ASNT Fall Conference 2002 San Diego CA

Leak Testing in Alaska

Ken L. Heaps
K2 Technologies
3400 Glenn Don Cr
Anchorage AK 99504
907-333-1666

Cover
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Introduction:
Leak Test Applications in Alaska
Vacuum Box Testing
Close Calls
Vacuum Box Failures
Vacuum Box Construction
Interpretation of the Vacuum Box View Pane

Intro
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Leak Testing in Alaska

¾ Pressure Change Testing
9 To determine leakage on a closed component or system
9 Air or water used as test medium (hydro or pneumatic)
9 Test pressures up to 25% of design or operating pressure

¾ Bubble Testing
9 Vacuum Box testing in the petrochemical industry (generally
performed on tank floor plates)
9 Bubble Testing (inflatable water craft)

¾ There is not a lot of other Leak Testing that takes place in Alaska

LT in AK
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Vacuum Box Testing in Alaska

¾ Primary application is new tank floor construction in the
petrochemical field
¾ Has been performed on ‘old construction’; i.e, painted lap
welds to confirm leakage conditions on tanks that are being
taken out of service
¾ Estimate approximately 10 relevant indications detected per
mile of weld vacuum box tested
¾ Average about 70 linear feet of weld per hour per crew of
vacuum box testing
¾ Use of low temperature bubble solution to test down to +20o F

VacBox in AK
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Close call…
¾ While vacuum box testing lap welds on fuel tank floor plates a close visual
examination (with improved lighting) reveals a layer of heavy sediment adhered to
the lap joint fillet welds
¾ A mechanical wire brush was required to remove sediment layer
¾ Once removed, indications were detected during the second vacuum box test over
the same weld

Pay close attention to…

¾ Surface Preparation (cleaning)
¾ Visual Examination
¾ Adequate Lighting (50 ft candles per ASME V Article 9)

Close Call
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Close call…
¾ While vacuum box testing butt welds on tank roof plates a depression area is noted
in the vicinity of the weld being tested
¾ Under the low test pressure (2 psi pressure change) a two inch crack is detectable
¾ Under the high test pressure (8 - 10 psi pressure change) the indication disappears
¾ The high test pressure created enough vacuum to reduce the depression area
enough to close off the air flow through the crack

Pay close attention to…

¾ Low test pressure hold times specified in the procedure (recommend no less than
15 seconds)
¾ Visual Examination

Close Call
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Close call…
¾ While vacuum box testing lap welds on tank floor plates a small relevant
indication is noted after a 15 second hold time at the high test pressure
¾ The indication was not visible during the first 10 seconds of the hold time.

Pay close attention to…

¾ High test pressure hold times specified in the procedure (recommend no less than
15 seconds)

Close Call
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Close call…
¾ While vacuum box testing lap welds on tank floor plates there were no bubble
indications evident in the film solution
¾ The viewing pane of the vacuum box exhibits peculiar patterns
¾ Gross indications were blowing through the film solution, even at the low test
pressure, and creating airflow patterns on the inside of the viewing pane

Pay close attention to…

¾ Interpretation of the inside of the viewing pane - specify this in the procedure
¾ If possible, watch the welding process - fast weld travel speeds often result in
gross indications

Close Call
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Vacuum Box failures…

¾ Faulty relief valves on gages can cause implosion of the
Plexiglas pane
¾ Localized stresses embrittle the grain structure on Plexiglas
viewing panes, especially in corner boxes, leading to
implosion
¾ All implosions have occurred on short Plexiglas panes on cast
aluminum bodies, primarily corner boxes

VacBox Failures
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Vacuum Box Construction (Alaska style)…

¾ Viewing pane – Lexan Polycarbonate
9 The longer the box, the more easily it conforms to irregular surfaces
¾ Body – Closed cell foam 2” h x 2” w makes up the body and the gasket
9 Do not cut bottom of foam at angle - leave square
¾ Gage – incorporate a good relief valve
9 Aids in interpreting at different pressures
Lexan polycarbonate viewing panes offer
9Good light transmittance
9Excellent impact strength
9Good weatherability
9Ease of machining
9Thermoformability
9UV resisting

Construction
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Vacuum Box in operation

Seal
Power
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Operating Vacuum Box at low

and high pressure differentials

Gage
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Pressure Conversion Table

It’s a great reference for the technician when he/she’s in the field and the
units on the gage are not in a familiar format. Put it in your procedure!

PSI(g) kg/mm2 inches Hg Torr kPa

(kN/m2)

2 0.0014 4.07 103.43 13.79

4 0.0028 8.14 206.86 27.57

6 0.0042 12.22 310.29 41.36

8 0.0056 16.28 413.72 55.15

10 0.0070 20.36 517.15 68.94

12 0.0084 24.43 620.58 82.73

Conversions
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Interpreting the inside of your view pane….

It’s a good idea to include it in your procedure!
¾ Look for circular patterns, or ripples, created by air flow on
the inside of the pane. These indications are from gross
discontinuities that break through the bubble solution and
otherwise cannot be detected.

ViewPane
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Interpreting the inside of the viewing pane

Two
Indication
s
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Gross indication blowing through bubble solution and

creating a circular pattern on the inside of the viewing pane.

Gross
Indication
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Vacuum Box Sensitivity

¾ Bubble testing can detect leaks in the laminar

flow range of 10-2 std cc/sec to 10-5 std cc/sec

¾ By interpreting the viewing pane, the higher end

of this range can be extended

Sensitivity
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Administering practical examinations

Shop Class
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Unique to Alaska…
¾ Request for bids that contain vacuum box testing in remote
locations often require a ‘bear watch’ for the ‘man watch’
¾ Technicians have watched bears wander into tanks and
wander back out
¾ There have been no vacuum box technician fatalities
associated with bears in Alaska

RFQ
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Bear 1
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Bear 2
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Bear 3
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Statewide, there may be as much bubble testing on

personal inflatable watercraft as vacuum box testing
on tank floors
¾ But most of this testing is performed remotely, with
unqualified technicians

RFQ
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Bubble Test specimen in Turnagain arm at low tide

Bubble Test specimen floating the Kenai river, watch

out for that moose!

Boat 1
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

Deflated Bubble Test specimen flown into

remote airstrip in the Wrangell Mts

Bubble Test specimen floating the Nebesna river, Wrangell Mts AK

Boat 2
ASNT Fall Conference 2002 San Diego CA Leak Testing in Alaska

There’s generally bubbles coming out of this boat all the

time!

Boat 3