Having been recently involved deeply in this subject, let me give you the long version - bear with

me all pipelines, flowlines and risers fabricated in accordance with API 1104 (all welds on API 1104
pipelines, flowlines and risers must be 100% radiographed or ultrasonically tested), must be
hydrotested in accordance with API 1110 (maybe other codes depending on your location), which in
turn refer to ASME B31.4 or 31.8.
ASME 31.4 permits a "waiver" of a hydrotest on tie-in welds between two sections of piping that have
been previously tested in accordance with the code provided that the weld has undergone 100% NDT
(see Section 437.1.4 (b) Testing Tie ins, ASME B31.4 2002 Edition).
Now, if you are talking about plant piping, very similar. All piping fabricated in accordance with ASME
B31.3 must be hydrotested in accordance with ASME B31. The latest edition of B31.3 permits the same
waiver of a hydrotest as 31.4 on Tie-in welds between two sections of piping that have been
previously tested in accordance with the code.
ASME B31.3 only requires 5% random radiography (or UT) on piping in Normal Fluid Service (whihc
means hydrocarbon service up to 2500 flange rating)
You didn't ask, but will throwi n that API RP14E, only recommends 10% radiography on class 150 to
600lb ANSI rated flanges, 20% on 900 to 1500lb ANSI rated flanges and 100% on 2500 and above.
Most companies will have a procedure to perform a "golden weld", but at the end of the day it's what
we always referred to as a final tie-in and basically you do a 100% RT or UT NDT on the weld in lieu of
pressure testing it.
Do you think I have recently been involved with these?
Greg Lamberson
Consultant - Upstream Energy
Website: www.oil-gas-consulting.com

MJCronin (Mechanical)

6 Apr 07 9:29

LBouzid....
There are no ASME recommendations, but there are usually many recommendations from the company
PR office....
Always have plenty of room around the "golden weld" location for dignitaries, corrupt politicians and
evil lawyers, hangers-on and wannabees. Set up a refreshment area... Put up a tent for the press. Give
the poor welder a bath and money for new clothes. Contact the local high school marching band.....
See page 4 of this link:
http://www.bp.com/liveassets/bp_internet/globalbp/globalbp_uk_english/reports_and_
publications/frontiers/STAGING/local_assets/downloads/bpf11p01-05contsnews.pdf
And this one...
http://azer.com/aiweb/categories/magazine/ai124_folder/124_articles/124_bp_develop
ments.html

-MJC

dcasto (Chemical)

6 Apr 07 14:50

I was at a TRRC seminar on pipeline safety and during the Q&A at the end, someone asked about a
repair of a line with a joint of tested pipe and the inspection as prescribed. His question came down to
this, how many joints could he install in a row before he needed to a hydrotest?
The reponse was great. The head of the commision said, " you could build a 100 mile line from scratch
with that technique and we could not stop you. But we'd find 1000 other things to make your life so
unbarable you'd wish you would have. (like show me you written policy on the subject and we will
hammer it for years)"
So there is a question of Generally Accepted Engineering Practices that will come into play and the
owners risk assesment.
GregLamberson (Petroleum)

7 Apr 07 0:52

Bottom line is, theoretically, you can install a pipeline without a golden weld, but most often the case is,
you cannot. Either you have more than 1 test section, you don't for some reason, have flanges on each
end, there are a variety of reasons that require a golden weld - they are the norm, not the exception.
There is nothing wrong with golden welds and they are allowable by code (see Section 437.1.4 (b)
Testing Tie ins, ASME B31.4 2002 Edition).
Greg Lamberson, BS, MBA
Consultant - Upstream Energy
Website: www.oil-gas-consulting.com

GasGuy (Mechanical)

20 Apr 07 1:02

GregLamberson: You mentioned "The latest edition of B31.3 permits the same waiver of a hydrotest as
31.4 on Tie-in welds between two sections of piping that have been previously tested in accordance
with the code."
I have been unable to find this in B31.3-2004. Could you give me a reference please?
GregLamberson (Petroleum)

20 Apr 07 8:11

GasGuy
My apologies, I have the 2002 edition, I don't have the 2004 edition, but would be surprised if it did not
contain the same provision. In it, section 345.2.3 Special Provisions for Testing states..."The final weld
connecting piping systems or components which have been successfully tested in accordance with para.
345 need not be leak tested provided the weld is examined in-process in accordance with para. 344.7
and passes with 100%
radiographic examination in accordance with para. 344.5
or 100% ultrasonic examination in accordance with
para. 344.6...."
Greg Lamberson, BS, MBA

Consultant - Upstream Energy

Website: www.oil-gas-consulting.com

25362 (Chemical)

