift is seen, or when reaching the limit of an acceptable/not acceptable criterion. They do not need calibration.
Moreover, these comparators being made of plastic, it is likely that the indications sizes and dimensional tolerances may vary over time.

Note that, regarding the evaluation of PT discontinuities indications, some specifications state that these comparators may be used as a substitute to some sizing gauges, if they are periodically checked.

One manufacturer supplies thirteen different types of comparators, marked in metric and/or imperial units that cover almost all the needs.
On request, this manufacturer delivers what is called a "certificate of conformation", which states that:

• The supplied items have been manufactured and inspected using tooling and gauges that are traceable to the National Institute Standards and Technology (NIST).

• Sizes on templates were inspected using a 7x power optical magnifier that has been calibrated and traceable to the NIST (the certificate number is quoted).

•  The undersigned deposes and says that the comparators have been inspected and conform to the dimensional tolerances of the design listed below:
> Dimension of .060 à .500 is within a tolerance of +/- .0.005
> Dimension of .045 is within a tolerance of +/- .004
> Dimension of .30 is within a tolerance of +/- .003
> Dimension of .010-.025 is within a tolerance of +/- 002
> Dimension of .005 is within a tolerance of +/- 002- .000

• The collective uncertainty of our measurement standards does not exceed 25% of the characteristic being calibrated or verified.

We think this is enough for indicators.

1.2- Type 2 reference test block of the ISO 3452-3 standard

The same procedure as above could be used with the reference test block of the ISO 3452-3 standard.

2- External recalibrating

Several PT material/equipment/accessories suppliers test panels cleaning and check, as a periodical "customer service".

Then, the test panels are sent back to the supplier for thorough cleaning, re-grit blasting to restore the initial roughness, and cracks measurement.

If the test panels can be restored to comply with the relevant requirements, they will be recertified.

3- Conclusion

We have seen that an in-house recalibration is possible for some reference test blocks.
However, we believe that an external recalibration by experienced PT materials/accessories suppliers that have top-quality calibration equipment is better as it provides a greater reliability.
Further, it is a good idea to require them to perform the reference test blocks cleaning, as this is a dramatically critical step that, if badly processed, will impair them and could make them unusable.

References

(1) Patrick DUBOSC and Pierre CHEMIN, Penetrant Testing: test panels/parts cleaning, DPCNewsletter N° 010 - March 2009. On our Website:
http://www.ressuage-magnetoscopie-penetranttesting-magnetictesting-dpc.info/site/en/dpc-news/2009/105-dpcnews-010-nettoyage-en-ressuage

(2) Nondestructive Test Manual Requirement-Fluorescent Penetrant Method, Fluorescent Penetrant Method (FPM) Master, Pratt & Whitney Aircraft (United Technology), P.O.Box 109600, West Palm Beach FL 33401-9600, USA.

Normative references

• ISO 3452-2:2006 Non-destructive testing - Penetrant testing - Part 2: Testing of penetrant materials, International Organization for Standardization, Geneva, Switzerland, 2006.

• ISO 3452-3:1998 Non-destructive testing - Penetrant testing - Part 3: Reference test blocks, International Organization for Standardization, Geneva, Switzerland, 1998.