Manufacture and Testing of a Large Zirconium Clad Vessel

1
Manufacture and Testing of a Large Zirconium Clad
Vessel
David Clift, P.Eng. Production Manager Ellett
Industries September 14, 2005
2
Abstract

• Serviceability depends upon the quality of the
  clad corrosion liner
• Careful application of design details, skilled
  trades systematic application of manufacturing
  controls and non-destructive examination.

3
Zirconium Clad Steel Reactor
3.6m diameter x 6.5m length (11 10 x 21 2
Shells are 22mm (7/8) thick SA516-70 with 3mm
(0.12) nominal, 2.28mm (0.090) minimum,
thickness Zirconium 702 explosion clad liner. The
hot pressed heads are of 37mm (1.5) SA516-70
with a 4.7mm (3/16) nominal thickness of clad
liner
4
Design Details
Separation of Longitudinal Circumferential Welds
Advantages

• Containment of possible breaches
• Efficient gas purging
• Simplified helium testing

5
Material Procurement

• Base Metal N.D.E.
• ASME SA 578 Level C
• grid pattern or continuously scanned - grid
  pattern is standard
• three inspection levels
• A- least , B - moderate, C - most demanding
  allows a discontinuity smaller than can be
  contained within a 25mm circle
• Cladding N.D.E.
• full compliance with ASME Section VIII IX
• production bend tests, PT examination and 100 RT
  before bonding
• PT after bonding and, in the case of the heads,
  after forming. All clad surfaces were visually
  examined

6
Ultrasonic Examination of Bonding
ASTM B 898 Acceptance Criteria

• Class C is the standard Inspection Class of B
  898
• Class B was specified for this project

7
Explosive Cladding

• Current Practice
• an interlayer of titanium when zirconium cladding
  exceeds 6.4 mm (1/4) nominal thickness
• Explosion detonation (booster) locations
• corner, side or center of plates
• an area of ultrasonic non-bond is typically
  located under the detonation point
• non-bond related to the size of charge

8
Vessel Plate Manufacture

• Central explosion detonation points for both
  shell and head plates on this unit
• thin cladding, thick backing plates
• ASTM B 898 Inspection Class B ultrasonic
  inspection (75mm maximum indication size)
• Results
• non-bond areas in shell plate were acceptable
• head plate detonation points were removed as
  cut-outs for centrally located nozzles

9
Plate Surface Defects

• Damaged areas of clustered gouges were noted on
  three of the six shell plates
• Gouge depth ranged from 1.2mm (0.047) to 2.4mm (
  0.094) deep
• Gouge size varied less than 13 cm2 (2 in2) in
  area
• Root cause - rock fall that occurred during
  underground explosive blasting

10
Repair Plan

• Weld repair
• shallow gouges (lt1.2mm deep) were weld repaired
  in conformance with ASTM B 898.
• Controls
• customer approval
• carbide burr removal
• qualified weld overlay repair
• PT inspect
• fully documented

11
Repair Plan - contd

• Deep gouges
• Individually ported batten style covers

12
Vessel Seam Weld Test Sequence

• Carbon steel welds are applied and specified NDE
  performed prior to batten strap attachment
• UT the cladding bond adjacent to the end of the
  longitudinal seam filler strips to find potential
  leak paths
• Fit weld longitudinal batten strips, silver
  braze to isolate and helium bubble test _at_ 1 bar
  pressure
• Circumferential welds performed in a similar
  manner
• Confirming PT of all ZR welds per ASME

13
Fabrication - Cleanliness

• Tool surfaces to be of alloy, plated or hardened
  steel
• Rolls brake forming surfaces - confirmed free
  of all surface defects contamination
• Weld joint design should minimize carbon steel
  welding grinding on process surfaces
• Zirconium weld zones are mechanically and
  chemically cleaned prior to welding

14
Final Cleaning

• Contaminated areas abrasively ground
• Acid wash with HF/HNO3
• Ferroxyl test per ASTM A 380
• spray application of the potassium ferricyanide
  test solution on suspect areas
• reacts with free iron to form a blue indication

15
Helium and Hydro Testing

• Helium Mass Spectrometer testing per ASME Section
  V, Article 10, Appendix IV was performed before
  and after hydro testing
• Purging batten strip cavities
• helium from bubble testing conducted up to a
  month earlier had to be purged with clean dry
  air to eliminate false indications
• Test conditions
• 25 of the vessels maximum allowable working
  pressure (50 psig)
• approx 20 helium concentration

16
Test Results

• Results
• All purge vents located behind nozzle liners,
  seams and internal batten style patches were
  vacuum sniffed
• All readings remained at a background or
  atmospheric helium concentration - equivalent to
  a leak of 5 x 10-6 atm cc/sec

17
Additional Research

• Preparation and analysis of a zirconium weld
  overlaid coupon to assess built-up metal quality
• CS - TI - ZR construction

After welding and step machining
Prepared block ready to weld overlay
18
Results

• ZR overlay welding on both the titanium
  interlayer and the zirconium cladding, min
  thickness .76mm
• Test results confirm a composition containing
  titanium and zirconium
• No detectable iron was present in any of the test
  samples
• 100 Zr when a 0.76mm thick layer was applied
  over a combined thickness of Ti and ZR of 3.05mm.
  The actual thickness of the ZR alone in this case
  was 1.3mm

19
Questions