วีดีโอ

You have to differentiate the plane of oscillation from the wave propagation direction. The plane of LW oscillation is in the propagation direction. The plane of SW propagation is perpendicular to the propagation direction. BUT: The propagation direction of shear and trans waves are usually NOT 90°. The propagation direction depends on Snell's law, so it depends on the incident angle and the speed of sound. As the speed of sound of LW is about double that of SW we have different angles in our components.

@JohannesVrana so the lines of refraction are lines to indicate the planes of oscillations, right? The lines of propagation are not represented. The angle between them is due to the difference in the propagation speed.

@@billy-cg1qq The lines you see (starting from 04:29) indicate the wave propagation directions. The angle between them is due to the difference in the propagation speed.

​@JohannesVrana but the difference in the angle between the two types of wave in the material is not only due to the difference in their speed. The difference is more fundamental. If we imagine a network of particles connected with strong bonds, if a particle moves up and down, it's gonna create a pressure wave in it's oscillation direction, but it will also create a transverse wave with the particles next to it from the right and left. So the angle of propagation is generally 90°. If it get slowed, the angle between the waves can't just get less that 45°. I think it should be at least 45° as the difference in the speed is 1/2. Thanks for your patience and responses. I hope I'm not missing something obvious here 😅

@@billy-cg1qq, SW can be created in a variety of ways. 1. What you are speaking about are SWs induced by an LW, as discussed in this video: th-cam.com/video/mr778r1uStw/w-d-xo.html. Usually, those are quite minor effects. 2. What is discussed in this video is something different. In this video, the creation of LW or SW by refraction is discussed. Those SWs are way stronger and their propagation direction can be calculated by Snell's law. For sure the speed of sound is caused by the lattice properties (solid state properties). 3. In civil engineering applications 0° SWs are created by transducers "scratching" on the surface.

This entirely depends on the material and the sound path. If you have a forged and heat treated material and the sound path it should be possible to let's say 1/2 meter

You are very welcome

You are very welcome ...

Muchas Gracias

The sound is generated by your active element, travels through some couplant, and is partially reflected where your sound enters your actual part. This reflection goes back through the couplant and is reflected again by your probe. This continues and leads at the beginning of every A-scan to an increased noise level. This is the dead zone. It is the zone where you can not detect anything as the noise, caused by this bouncing back and forth between the active element and the entry surface, is too high. And it is the same effect for conventional and Phased Array probes. On the other hand, the near field is the natural focus point of your beam, which can be calculated. This video can help: th-cam.com/video/xWRABwg1TB8/w-d-xo.htmlsi=M8IclHFQWzDRgFhr

​@@NDE40Ok thank you very much. In other words the PA transducer has a dead zone like the conventional probes. We are talking about 3-6 mm of dead zone below the surface. From this point of view there are no advantages in using the PA instead of the conventional transducers. Correct? Thank you again. Best Regards. Massimo.

@@MassimoPradetto Exactly - phased array and conventional probes with the same aperture also have a similar deadzone. Actually, I found that the deadzone of phased array probes is a tad longer. This is where TFM can help (but only if done correctly).

You are very welcome

Not really - I am a consultant on UT and NDE 4.0

The wavelength determines the minimal size of the objects you want to find - a wave with 10 kHz in steel (~6000m/s) allows you to find objects with ~0,5 METER (the lower limit would be something like 0,15m). With 2 MHz you can typically find indications of ~1 mm and larger.

Unfortunately, I do not have any experience in nonlinear UT. For crack detection and wall thickness measurements I would always use linear methods to do their ease of use. However, in the detection of early damage, we are most likely speaking about damages that are #1 clearly smaller than the wavelength and #2 still quite "integrated" into the material and therefore create weak signals. And this is where nonlinear UT or even nonlinear reconstruction could help. Also, take a look into some medical applications.

Thank you so much for the suggestion. And yes, by early damage I meant defects smaller than the wave length & producing weak signals.

again thank you for your videos. I've watched most of them to get a grasp on UT methods. Specially, integrating NDE and AI seems really promising, and I'm planning on looking into that field more.

@@abrarsalekinraiyan3170 COmbination of nonlinear UT, reconstruction, and some AI could really be a fascinating topic ....

Thanks

Not necessarily. TFM Speed scales linearly with the number of elements and PAUT with the number of focal laws. Moreover, PAUT usually generates more ringing as the beam is directed. Therefore, the PRF of TFM could be higher. The third limiting factor is the speed of data transfer - so You need to be careful which instrument to buy.

You can find on the slide that my calculations are based on Plexiglas (2670 m/s) and Steel (5920 m/s / 3200 m/s). For immersion testing, you have to recalculate with 1500 m/s.

Because currently nobody knows what 5.0 will eventually be. However, at the end of the video you will find a short outlook of what I think it could become ...

You are very welcome...

Sure - you can purchase through ASNT: source.asnt.org/1s8ivdm/

Yes, you are right. Amplitude-based techniques can be misleading. #1 The calibration needs to be performed accurately. #2 The check of the equipment needs to be done accurately (for example if the probe frequency is not the frequency stipulated by the manufacturer DGS can be off drastically). #3 Larger, angled, indications have a quite directed reflection. Therefore, if they are not perpendicular to the ultrasonic beam only a fraction of the beam is collected by the probe. However, once we get to indications that are in the order of magnitude of the wavelength this angle dependency is quite minimal - also the structure of such small defects is quite insignificant (as this structure is way smaller than the wavelength it is not "seen" by UT. Meaning those indications can be sized quite accurately. And if all the other factors are also controlled you can achieve a surprisingly good sizing accuracy. And in case you wanna do better: th-cam.com/video/-dPkgov5E-M/w-d-xo.html

You are very welcome

You are very welcome

Great !!!

Any time!

Thanks so much!

Hi - just look for the work of Patrick Fuchs (www.ndt.net/search/docs.php3?showForm=OFF&MainSource=-1&AuthorID=38978). He has done some great work using U-Nets on Computed Tomography results.

You are absolutely correct. AI does not eliminate human factors. With AI the human factors are shifted from the inspector to the programmer/trainer. The benefit: if done correctly this can bring a great boost in reliability but if it is done incorrectly the issue will be applied consistently and all results might become questionable.

You are very welcome

Thank you

Thanks a lot

The High Frequency in the A-scans is not about the ultrasonic frequency. So-called HF (high frequency) A-scans are not rectified, are usually not compressed, and use a sampling rate that is at least 5 times the frequency of your ultrasonic frequency.

@@NDE40, awesome! That makes perfect sense to me now. Thanks a bunch!

Thank you