X-Ray Forum for NDT Users
It is currently 28.11.2021, 15:34

All times are UTC + 1 hour




Post new topic Reply to topic  [ 4 posts ] 
Author Message
PostPosted: 22.11.2019, 11:45 
Offline
User avatar

Joined: 19.10.2019, 15:47
Posts: 124
E746 is a practice with uses the concept of relative image quality response (RIQR). The results of the evaluation can be transfered to the equivalent penetrameter sensitivity (EPS) described within Practice E1025.
The basic idea of this test is to get a statistic within one IQI evaluation. Therefore a set of multi hole plates are used where the inspector has to count in a row of holes with same diameter the amount of holes which he could clearly identify. The 50% level of the plate type and row number leads to the EPS value.
Within E746 there are two versions of sets of the plates, one made of stainless steel and one made of plastic.
In E1735, which is intended for energies above 4 MeV, is a very similar set of stainless steel with other thicknesses of the plate.
Here is a photo of the four E746 plaques from stainless steel on a 19mm steel plate and an X-Ray image taken with a DDA in positiv display with a small highpass filter to emphasize the visibility of the small holes:

The four plates come in a wooden box.


The thickness of the plates decrase (yellow: E1735) .................................. and same for the hole sizes in the rows (yellow: E1735)


As it is easy to see all holes when you radiograph them directly, E746 requires that you put the stainless steel plaques on an absorber steel plate with 19mm thickness or the plastic plaques on a 35mm plastik plate made of polymethylmethacrylate (PMMA). In your own interest the absorber plates should cover the complete detector and the two surfaces of the absorber plates shall be smooth - ground finish Ra 6.35µm is required. If it would be larger it may be visible in the image as grain structure or simply noise. For E746 the acceleration voltage is set to 200kV for the steel plaques and to 30kV for the plastic version. Place the plaques on the source side of the absorber plates.
The absorber base plate for the E1735 version shall have a thickness of 15cm (6 inch) and shall be made of stainless steel. The requirements for size and ground finish are similar to E746.

The contrast in [%] for the different plaques is shown in the table above. It starts at 2% and goes down to 0.68% which is really challenging for all X-Ray techniques without DDAs. In the MeV range the requirements are even tougher ... .
In the it is shown that the visibility depends on contrast to noise and spatial resolution. To avoid that the limited spatial resolution dominates the visibility, you should have enough pixel insides the holes. The right table shows the amount of pixels which cover the same geometrical area as the hole - depended on the pixel size of the used digital system in µm. E.g. if you want to see the smallest holes with a diameter of 0.5mm you will have with a 100µm pixelsize detector 20 pixels covering the hole - and the edge pixels are inside this number.

The energy level for E746 is very restrict - chapter 6 describes a method for calibration of the X-Ray source with half value layers of copper for the 200kV and with an Aluminum plate for the 30kV version. Due to my experience the X-ray generators are quite good in providing the correct kVs - if you could prove the energy level in a different way, it is okay (see 6.5). For the MeV Sources there is no requirement to validate the energy level.


Last edited by admi on 17.03.2020, 21:20, edited 5 times in total.
added contrast in table 1 and #pixels in table 2


Top
 Profile  
Reply with quote  
PostPosted: 22.11.2019, 18:20 
Offline
User avatar

Joined: 19.10.2019, 15:47
Posts: 124
Taking the images:
For the exposure remove all filters or anything out of the beam - without the absorber plate and the plaques. To avoid unsharpness due to the size of the focal spot, the sourve detector distance (SDD) shall be a multiple of the focal spot size (FS). The absorber plate has to be placed directly on the detector, the plaques face the tube.
For the steel plaques the SDD shall be 381*FS + 19mm but more than 1m. E.g. for a 1mm Focal Spot SDD shall be the larger value of (381*1mm+19mm =) 400mm and 1m - which is 1m. Hint: As long as your focal spot is <2.5mm the 1m distance is required.
For the plastic plaques the SDD shall be 698*FS + 35mm but more than 1m. Hint: As long as your focal spot is <1.3mm the 1m distance is required.
For the MeV version the 1m minimum is also valid but additionally the geometrical unsharpness Ug shall not be larger than 0.15mm. It can be calculated with Ug=FS(v-1) with v =magnification and would lead to a minimum value for the SDD of 1000*FS+150 [mm].

Both standards require that three exposures are done with identical settings. For digital systems the digital value (Pixel Value PV) shall be within 40% to 60% of max. PV. For DDA exposures it may be difficult to reach such a high value and it is not required because DDA systems have no dose limit; in this case other PV values can be selected when approved by the cognizant engineering authority. Same situation when using digital filters as shown in the post before.


Top
 Profile  
Reply with quote  
PostPosted: 22.11.2019, 21:44 
Offline
User avatar

Joined: 19.10.2019, 15:47
Posts: 124
Evaluation of the images:
Three (different) readers should evaluate each of the three exposures independently. For each of the 14 rows with 30 holes each the three readers note in a table as shown in Fig 3 of the standard the amount of holes which were identified.
The following explanation is given for the steel plaques, it is similar with the plastic or MeV plaques.

In the example the holes from row 5 of plaque 10 are already reduces (less then the 30 of 30); the reading decreases further, at row 10 of plaque 5 only 6 of the 30 holes (0.2 or 20%) were identified. When the graph is done draw a horizontal line in the exact middle of the y-axis until you touch the graph. Follow from this point vertical to the x-axis and you get the EPS value - in the example it is between 1.33 (8) and 1.40 (7) and 1.36% seem to fit best between 1.33 and 1.40.

There is an alternate method for evaluation of the results. All values are tabled and the value left and right beside the 45 (50% of the max. 90 visible holes) shall be taken for further evaluation. Here is an example where the sum of three readers are in (in E746 chapter 9.2 only the results of one reader is taken - 15 holes out of 30):

From the example 52 (array #7) and 38 (array #8) are the correct values to select. The formula
is the standard formula for evaluation of the suitable value between two reference values (same formula as in E2002 for duplex wire evaluation).
In our example we have as first value below the 50% the array #8; array #8 has an EPS value of 1.33. As we didn't reach the 50% we have to reduce the quality (increase the EPS value) and this reduction is the term after the '+'.
C = 1.33 + [(45 - 38) * (1.40 - 1.33) / (52 - 38)] = 1.365
If we show the results in a graph, the result looks feasable. And it looks like the first diagramm and also the result value is quite similar.


Top
 Profile  
Reply with quote  
PostPosted: 22.11.2019, 22:39 
Offline
User avatar

Joined: 19.10.2019, 15:47
Posts: 124
Here is an example where the dose limit of the detector has happened. Left just one frame of a good DDA and right 60 frames integrated (with same conditions):

(the images are shinked by factor of 2 for this forum).

And here is another example, where the imaging system was noisy (with noise also on the lead letters ...):


No digital filter applied on the images.


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 4 posts ] 

All times are UTC + 1 hour


Who is online

No registered users


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
cron



Hosted by iphpbb3.com

Impressum | Datenschutz