French (Fr)English (United Kingdom)

DPC NEWS: a website dedicated to Penetrant Testing and Magnetic Testing



visits on site since April 2008

Log in


Receive HTML?

DPCNews 008 - ASNT Fall Conference 2008

Written by Administrator
Thursday, 01 January 2009 18:17

Season's greetings from

Patrick DUBOSC - Pierre CHEMIN

January 2009

Report and comment on ASNT fall conference and quality testing show in Charleston, South Carolina, USA, November 10/14, 2008

One of us, Patrick DUBOSC, had the opportunity to attend this conference and to visit the exhibition held on the same days. He also went to the meeting of the PT/MT group, on Monday 10.Some 15 people were there, only one non-American.

This meeting was not that important for European users. Nevertheless David G. MOORE, the President, let us know that Noël TRACY, a well-known expert on PT/MT methods, has sent enormous quantities of files about old MIL, NAV, etc specifications. These documents will be recorded on a DVD after scanning page by page. These are invaluable documents for history.

On the other hand the several conferences held on Nov 11 on PT (none for MT) were useful. There was also a very interesting conference held on Nov 12 about POD (probability of detection).

The first conference was about rebleed (wipe-off technique*). The main topic: what if several rebleeds?

Measurements of indications detectability were done using a luminancemeter. This conference was "typically American" in that all the units used were non SI ones: very small fractions of inches (you may easily read 0.001" for 0.0001"), mills, foot-lamberts for luminance when the SI luminance unit is the candela/m².

AMS-SAE 2647 specification in its paragraph deals with wipe-off. A new revision should be available January 2009.

A comparison was made using acetone or isopropyl alcohol (IPA) as solvents or as the basis of non aqueous wet developers (NAWDs).

A comparison was also made using cotton swabs or wipes with these solvents and acetone or IPA preimpregnated wipes from CONTEC®.

As anticipated several rebleeds impair sensitivity results: for very small indications(crack length: 33 mils, i.e. 810 µm or 0.81 mm), one rebleed allows for detection, but a second one is detrimental!

The "seeability" was measured by a photometer the response of which does not equate to detection by an inspector. My conclusion is that means once again nothing is better than the "obsolete system" comprising the human eyes as sensors and the human brain as signal processor!

The second conference was given by George HOPMAN, who during the Fall Conference 2007 in Las Vegas delighted the audience with his thorough explanations of MT specifications' impossible-to-meet requirements.

This year he gave a similarly entertaining translation of several PT specifications' requirements. Using one of his sentences: "Through the process of evolution and NADCAP, something very simple became very complex".

He gave some examples:

Rinsing water temperature shall be 95 °F (35 °C) maximum. His tests using 185 °F (85 °C) showed no difference in sensitivity.

He used 33 °F (0.5 °C) cold water: similar result.

Overdry 30 minutes at 140 °F (60 °C): only the smallest indication on the PSM-5 is fading.

He states also that it is necessary to "de-emphasize the Known Defect Standard process control". In this we completely agree with him. The PSM-5 was never intended to be the "3-indication-for-Level 2s, 4-indication-for Level 3s, 5-indication-for-Level 4s tool" that many specifications require.

He claims that the main points for PT failures are:

Precleaning (we think it is quite often totally inappropriate for PT purposes).

Insufficient knowledge of acceptance criteria by inspectors.

The inspector did not look at the part thoroughly.

A third conference was given by BLUE-LINE, a company almost "devoted" to the use of blue light instead of UV-A light for many fluorescent applications. The author reminds some basic but important points (often badly understood by users) about fluorescence.

For instance the emission spectrum (the "light" emitted by the fluorescent material) is independent of the energy of the exciting radiation.

(For us this is somewhat disputable, as a material may give rise to different colours when UV-A or UV-C for instance are used. But the author was talking about 365 to 450 nm radiation.)

Different materials may have different responses to 365 and 450 nm as regard to the INTENSITY of reemission, some responding more to the blue (450 nm) than to the UV-A. Several examples were given:

- One penetrant which, when under a radiation between 400 and 500 nm reemits 80% of the intensity measured under a 365 nm radiation.

- Magnetic powders: the ratio of reemission under a blue light vs a 365 nm radiation may be surprising:

MG 900: 1.3 ratio.

MG 93:   2.1 ratio.

MG 800: 0.8 ratio.

That means the MG 900 and 93 fluoresce more under a 450 nm radiation than under a 365 nm one.

