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 010 - Test panels/parts cleaning

Written by Administrator
Sunday, 01 March 2009 11:19

March 2009

Penetrant testing: test panels/parts cleaning

I- Scope

This paper deals with ALL KINDS OF REFERENCE PARTS: test panels as described in standards (such as the ISO 3452-3) or real parts taken out of a production line and kept apart because they display specific discontinuities, or hard to detect ones, or a specific roughness, or for any good reason. In some industries (such as automotive) they are called "ghost parts".

II- Introduction

What are the most widely used test panels? There is not one answer: this depends on the industry branch, on the country.

For instance the US aerospace industry uses the PSM-5® (Penetrant System Monitor-5: (5 comes for the 5 star-like defects indented on the panel) almost exclusively.

In France, the nichrome panels, described as the Type 1 Test blocks in the ISO 3452-3 standard, are widely used in the aerospace industry and in the nuclear industry.

The PSM-5® is also used in the French aerospace industry on US manufacturers’ requirements. In Germany the aerospace industry relies upon the PSM-5 and the nichrome panels while the car industry uses the Type 2 Test block described in the ISO 3452
-3 standard.

A lot of aircraft subcontractors working for the US industry in the world use the PSM-5® panel.

The ISO 3452-3 Type 2 panel is unique in that it displays four different surface roughnesses (Ra) giving a very thorough information about washability, emulsifier efficiency, etc. On the other hand the 5 star-like defects are far larger than on the PSM-5®. Therefore they are less prone to alert the user on an adverse change in a PT installation performance.

The PSM-5® is also known as the TAM 146460 panel, after a document by Pratt and Whitney which exactly describes what it is.

These test panels, especially the PSM-5® were NOT initially intended to "measure a sensitivity": their purpose was to establish a "baseline" with brand new penetrant materials in an installation and all parameters set at their nominal values (water temperature and pressure, drying oven temperature, timers of all kinds, UV-A irradiance in the washing booth and in the inspection booth, ambient visible light illuminance in both, etc). Then every day, or every shift, or every time the installation is put back to service, the test panel comes along the first set of parts. When inspected in the inspection booth it should display the same pattern of indications. If there is any difference, it should be reprocessed. If the difference is confirmed, either a "master" panel is to be used to check whether the difference comes from the panel itself or from the installation, or directly the Quality manager shall consider there is a problem on the PT line, up to him (her) to decide what to do.

Along the years we have noticed a drift in the requirements. As far as the PSM-5® is concerned, more and more the following requirement is stated:

- A Level 2 penetrant process shall clearly display at least 3 stars.

- A Level 3: at least 4 stars.

- A Level 4: all the 5 stars.

This was not, once again, thought of in the design of this panel.

Test panels #1, as per ISO 3452-3 standard (also known as Nichrome panels) have been used for many years by the French aeroengines manufacturer SNECMA to establish fluorescent penetrants sensitivity. Until mid ’70s there was no very efficient means to classify penetrants. Jean VAERMAN, then the person in charge of NDT methods in the Materials Laboratory, decided to use these reproducible panels on an automatic machine. The four panels (so-called: 10 µm, 20 µm, 30 µm and 50 µm, for the plating thickness of the nickel + chromium in which cracks are induced) were used and, depending on the percentage of cracks detected on all the four panels by a sensor, the machine decided to classify the tested penetrant at sensitivity Level 2, 3 or 4 (Levels called S2, S3 and S4 as per SNECMA specification DMC0010).

This system was far better than the equivalent in the MIL-I-25135 C, D and E revisions in that the discontinuities were far smaller, reproducible and the indications were recorded as seen (1) or not seen (0) by the automatic sensor + signal processor. No "maybe" or "maybe not" situation. No comparison with "artificial penetrants" whose formulae were designed ... by the US Air Force Materials Laboratory.

Later the US SAE-AMS 2644 specification used an other type of reference parts, and chose commercial, tried and tested penetrants as references for every level of sensitivity.

No one now can imagine using a penetrant line without one reference part. Nevertheless real parts may be used instead of, or along with, the widely used test panels. In fact real parts are closer than test panels to the real material, surface roughness, shape, location of likely real defects: they are a very good means to test on a regular basis the inspectors' ability to process the parts manufactured by the company. But these parts cannot be reproduced and the results they show cannot be compared with those got on other lines.

Since very few years now the best way to compare a test panel processed on the spot to this same panel baseline is to have a reference photograph. After years when we pushed hard finally it is now quite always mandatory that:

- The photograph shall be at full scale against the real panel.

- The photographs shall be visible under UV-A radiation if fluorescent penetrants are used.

- They shall be in colour.

The photograph quality is now a concern for many users who think they can do their own pictures as digital still cameras seem to allow for picture taking even in the most difficult conditions.

The quality of test parts is one of the most frequent causes for non-compliance-report (NCR) during audits (NADCAP or others).

And as a last point cleaning these parts after tests is of the UTMOST importance as regard to reliability, consistency and reproducibility of tests results. Inappropriate cleaning is the #1 factor leading to non-consistent tests, cause of costly outing-of-service of penetrant lines when in fact the line is in good condition--but not the tests parts!

