Period 1 - Pipeline Corrosion Control

Download Report

Transcript Period 1 - Pipeline Corrosion Control

Period 1
I am trying to bring our science into the understanding
that everyone has in their every day lives. I make no
apologies to the scientists amongst us because they
will learn where they have failed to communicate their
knowledge to the end users.
Orac 2
• This arrangement of dry cell batteries is in
fact similar to the arrangement of
corrosion cells that confronts us every day
on pipeline networks. We use a system of
measurement called ‘pipe-to-soil potential
measurement’ that is a voltage obtained
using a voltmeter or oscilloscope. The
measuring circuit is completed by touching
the ground with a copper/copper-sulphate
electrode.
Orac 2
Orac 2
•
The idea is that we know the potential of the reaction between copper and
copper-sulphate and can compare it with the reaction of the pipeline metal
to the ground at that location. This is comparing two reactions and is not
measuring a ‘half-cell reaction’ as they do in a laboratory.
•
We can represent this part of the measuring circuit by putting a new battery
(of known value) in a battery holder with a depleted battery (in series) and
measuring the total voltage. We then subtract the known value and
discover the value of the depleted battery.
•
In closed circuit measuring conditions we can use this measurement to
determine when corrosion has stopped and it is that reasoning that is
behind the use of the ‘pipe-to-soil potential’ as a standard criterion for the
achievement of ‘protection’ in our industry.
•
You will now see that this is wrong thinking that has never been supported
by science but is used as the basis of all cathodic protection design,
monitoring and maintenance at this time.
3 nails demonstration
•
We can see the displayed voltages as we apply this reasoning to the
demonstration involving the corrosion of three nails.
•
It is quickly apparent that the measuring circuit has more than two potentials
in series because the readings change when the electrode is moved. The
negative of the meter is permanently attached to the negative pole of the
battery and the other is permanently attached to the electrode that we are
told to regard as a reference potential. The only change in the measuring
circuit is the location of the electrode point of contact so it must be that each
location itself has a different potential. We are measuring the potential of
the ground itself, and this is an unknown variable. The voltage on the
display is therefore between the pipeline metal/electrolyte potential and the
copper/copper-sulphate reaction potential AND whatever other influences
affect the potential of the ground on which the electrode is placed. This
means that the data cannot be analysed mathematically or on a computer
without further data.
3 nails corrosion
This is identical to the measuring
process we use in field work
• It was recognised in the 1970’s that the criterion was in error and
pipeline operators investigated the reason why leaks were occurring
in places that had been regarded as ‘protected’
• Field engineers simply believed what their clients specified and the
pipeline owners contracted others to apply the science that had
satisfied the theories in laboratory experiments.
• The term ‘half-cell reaction’ was wrongly taken to infer that you could
make a ‘half-cell’ and use it as a reference against which you can
measure another potential. In fact this can only be done in
controlled circumstances in a ‘closed circuit measurement’ condition
but the demonstration (involving the three nails) proves conclusively
that you cannot do this in field work.
Present field practices.
• Pipelines have continued to fail due to corrosion that
continues because of measurement errors and the
pipeline owners engage consultants who confirmed that
the copper/copper-sulphate electrode is acceptable as a
reference.
• The corrosion control industry has confirmed that the
measurement should be taken using a standard
reference electrode and nobody could understand what
was wrong.
• Field engineers teach cathodic protection technicians,
who in turn form their own service companies, who
continued using similar techniques.
A different scientific approach was
required.
•
Specialised equipment is used, world-wide, and is very efficient, but there are many
disastrous failures, which could be prevented by proper measurement of the
corrosion voltages. This has been recognised by NACE and ISO and many people
in the industry, but they have not presented a solution to the problem.
•
I recognised the error within 20 minutes of being shown how the measurement is
made but it took me some months to find our how to accommodate this
understanding within the work that I was being paid to carry out.
•
It took me a couple of years to realise that a solution to this problem had to be
devised or corrosion will never be controlled by applied cathodic protection.
•
In medical science it is common practice to experiment with cultures of the bacteria
and samples of the virus that is causing the trouble and it was apparent that we
needed a corrosion cell that we can measure and control so that we can see the
success of our work.
The original design for the
Alexander Cell (corrosion cell)
The general purpose Alexander
(corrosion) Cell
The criterion for cathodic
protection.
•
The ‘criterion’ for achieving cathodic protection is widely accepted to be a voltage
measurement that is supposed to indicate an equilibrium between the metal and the
electrolyte at which no metal is going into solution. No corrosion is taking place
because the electrochemical reaction has been halted.
•
This criterion has been under continuous review for at least 50 years, that I know of,
and there is still no advice that is backed by demonstrable evidence, published by
NACE, ISO or any of the corrosion control institutions of which I am aware.
•
In the 1980’s the Alexander Cell was suggested to the National Physical Laboratory in
the UK but they were told of an arrangement of coupons that is NOT the one that I
invented.
•
I presented the Alexander Cell to the Institute of Corrosion Science and Technology in
London in the mid 1980’s and this was published in two international journals
specialising in corrosion control and pipeline management.
The measurement of actual
corrosion in field work.
• When I personally submitted the Alexander cell for examination by
The National Physical Laboratory in the UK they dismissed the idea
as "unnecessary, due to the satisfactory performance of existing
practices".
• They made a report that they sent to specialists in corrosion control.
The Chairman of the British Standards Institute committee for
Cathodic Protection, Jim Gosden, officially demanded that the report
be withdrawn and highlighted no fewer than 17 gross scientific missstatements.
• This did not prevent the report being sent to Libya where it misled
Sirte Oil company into putting a major contract on hold.
• There were many other well documented instances of disinformation
to suppress the use of the Alexander Cell and thus prevent proper
scientific measurement of the corrosion reaction in the field.
• However, Alexander Cell has now been extensively field trialled and
in 2009 I made presentations at two international conferences in
Brazil, announcing it as a definitive criterion for cathodic protection.