20 Apr 07 9:59

GregLamberson, not belonging to this engineering field, kindly confirm for my own education whether a
"golden weld" means a tie-in that has been 100% radiographed, or 100% U/S inspected, or subjected
to a magnetic particle inspection ?
GregLamberson (Petroleum)

20 Apr 07
10:34

A golden weld is a weld that has not been included in the hydrotest for a variety of reasons, but has
been subjected to complete NDT.
We used to refer to these simply as final tie-ins, which is a much more accurate description. There is
nothing special nor golden about the welds.
Greg Lamberson, BS, MBA
Consultant - Upstream Energy
Website: www.oil-gas-consulting.com

Gator (Industrial)

20 Apr 07
23:43

Pasted below (in no particular order) are some relevant comments posted to my PipingDesign group
from February this year. I'm not sure if this addition will help.
The group is a closed, email-based list, so I can't just post a URL.
Paul
www.pipingdesign.com
=======================================================
Hi Guy's,
with regard to the 'Golden Weld' theories
going around, the latest bit of info is that such
welds are now called 'Closure Welds' as the old term
is not representative, in this technically correct
world of today.
Anyway, being an 'old bod' I will stick to the term
for the sake of continuity.
In my experience with Golden Welds is that they were
always treated with the utmost repect for the obvious
reason, that they were not proof tested by going
through the hydrotest process.
As for the substitue measures that can be taken with
such welds, there are many and usually there is a
method statement generated and approved by the Client
to spell out the process for completing such welds.
I have in the past been involved with the witnessing

of such welds and it is usually up to the Welding

Engineer to decide on what length to go with these
joints, in the past I have had to follow a whole
variety of stringent controls to ensure that the welds
were carried out in the best possible way.
Some of the measures are as follows:
1). A qualified Inspector being present throughout the
welding operation to confirm the visual condition of
the weld at all stages of completion.
2). A visual inspection of the root of the weld, this
was achieved by leaving the final length of root bead
open, (100mm), for a visual inspection with a pen
light, to ensure defects such as lack of root fusion
were not present.
3). Controlled sequential welding was applied to
lessen the unnecessary stress that could be put into
the weld by not applying such method.
4). The Hot Pass of the weld was sometimes MPI'd or
DPI'd, dependant on the pre-heat requirements or the
material being welded.(never the root as it is too
sensitive an area and it is best to be covered over
with the hot pass bead as quickly as possible for
added strength).
5). Heat inputs were sometimes taken at certain
intervals during welding by the Inspector, as would be
during a Weld Procedure Qualification test, using
'Tong testers'/stopwatch/tape measure/calculator, to
ensure that the welder was working within the
peramaters stipulated in the relevant welding
procedure.
6). Joint interpass temperatures would be taken with a
digital thermometer at the end of each pass, to ensure
that the interpass tempareture was in line with the
welding procedure tolerances.
7). Finally, when the 'Golden Weld' was completed
there would be a whole range of NDE to complete such
as Visual inspection, MPI or DPI, Radiographic AND
Ultrasonic Inspection, both prior to and after Post
Weld Heat Treatment, (if PWHT was applicable).
Still being in the business after 35 years, I look in
dispair today when it is notedthat such critical
joints are not treated with the same respect today.
It is in my book better to be safe than sorry by going
the extra yard with the controls, in such critical
instances.
=======================================================
Speaking from the Inspector standpoint, I might only add one thing.

The processes for treating final closure welds are well spelled out
in some construction codes which I would consult for strict
compliance purposes. For instance, ASME B31.3 has extensive
directions which I have seen often change the decision on whether or
not to hydrotest when it is at all possible. As for the golden
term...I learn something new everyday!
=======================================================
In a new piping fabrication/installation/replacement jobs, generally all
the weld joints are hydroteted (along with other NDT methods) in phases.
However, sometimes due to some layout/site problems, it is almost impossible
to do hydrotest of the final weld joint. We refer that weld joint as golden
joint.
Generally, Dye penetrant test of root weld, radiographt etc.. is used for
quality control of that joint
Say, something like a hook-up to an existing line etc..
=======================================================
Golden weld means, InA tie-in work whatever joints were welded at the
fabrication shop,Balance weld must be only one joint at site
welding.This weldind joint cal be called as golden joint.This joints
should be 100% confidence with the all NDT.Welder could be very very
experienced man.
=======================================================
This must be the modern day replacement for the fabled golden rivet from
shipbuilding days. When inspecting a golden weld be sure not to bend over.

unclesyd (Materials)

20 Apr 07
23:57

Gator,
You forgot the Golden Spike in RR construction.
We have never had any problem with getting the AI to OK NDT tests in lieu of a hydrotest upon
completion of the closure weld.
Gator (Industrial)
See http://en.wikipedia.org/wiki/Golden_rivet, it's referenced there.

21 Apr 07 0:05