where expertise comes together - since 1996 -

The Largest Open Access Portal of Nondestructive Testing (NDT)

Conference Proceedings, Articles, News, Exhibition, Forum, Network and more

where expertise comes together
- since 1996 -

GE Inspection Technologies
Inspection Technologies, a business of the Baker Hughes, a GE company (BHGE IT), is one of the world's leading suppliers of nondestructive testin ...
926 views
Technical Discussions
Martin Callaghan
Other,
USA, Joined Jan 2009, 3

Martin Callaghan

Other,
USA,
Joined Jan 2009
3
22:10 Apr-14-2009
Phased Array inspection of Circ Welds 45mm W/T

Just wondering on what the ideal acquisition rate would be set to carry out a sectorial scan on welds this size.

The welds are been scanned from 90 and 270 to look for cracking.

The people carrying out the scans are using an Omniscan MX

    
 
 Reply 
 
Roger Duwe
NDT Inspector, API-510, 570, 653
MISTRAS, USA, Joined Jan 2009, 148

Roger Duwe

NDT Inspector, API-510, 570, 653
MISTRAS,
USA,
Joined Jan 2009
148
03:11 Apr-15-2009
Re: Phased Array inspection of Circ Welds 45mm W/T
In Reply to Martin Callaghan at 22:10 Apr-14-2009 (Opening).

IMHO, they are using the premier instrument. Good start. Hopefully they are using a 64-element linear probe, like the '5L64', or bigger. More elements give better resolution.

Next, have them map out their Scan Plan graphically, to verify coverage. It should be possible to get full coverage with 2 passes on each side of the weld; one indexing against the cap [or very close to it], one indexing about 100 mm back from the center of the weld.

Third, acquisition is 3-dimensional. The sectoral sweep should be between 1-degree and 0.1-degree increments. I would use 1/2-degree, myself. The aquisition from the linear encoder should probably be set to 1 mm increments, certainly no more than 2.5 mm increments.

Good luck and good flaw hunting.

    
 
 Reply 
 
Ed Ginzel
R & D, -
Materials Research Institute, Canada, Joined Nov 1998, 1268

Ed Ginzel

R & D, -
Materials Research Institute,
Canada,
Joined Nov 1998
1268
13:43 Apr-15-2009
Re: Phased Array inspection of Circ Welds 45mm W/T
In Reply to Martin Callaghan at 22:10 Apr-14-2009 (Opening).

Martin...the acquistion rate is a bit of a moving target. This will dictate the number of times per second that the waveform is moved on to the memory or display. Soe equipment can be "instructed" to maximise the acquisition rate. The number of acquisitions "possible" will mainly be a function of main clock rate, time of the digitisation window, analogue to digital conversion (ADC) rate, number of focal laws and any averaging or pre-processing being done on the signals. If the system is encoded the acquisition "cycle" will need to be adequate to ensure that all the functions are carried out prior to moving to the next encoded position, i.e. higher acquisition rates are required for higher scanning speeds.
If you are addressing 2 probes simultaneously the demands on the timing are increased so the acquisition rate will probably need to be reduced.
Contrary to other advice, number of elements has virtually nothing to with the acquisition rate and more elements does not necessarily mean better "resolution". The angular resolution in a sectorial scan should consider the spot size at the point of interest in the swept plane. For an unfocused focal law (as you would use for most weld inspections) it would generally be adquate to sample at intervals of the 6dB spot size which for a 10mm aperture would be approximately 2.5mm. At a 100mm path length th 2.5mm increment would be approximately 1.4°. More than that and you MIGHT miss a peak amplitude but sampling under 1° at that sort of path length is a waste of memory and operating speed.

    
 
 Reply 
 
martin callaghan
Other,
USA, Joined Jan 2009, 3

martin callaghan

Other,
USA,
Joined Jan 2009
3
14:43 Apr-15-2009
Re: Phased Array inspection of Circ Welds 45mm W/T
In Reply to Ed Ginzel at 13:43 Apr-15-2009 .

Hello Roger and Ed,

Thanks for getting back so quickly. We are using a 5L64-A2 on a N55-IHC wedge curved to the OD. The acquisition rate has been upped to 2mm and this has helped to reduce the amount of lines of mising data on the saved scans.

The scans and stand offs have been predetermined by Beamtool3 to ensure coverage of the weld volume and the HAZ and a single sectorial scan is been taken in each case. We are taking one encoded scan at 90 and the other at 270 (but not simultaneously) 40 - 70 degrees with 16 elements starting from element one (unfocussed)


We have noticed a feature on the Omniscan in that whilst scanning and monitoring the C Scan it appears that there is no missing data but once the freeze button is hit some black lines immediately show up. What is the cause?


Ed wrote -

"For an unfocused focal law (as you would use for most weld inspections) it would generally be adquate to sample at intervals of the 6dB spot size which for a 10mm aperture would be approximately 2.5mm. At a 100mm path length th 2.5mm increment would be approximately 1.4°. More than that and you MIGHT miss a peak amplitude but sampling under 1° at that sort of path length is a waste of memory and operating speed"

We will try 2.5mm tomorrow and check the results.

One other question on PAUT. Whilst calibrating I have seen a situation where an echo is been returned from the front of the wedge face (the face that couples to the block)
While collecting the TCG points this spurious signal prevents one of the TCG points from been used on the TWI PAUT CAL block. In other words the cursor "locks on" to the signal from the front of the wedge rather than the hole in the TWI block. (Yes the gate is set on the correct hole during calibration!!) Can anyone provide an explanation?
Its baffling because the wedge is brand new and is the transducer is properly coupled to it without air bubbles?!

    
 
 Reply 
 

Product Spotlight

NDTkit RT

NDTkit RT, TESTIA's Digital Radiography software The NDTkit product line software for X-ray analysi
...
s. NDTkit RT is a software benefiting from the Ultis kernel which is dedicated to radiographic image analysis. It offers a set of tools and filtering processes to assist RT operators in finding relevant flaws.
>

High-performance Linear Phased Array Probes

Available to order from stock in a range of 5MHz – 7.5MHz and from 16 to 64 elements. Designed w
...
ith piezo-composite elements, Phoenix phased array probes provide high-resolution imaging to maximise sensitivity; accurate ultrasonic detection and sizing of defects in welds; and effective corrosion mapping. Housed in a rugged stainless steel case for on-site industrial NDT applications.
>

EKOSCAN Phased Array

In order to always fit your needs, EKOSCAN can manufacture any type of UT transducer, either convent
...
ional or Phased Array. As an ISO 9001: 2015 certified company, EKOSCAN is extremely careful as far a material selection and manufacturing processes are concerned. Our probes guarantee our customers the benefits of latest innovations regarding piezo-composite, backing, impedance adaptation layer, etc. Specific probes for hostile environment: high temperature, high pressure, corrosive environment,etc. Specific probes designed to fit your specific application: optimization of every parameter to guarantee you the best detection.
>

PROlineTOP Plug & Play Ultrasonic inspection device

As Plug & Play solution it units all control and operation elements in a small housing and therefo
...
re replaces the typical control cabinet...
>

Share...
We use technical and analytics cookies to ensure that we will give you the best experience of our website - More Info
Accept
top
this is debug window