III- Handling precautions

Test blocks handling requires some sine qua non, sometimes basic precautions so as to prevent any state of irreversible damage such as scratches, new cracks, existing cracks widening. When Type 1 (nichrome) panels are brushed use only a soft bristle brush. Brush parallel to cracks to avoid chipping them. Do not let panels lie about: store them in their case or in a pouch.

Hang panels by the edge: contact with skin is detrimental due to the lipo-acid layer which protects skin from infections and dehydration. This lipo-acid chemical let a "fatty" deposit on the surface which may prevent penetrant from wetting this area.

It is necessary to bring panels back to their initial cleanliness after EVERY use. After a penetrant test is processed there is ALWAYS some penetrant strongly retained in the cracks by the capillary effect. These residues may interfere with a further test, sometimes enhancing "sensitivity", more often leading to non reliable results.

IV- How to effectively clean test parts

IV-1- Chlorinated solvents

Indisputably the best way! And the "best of the best" is using the vapour-phase process! When available!! See underneath Health, Safety and Environment Chapter.

Do NOT keep panels in any chlorinated solvent: these solvents are all unstable when in contact with moisture (water from the atmosphere), when under visible or UV-A radiation, or heat. These factors lead to the split of the carbon-chlorine bond and the chlorine free radicals are very strong corrosion agents. The panels are quickly corroded!

IV-2- Alkaline cleaners

Also referred to as "alkaline detergents".

Soap and water ARE NOT to be used! Soap is a smooth detergent for skin....but there is no corrosion inhibitor in the formula! Further soap contains saponified chemicals which are prone to clog cracks if uncomplete removal occurred. Soap is NOT solvent-soluble; therefore it cannot be taken out of cracks when using a solvent.

Soaps are NEVER used to clean parts surface before Penetrant Testing. Alkaline cleaners are used on titanium parts and may be used on light alloys parts, provided that the formula is compatible with these materials.

Alkaline cleaners are used at temperatures in the 70/80°C range (158/176°F). They shall be:

- corrosion inhibited

- silicates-free (or very low in silicates)

Silicates are cheap minerals which lead to very efficient cleaners for specific hard-to-remove pollutants. The unacceptable drawback, when PT is performed after a cleaning with these high-in-silicates products, is that when the parts at room-temperature are immersed in, or sprayed with, the hot detergent, silicates give a thin, impossible-to-remove layer on the parts. Penetrant cannot enter discontinuities ... and parts are then all "accepted" as no indication is ever seen!

After the required time of degreasing parts are rinsed with water (cold or hot, one or several rinses). Degreasers contain surfactants, chemicals which substantially lower the water surface-tension. Degreaser very easily comes in the discontinuities while rinsing water, deprived of such surface-active agents, has a very high surface tension; it cannot easily go into the discontinuities and cannot dissolve the soluble chemicals which came along the degreaser.

After drying, if some soluble chemicals stay in the discontinuity they will make it more difficult, even impossible for the penetrant to enter the discontinuity.

As some rinsing water nevertheless found its way in the discontinuities it is one of the major concerns to make it COMPLETELY EVAPORATE DURING DRYING TIME BEFORE PENETRANT APPLICATION. Drying MUST BE EFFICIENT. Numerous experiences show that a drying temperature in the 105/125°C (221/257°F) range, and a time long enough (often more than one hour) are required to ensure a complete water evaporation.

Tests parts should not face such high temperatures.

Further, once again experience shows that ANY ALKALINE DEGREASER HAS A DETRIMENTAL EFFECT on test panels.


IV-3 - Non-halogenated solvents

What is a "non-halogenated solvent"? A solvent free of chlorine, fluorine, bromine, iodine. Generally the non-halogenated solvents are hydrocarbons, though some oxygenated solvents may be used (different alcohols, except methanol; acetone, for example).

Hydrocarbons-only solvents are not very good cleaners prior to Penetrant Testing. These chemicals do not dissolve water, nor water-soluble minerals; they have a lower density than water, and hence if water is in a discontinuity (for instance moisture which stayed a full night on a part) the hydrocarbons will come on top of water, not displace it.

The best degreasers/solvents are based on a mix of hydrocarbons + a light alcohol (such as isopropyl alcohol) + a ketone (such as acetone). Such a mix is at the same time polar and non-polar and may dissolve hydrocarbon-based pollutants as well as water and many water-soluble chemicals.

Please note that the American SAE-AMS 2647B (R) specification, ‘‘Fluorescent Penetrant Inspection Aircraft and Engine Component Maintenance’’ states that test parts shall be processed (cleaning and storage) as per the manufacturer's instructions or as per AMS 2647B para to which, in summary, recommends: remove developer with water and a soft nylon brush, then dry then immerse the panel in a volatile solvent such as isopropyl alcohol, acetone or methyethylketone (in Europe, this chemical is now forbidden for use in open tanks) for at least 10 minutes. Then thoroughly dry the parts as any trace of solvent in the discontinuities may prevent penetrant from going in.

