3../../../ Radiographic Investigations of Micro-Ruptures in Technical Diaphrams

NDT.net Home
  • Home
  • Database Search

  • This simple HTML-Text version serves as a quick to access preview of the PDF file Simple HTML-Text Preview of Open the PDF file of the Article PDF-File [272KB]

    Source: DGZfP 2005 [more]
    German Society of NDT - 2005 - Rostock [Home]


    DGZfP-Berichtsband 94-CD

    DGZfP-Jahrestagung 2005

    Plakat 11

    2.-4. Mai, Rostock

    Radiographic Investigations of Micro-Ruptures in Technical Diaphrams

    D. Thomsen1), T. Kurz 2), R. Kaesler2), T.M. Buzug1)

    1) Department of Mathematics and Technology, RheinAhrCampus

    2) Integral Accumulator, Remagen

    Abstract

    Accumulators are used in lots of applications. So they are used as a pressure accumulator for quick energy sup- ply to hold a constant pressure (equalization of leakage), for pulsation damping and as an element of suspension. We can differ between 3 types of accumulators: 1. Diaphragm Accumulator, 2. Piston Accumulator 3. Metal Bel- low Accumulator. At applications with high requirements to permeation the diaphragm accumulators are equipped with so called multi-layer diaphragms. During test series some of these accumulators failed because of big gas losses. At a first look, no external damage could be seen, so that the damage was assumed in the internal layer.

    In this contribution the results of a detailed X-ray based non-destructive testing (NDT) of accumulator mem- branes are reported. This includes digital X-ray fluoroscopy using a Philips Computed Radiography (PCR) sys- tem as well as conventional and micro CT investigations.

    1

    Introduction

    aggressive liquids and prevent the PVA-foil compo- nents from diffusion.

    Accumulators were originally developed as energy accumulators. Modern hydro-accumulators are also used for damping, to balance out leakage and volume and for pulsation damping. There are diaphragm and piston accumulators and valve units used in conjunc- tion with diaphragm accumulators.

    In extreme situations a membrane leakage has been observed in a testing environment. A first optical in- spection did not show any damage of the issue. Thus, due to the multi-layer structure of the membrane an inner-structure damage is suspected to be the cause of the leakage.

    The internal structure of the multi-layer diaphragms is shown in figure 1.

    2

    X-ray Fluoroscopy

    As a first low-cost easy to use method in the investi- gation of gas losses, we used a digital X-ray fluoros- copy system. In Figure 2 the setup and the data of the PCR system are shown. The data do not reveal any significant difference between a new diaphragm (up- per part) and the damaged diaphragm from the test series (lower part). It shows no details about the in- ternal structures of the component parts and espe- cially no indication about the cause of the gas loss. The reason for this is the low contrast of X-ray fluo- roscopy systems compared to CT X-ray systems.

    Figure 2: Setup of the X-ray fluoroscopy system with a digital cassette (resolution 2364 x 2964 pixel) and data of reference membrane (top) and membrane after test series (bottom).

    multi-layer diaphragm

    barrierfoil

    Material Function

    PVA

    gas barrier

    EVOH 1. shields the PVA-foil from chem. agressive liquids

    2. hinders the diffusion of PVA-foil components

    EVOH

    Rubber: NBR, IIR etc.

    Rubber

    PVA

    Figure 1: Overview of multi-layer diaphragm assembly.

    The diaphragm used in the accumulator consists mainly of nitrile rubber (NBR) and contains an in- ternal layer of polyvinyl alcohol (PVA) which works as a gas barrier. This layer is covered by two thin lay- ers of ethylene vinyl alcohol copolymer (EVOH) which have to shield the PVA-foil from chemically


    DGZfP-Berichtsband 94-CD

    DGZfP-Jahrestagung 2005

    Plakat 11

    2.-4. Mai, Rostock

    3

    Conventional CT

    4

    Micro CT

    4.1 Spatial Resolution

    To combine high contrast and high spatial resolution a micro CT was used.

    The relevant components of such CT systems are mi- cro focus X-ray tube and a high resolution 2D-CCD detector array. The spatial resolution is limited by the size of the focus (see figure 5) and the size of the detector elements.