Our comment is that this study is very interesting; nevertheless keep in mind that the current fluorescent penetrants are ALL optimised for the most powerful response to 365 nm. Even if 80% response under blue light is impressive ... 80% is the bottom limit acceptable when testing penetrants in tanks for reemission as per several specifications!

However concerns about UV-A use safety may lead to the more acceptable 405 or 450 nm sources...making it necessary for the penetrants suppliers to adjust formulae for a better response to these wavelengths.

A fourth conference about PT dealt with developers application, especially that of dry developers.

All the dry developer installations shown in the paper use either compressed air or electrostatic spray (with compressed air to push the powder) to apply the developer on the parts. All these installations, small or large, have exhaust fans. All these installations give poor results on surfaces when they are downward, sensitivity being impaired, the even coverage of the entire part being not got as downward surfaces get far less powder than upward facing surfaces.

This is a concern for the entire aerospace industry the world over.

Patrick DUBOSC made a double-point remark at the end of this conference:

Using compressed air leads to moisture problems; moisture then "glue" the developer particles.

Using an exhaust fan makes the very fine particles drawn out of the chamber while the largest ones stay in a far larger quantity; this modifies the formula and lowers its ability to detect very fine cracks.

Patrick DUBOSC reminded the audience of a paper given some 6 years ago (then titled: Dry developer: the French paradox) in which the design of what he called the "fan-at-the-bottom" chamber was presented. He explained that this concept may be used in very large installations (he gave the example of a 17m, 20 yard-long, 7m, 8 yard-wide and -high chamber, with four fans). A very small quantity of dry developer is used, new powder being added each shift. In the very large installation, he said, users pour 8 wine-glasses every half-shift.

A large part of the audience was delighted by the idea of French people using wine-glass in a penetrant installation! He suggested to supply the lab in charge of the study with the glasses! George HOPMAN seemed also interested!

An other point not said that day: when using dry developer there is no need to evenly cover the entire surface of a part. What is important is that there is powder EVERYWHERE, even a very small amount. The dry developer will be stuck by the penetrant meniscus which is at the top of the crack--provided removing the excess of penetrant has not been too harsh! The "white-surface" after developer application may even be counter-productive.

The last conference was a kind of general information about Probability of Detection (POD), applicable to all sorts of NDT methods.

The title: POD IS NOT RELIABILITY is self-explaining.

Fracture mechanics was a revolution in Materials Science and Structure Design.

A revolution in NDE was driven by fracture mechanics.

The outcome of importance:

NOT ... the smallest flaw ever detected ... BUT ... the longest flaw never detected.

POD depends on:

Test object.

NDE Engineering:






Calibration variables.

Acceptance criteria/decision variables.

Human or auto inspection.

Human factors: guy to blame!

Reliability is characterized by:

Reproducibility: calibration.

Repeatability: process control.

Capability: POD.

POD is somewhat critical to "third party QA process", to generic "certification " schemes.

As it is often understood as reliability, there is some confusion in its use/purpose. The author said it would be better to focus on better NDE engineering than on "raw" POD.

The exhibition was also full of "new equipment". New equipment is often an improvement of software, new lighter, longer-life batteries. The penetrant and/or magnetic materials suppliers are a minority; their booths are not full of "electronic items"...but these methods are still there for a long time as cheap, invaluable, irreplaceable methods.

Note also that an interesting part of the exhibition the "free buffet", full of very exciting things...and high-quality American wines! But one may also see a lot of visitors/exhibitors eat meat, vegetables, fish, etc while drinking....colas! A pity!

(*) Will be the topic of a future DPC NEWS.

We, Pierre CHEMIN and Patrick DUBOSC, welcome any comment, any idea. If you have some examples you would like to see discussed here, please give us all the useful indications. If you require confidentially, we would modify locations, names and some parameters to prevent any traceability.
Nevertheless, we are convinced that our site may be a kind of surge-valve: the topic is NOT to target this company, or that auditor; but it is always to make users think, to make them ask themselves, or others, the right questions.

We may also give advice, once again on a confidential basis if needed: please, feel free to ask questions, to document our data basis: about Material Safety Data Sheets (MSDS), about environment, a chemical name you don't understand, a Penetrant process you have heard about, etc.
We have plenty of examples, some being out of all the specifications/standards, which led to the discontinuities detection, when the "current, normal, processes" prevented discontinuity finding.

Last Updated ( Monday, 23 May 2011 16:49 )