12:09 Nov-07-2009 Bill Chestnut NDT Inspector, Self Employed, United Kingdom, Joined Mar 2006 3
Phased Array Probe Checks
When carrying phased array probe performance checks, like a conventional ut probe we can check for amplifier linearity, timebase linearity signal to noise and so on, but is there any hard and fast guidlines for the following,
1. Number of allowable dead elements within an array,
2. The power output of the individual elements within an array,
I can recall recomendations such as no more than 25% of elements within an array shall be inactive and no more than 2dB difference in output for elements across the range.
can anyone advise if any of the standards cover this in any detail,
14:36 Nov-07-2009 Ed Ginzel R & D, - Materials Research Institute, Canada, Joined Nov 1998 1198
Re: Phased Array Probe ChecksIn Reply to Bill Chestnut at 12:09 Nov-07-2009 (Opening).
Bill: There is an ASTM Standard (E-2491) that describes how to go about several checks on phased array systems. But quantitative limits are not stipulated. That is left to the users and contracting parties. The guidelines that you refer to are reasonable, but until the specifics of the application are determined, guidelines are all that can be given. Lack of uniformity of individual element output and in-active elements have a similar effect on the phase-generated wavefront. The effect on the application may or may not be noted during calibrations and it is the calibrations that ultimately determine suitability. 25% of the elements missing when scattered over a 32 element aperture may have virtually no effect on an unfocused beam but if the 25% was grouped at one end and the beam was highly focused then there could be a significant effect. But this concern for phased-array probes should not be considered any different from the conditions that could cause similar problems in monoelement probes. I have seen grit trapped between the probe and refracting wedge causing a score on the probe-face that would disrupt the wavefront generated and the same weakening and distortion effect can occur as the coupling between the probe face and refracting wedge dries out in areas.
An in-house programme to verify quality of new probes can verify the manufacturer's report. After that, instability of your calibrations will probably be your first indicator of concern.
As a point of experience, I have found that the probe elements themselves rarely fail. More often I have seen the cables and even the connector pins as the source of variation. In one situation I ran an element activity check and counted 6 or 7 elements not active. I put the same probe in the other connector of the Y-splitter and a different set of elements was inactive. The problem was in the phased-array unit itself and only required that the connectors be better seated. So even when you have identified a problem with the element activity, throwing out the probe may be an expensive and inappropriate solution.
To add to Ed's thesis on dead elements, EPRI did a study on matrix and 1D arrays and showed that up to 25% of the elements could be lost without major degradation. Not surprisingly, it did depend on the locations of the elements in the array. For practical purposes, it should be feasible to perform regular calibration checks on the array, and watch the cal amplitude drop, the noise rises, the beam start to deteriorate etc. as the elements die. Incidentally, Ed is correct in stating that most of the problems come from the cabling, not for the elements themselves.
10:05 Nov-09-2009 Dave Lines R & D, Diagnostic Sonar Ltd, United Kingdom, Joined Nov 1998 8
Re: Phased Array Probe ChecksIn Reply to Michael Moles †2014 *1948 at 15:13 Nov-08-2009 .
I would echo both Ed's and Michael's comments but with one caveat. If the beamformer is using aperture shading for optimizing the beam profile, then the degradaton in the beam's characteristics usually starts at a much lower threshold than 25% of elements.
Please can you, or any other learned person, explain what is "aperture shading", its effect, whether it can be controlled by commercial phased array equipment and if so how to specify it in a written test instruction or check its use in a verification circumstance.
Aperture shading performs a similar function to signal processing window functions (such as Hamming, Hanning, Gaussian windows) in that it attempts to reduce spurious responses. See http://en.wikipedia.org/wiki/Window_function for a quick outline description.
In the case of an ultrasonic array, the aperture shading is a weighting function that changes the contribution from an element so as to reduce grating lobes. Typically the shading function progressively reduces the contribution from the elements as you move away from the centre of the aperture. Whilst reducing lobes, it also reduces the effective aperture and hence lateral resolution. You pay your money and take your choice!
Aperture shadings are relatively easy to implement on receive but are significantly more difficult on transmit. The best way to establish if your instrument uses shading, and whether it is on transmit as well as receive, is to ask the manufacturer.
The original topic was about the effect of the loss of one or more elements. It is the smooth variation of the shading that controls the lobes and so losing an element at the edge of a shaded aperture will have less of an effect on sensitivity and lobe performance than losing one close to the centre. Modelling of the beam profiles allows the nature and extent of the degradations to be checked.
Our instruments have a self-test mode that allows identification of a faulty channel or transducer element but I'm not clear how widespread this is across other manufacturers. This is a system check and so won't identify local loss of coupling beneath the array whilst scanning which, as the other contributors pointed out, will affect the beam profile of conventional transducers as well as arrays.
hello every one, we have 32 element PAUT probe. probe description is,5L32 PWZ3('Olympus' make). as per the manufacturers recommendation,it is only for sectorial scan not linear scan.but, according to ASME Section V Article 23,SE 2491, it recommends, to configure with Electronic (linear)scan to determine dead element.can anyone please help me to solve this issue....which means... is any other standardized ways to check dead element..
That recommendation is probably based on the low element count and large element size and for scanning / inspection purposes. It will not hurt the probe to do a linear scan and it will accomplish the purpose of an element verification test.
Paranthaman, as Nicholas notes, the small aperture is probably the reason the manufacturer does not "recommend" it be used for linear scanning for inspection. But do not confuse inspection requirements with the dead element check function. The linear scan required for the element check is a 1 element delay-law with 1 element steps over the entire length of the array. In a B-scan display for such a configuration you will get a single A-scan line for each element. With the probe placed on a wedge or on a block you will get a backwall signal and you need only look at each backwall signal to determine if the element is active. Figure A3.1 in ASTM E2491 is reproduced in ASME Art. 23 SE 2491 and you can see an example of the display used.