    Anode

    To increase contrast we used a Philips Secura CT, a conventional medical spiral CT-System. The 3-D re- construction of the CT-data in figure 3 shows differ- ent areas of reduced absorption which can be divided into two categories:

    - large detachments of the inner rubber layer mainly in the center of the diaphragm (see also figure 4 - section through the center of the membrane)

    - multiple isolated lineaments located in the concentric strengthened middle part of the diaphragm. These lineaments are orientated on meridians.

    With their high contrast the CT-data reveal detailed information on structural irregularities in the mem- brane. However, for further investigations of the lineaments presented in the dataset a higher spatial resolution is necessary.

    Figure 3: 3-D reconstruction of the CT data of reference membrane (left) and membrane after test series (right).

    Figure 4: Section from the center of the membrane.

    blurring

    X-ray focus

    D e t e c t o r

    "deepest shadow"

    blurring

    Cathode

    Figure 5: The inclination of the anode controls the opti- cal size of the X-Ray focus. A larger X-ray focus in- creases the blurring of the image.

    The so called modulation-transfer function MTF gives the resolution for spectral frequencies q in line pairs per millimetre [1]:

    q b

    =

    Typical focus sizes are bf = 1 mm for conventional medical CTs and bf < 10 μm for micro CTs.

    Figure 6 shows the characteristics of the MTF of a conventional medical CT and a micro CT, respec- tively.

    101

    π

    MTF

    system

    ) ( detector focus

    MTF q

    ) (

    MTF q

    sin( ) (

    q

    π ) sin(

    )

    q b

    q b

    =

    π

    q b

    π

    F

    F

    D

    D

    1

    0.9

    0.8

    0.7

    MTF(q)

    0.6

    0.5

    0.4

    0.3

    0.2

    0.1

    q / (lp/m)

    0

    102

    103

    104

    105

    Figure 6: Frequency dependent modulation-transfer- function of conventional CT (doted curve) and micro- CT (dashed curve).


    DGZfP-Berichtsband 94-CD

    DGZfP-Jahrestagung 2005

    Plakat 11

    2.-4. Mai, Rostock

    As a consequence of the small focus size of micro CTs the anode current is limited to I < 100 µA. This affects the intensity of the X-ray spectrum and limits the range of possible probe materials.

    The micro CT by the Belgian company SkyScan shown in figure 7 uses a 12 Bit CCD-Chip with a pix- el matrix of 1024 x 1024 and a pixel size of about 10 µm which is coupled via fibre optics to a scintilla- tion crystal. In conventional CTs the detector dimen- sion bd is typically found in the order of 1 mm. SkyScan gives a specification of spatial resolution of 10 µm for the 1072 scanner used in the experiments shown here [2,3,4].

    Figure 7: Setup of micro-CT.

    4.2 Imaging Results

    Figure 7 displays the setup of the micro CT which can be used to investigate objects up to a size of 2 cm3.

    The 3-D reconstruction of the micro CT data of one of the lineaments shown in figure 9 reveals that there are ruptures especially in the covering EVOH layers.

    Figure 8: Single tomographic section of the mem- brane rupture.

    Figure 9: 3-D reconstruction of the membrane rupture.

    5 Microscopic Evaluation

    The microscope pictures in figure 10 show a part of the inner EVOH-layer in different enlargements. The higher magnification in figure 10b clearly indicates that the material becomes brittle and fragile along the fold structures.

    Figure 10: Microscope picture of the fold structure (width approx. 1 mm).

    6 Resume

    The production of the membranes apparently leads to folds in the internal EVOH/PVA-layer. The cause of gas losses are ruptures in these folds.

    References

    [1] T. M. Buzug, Einführung in die Computertomogra- phie, Springer-Verlag, Heidelberg, 2004.

    [2] A. Sassov 1999, Desktop X-ray Micro-CT, in: Proc. of the DGZiP BB67-CD, Computerized Tomography for Industrial Applications and Image Processing in Radi- ology (Berlin) 165.

    [3] A. Sassov 2002, Desktop X-ray Micro-CT Instru- ments, Proc. of SPIE 4503, 282.

    [4] A. Sassov 2002, Comparison of Fan-Beam, Cone- Beam and Spiral Scan Reconstruction in X-Ray Mi- cro-CT, Proc. of SPIE 4503, 124.