Some comments: the named volatile solvents may evaporate within one hour at room temperature. As the test panels are supposed to be used every shift at most (i.e. ca every 8 hours) this allows for complete evaporation between two successive uses.

Traces of solvent do not prevent penetrant from entering the discontinuities; what occurs is a dilution of the in-going penetrant by the solvent leading to a likely less "sensitive" result.

V- How to come back to the initial cleanliness: a recommended procedure

Through a 30-year+ experience, the following procedure is the one which:

- gives users the best reliability,

- allows for a long use of test parts,

- is applicable either on artificial test panels or on ghost parts.

To understand the process we have to remind readers of some points:

- penetrant enters any discontinuity, even those we do not detect after penetrant inspection process completion.

- colour contrast penetrants or fluorescent ones have exactly the same ability to enter discontinuities.

- the eyes and the human brain detect far smaller quantities of fluorescent dyes than of any colored dye.

- any remaining trace of penetrant in discontinuities has consequences on the following penetrant inspection process: remaining trace may adversely react with the penetrant put on the surface; on the contrary, if the penetrant used for the current inspection has been severely polluted since the previous inspection, the remaining traces of the "good" (i.e. non-polluted) penetrant may enhance the sensitivity so as to make the inspector accept the test as "OK", while using a badly underperforming penetrant.

Now carefully read this procedure: it is cheap, easy to perform in any plant.

1- Remove the developer layer using a soft bristle brush.

2- Immerse the part in a volatile non-halogenated solvent (isopropyl alcohol or acetone are the most appropriate) for at least 4 hours.

3- Take the part out, let it dry (at least 10 minutes: don't forget that a rapidly evaporating liquid cools down the surface; if atmosphere's hygrometry is close to 80/100%, this may lead to moisture wetting the surface! The surface shall be ABSOLUTELY DRY when going to step #4.

4- Apply a THICK layer of a non aqueous wet developer (NAWD).

5- Put the part in an oven set at 70/80°C (158/176°F) for at least 10 minutes. Heat added to the NAWD help to draw the penetrant traces out of the discontinuities.

6- Letting the developer on, let the part cool down and inspect it as if a tested part: UV-A radiation or white light with the asked-for viewing conditions.

7- If any indication come back to step #1 and reprocess until no indication is visible.

When test parts are resistant to this process, that means:

- Parts may have been let without any cleaning process for a long time.

- Minerals may have entered discontinuities (due to alkaline cleaners for instance!)

-  Corrosion may affect the surface or the discontinuities.

If minerals or corrosion are involved, the cleaning process is not that easy! Corrosion is a cause for putting the parts in a dust-bin (trash-can for our American friends) and buying brand-new ones! Happy suppliers, who like users using corrosive "cleaning means": alkaline cleaners, acids, etc (KEEP IN MIND THAT TEST PARTS SHALL GO THROUGH THE PENETRANT PROCESS ONLY, NOT THROUGH THE COMPLETE SURFACE-PREPARATION PROCESS THAT A LOT OF PARTS GET PRIOR TO PENETRANT INSPECTION). No way to come back to part's initial condition with corrosion.

If the problem comes only from minerals (except from silicates, once again), the best way is:

- Immerse the parts in demineralised water in an ultrasonic degreaser tank, and put on the US for something as 30 minutes. Check for water level, as water will heat and level will go down. NEVER USE A US CLEANING TANK WITHOUT ANY LIQUID IN.

- Take the part out of the tank, rinse thoroughly with demineralised water at room temperature.

- Dry parts in an oven set at 70/80°C (158/176°F) for 30 minutes.

- Let part cool down to room temperature.

- Go back to step #1!

Using ultrasonic cleaning too often may modify the outer part of the discontinuities. We write "may" as opinions differ on this topic. So it's up to every Quality Assurance Manager to decide.

VI- Health, safety, environment

We are to put here some reflections which may help you to choose among different cleaning processes (for your manufactured parts as well as for the test parts).

Non-halogenated volatile solvents ARE FLAMMABLE OR EVEN EASILY FLAMMABLE. If a non-halogenated solvent is not classified as at least "flammable", that means its volatility is very low (the flash point figure is a good indicator): it will not evaporate rapidly; and this is a concern for penetrant inspection efficiency.

Halogenated solvents (mainly chlorinated ones, but also some with fluorine in the molecule) are non flammable, with very few exceptions. But their use is more and more regulated. Further, as seen previously, some of them may give rise to corrosion on parts.

Alkaline detergents are costly: not the products themselves, but their use require a lot of square meters (or square feet!), a lot of energy (for heating), a lot of water (water evaporates from the tank, as these cleaners are generally used in hot water), a lot of rinsing water, better if at least the last rinse is by demineralised water and these products lead to an impressive volume of waste water. Treatment cost is not the cheapest line in the balance sheet!

Using an ultrasonic tank with a flammable solvent is forbidden. Nevertheless it is quite easy to use in a lab, and not that dangerous if flammable vapours cannot accumulate! But let lab people manage that; do NOT use such a combination in the plant, along the penetrant line, to clean your test panels!!!

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 16:29 )