where expertise comes together - since 1996

Web's Largest Portal of Nondestructive Testing (NDT)
Open Access Database (Conference Proceedings, Articles, News), Exhibition, Forum, Network

All Forum Boards
Technical Discussions >
Evaluation of an UT machine
Career Discussions
Job Offers
Job Seeks
Classified Ads
About NDT.net
Articles & News

806 views
08:41 May-24-2001

Carlos Valdecantos

Engineering
MTORRES, Ing.,
Spain,
Joined May 2001
9
Evaluation of an UT machine

Let me put a small problem on the table. Some of you may found it interesting.

An automatic UT water jet machine is designed for the inspection of composite parts and is able to operate both in through transmission and pulse-echo techniques. During the acceptance trials the entry surface resolution in pulse-echo is measured by the customer under the following test conditions:

- aluminium test block with flat bottomed holes
- standard, commercial probes
- manual scan in contact, light oil coupled.

The results does not fulfil the specification requirements (1.2 mm diameter hole at 6 mm from the entry surface is not detected at 5 MHz).

On the other hand, when the system is operated in automatic mode, water jet, pulse-echo using manufacturer probes, the C and D-scan of a 3.8 mm thickness composite reference block with artificial defects show all the embedded delamination-like defects (two of them less than 0.5 mm from the entry surface). Both size and depth of the defects are correctly measured to standard UT accuracy.

Please send your opinion (true, false, comments) on the following topics:

1. The entry surface resolution can be measured with standard contact probe-aluminium block, no matter which material is intended to scan under production.
2. To measure entry surface resolution a sample of the same material and testing with the same probes is always advisable.
3. Looking at the results on aluminium test block a table can be build to estimate the resolution on composite parts.
4. No correlation exist between the resolution on the aluminium test block and the ability of the system to detect near-surface defects on composite laminate.
5. The system shall be rejected because it does not fulfil the specification requirements.
6. The system shall be accepted because it detects all relevant defects.
7. Others topics are welcome. Please, specify.



 
00:12 May-24-2001

Tom Nelligan

Engineering,
retired,
USA,
Joined Nov 1998
390
Re: Evaluation of an UT machine You raise some interesting questions. Here's one opinion.

As a general rule, in any ultrasonic flaw detection application, performance should always be verified using the same transducers that will be used during actual tests, on reference standards made of the same material used in the actual tests. I realize that this is not always possible, but it's definitely the optimum approach. I would not say that in your case there is no correlation between response on aluminum blocks and composites, but it's going to be different, and to accurately establish the correction factors will require some thought and experimentation.

When you change either the transducer or the target, you must very carefully consider how the change affects your test results.

1. Contact and immersion transducers will not necessarily generate precisely the same beam profiles, even if center frequency and element diameter are the same, because of manufacturing considerations including changes in ringdown and bandwidth that can result from the difference between wearplates (contact) and matching layers or lenses (immersion). There are often changes in backing as well. If the immersion transducers are focused, their beam profiles will be VERY much different from contact transducers. Thus, the response to flat bottom holes or other reference standards should be carefully examined experimentally for each transducer, and any differences noted and considered when setting up a test procedure. It will likely be necessary to change pulser/receiver parameters such as pulse energy, damping, receiver gain, and filtering as well when you switch tranducers.

2. Sound coupling from a contact transducer into aluminum, versus from an immersion transducer via a water path into composite, represents two very different acoustic situations. Aside from differences in the target's position with respect to pressure gradients the transducer's near field, a contact transducer's near surface resolution will be limited by excitation pulse recovery, whereas in the case of an immersion transducer the limiting factor is interface echo recovery. These are not the same thing, and thus the minimum thickness in a particular material at which a given size flaw can be found will not be the same..

3. Sound velocity and sound attenuation in aluminum and composites are not the same. Aside from the transducer considerations noted above, a target of a given size and depth that can be found in one may not be detectable in the other due to these differences in acoustic properties between the materials. However, through experimentation it should be possible to determine what size and depth reflector in aluminum corresponds to the minimum critical flaw size in composite.

I'm certainly not in a position to say whether your system should be accepted or rejected, because that's a contractual matter between the manufacturer and the customer. However, in my opinion, the acceptable response under actual test conditions should hold much greater significance than the unacceptable response under completely different conditions.

--Tom Nelligan




 
01:55 May-25-2001

Udo Schlengermann

Consultant, -
Standards Consulting,
Germany,
Joined Nov 1998
174
Re: Evaluation of an UT machine reply by Udo Schlengermann :

(Additionally to the reply from Tom Nelligan).

Questions 1 to 6 are not in a logical flow.
Most important are questions 5 and 6:
5. If there is a written specification, the system has to fulfil all conditions listed in this paper. If it does not it has to be rejected.
If there are conditions within the specification which cannot be met because they are erroneous or technically impossible they have to be discussed between the contractors.

6. if no 5 is mandatory, question 6 is obsolete.

From questions 1 to 4 only no 2 is important:
Transfer of ultrasound from transducer into the object to be tested and vice versa depends strongly on the materials involved, they are acoustically matched. Their behaviour is determined by their acoustical impedances.
Transfer from water to metal is not a good matching. Most of the energy is reflected back to water, only a small amount of energy enters the metal object. So the interface pulse is strong and long which results in a poor near surface resolution.
The transfer from water to a low density plastic material has a much better acoustical matching. The interface pulse is small and short. So near surface resolution is much better than in the water-metal-case.
If both materials are equal (e.g. a perspex delay block on a perspex plate) there is nearly no interface echo and near surface resolution is ideal.

4. Of course there is a correlation between all these cases, The problem can be solved theoretically. But this will not replace the practical proof.

My summary:
Verification of parameters given in a specification always has to be done experimentally under the same conditions listed in the respective specification, e.g. same materials, same delay paths, same probe size and frequencies, etc.

Kind regards
Udo Schlengermann

+++++
-


: Let me put a small problem on the table. Some of you may found it interesting.
.
: An automaticUT water jet machine is designed for the inspection of composite parts and is able to operate both in through transmission and pulse-echo techniques. During the acceptance trials the entry surface resolution in pulse-echo is measured by the customer under the following test conditions:
.
: - aluminium test block with flat bottomed holes
: - standard, commercial probes
: - manual scan in contact, light oil coupled.
.
: The results does not fulfil the specification requirements (1.2 mm diameter hole at 6 mm from the entry surface is not detected at 5 MHz).
.
: On the other hand, when the system is operated in automatic mode, water jet, pulse-echo using manufacturer probes, the C and D-scan of a 3.8 mm thickness composite reference block with artificial defects show all the embedded delamination-like defects (two of them less than 0.5 mm from the entry surface). Both size and depth of the defects are correctly measured to standard UT accuracy.
.
: Please send your opinion (true, false, comments) on the following topics:
.
: 1. The entry surface resolution can be measured with standard contact probe-aluminium block, no matter which material is intended to scan under production.
: 2. To measure entry surface resolution a sample of the same material and testing with the same probes is always advisable.
: 3. Looking at the results on aluminium test block a table can be build to estimate the resolution on composite parts.
: 4. No correlation exist between the resolution on the aluminium test block and the ability of the system to detect near-surface defects on composite laminate.
: 5. The system shall be rejected because it does not fulfil the specification requirements.
: 6. The system shall be accepted because it detects all relevant defects.
: 7. Others topics are welcome. Please, specify.
.



 


© NDT.net - The Web's Largest Portal of Nondestructive Testing (NDT) ISSN 1435-4934

Open Access Database, |Conference Proceedings| |Articles| |News| |Exhibition| |Forum| |Professional Network|