Addressing the challenges to sustainable operation of gas turbines

01:03:16
https://www.youtube.com/watch?v=Ijqa6ClA9gw

Summary

TLDRIn this webinar, John Douglas and Laurie Brooking explore the challenges of maintaining sustainable operational performance in gas turbines. They emphasize the use of physics-based models in asset management to optimize turbine longevity and performance. Laurie discusses the behaviour of high temperature materials, including single crystal super alloys, plagued by degradation and stress factors like creep and fatigue. The presentation includes insights into Fraser Nash's growth and capabilities in engineering and consultancy, focusing on integrating digital tools with robust engineering principles. The evidence indicates that adopting these physics-based strategies can better inform operational decisions, potentially leading to more efficient and flexible management of gas turbine assets, suggesting extensions in operational life and improved reliability.

Takeaways

  • 🔧 Use of physics-based models and digital assets to manage gas turbine operations.
  • 🌍 Fraser Nash's international growth, with operations in the UK and Australia.
  • 📈 Importance of physics models in extending gas turbine operational life.
  • ⚙️ Corrosion fatigue in high temperature materials like single crystal super alloys.
  • 🔍 Emphasis on research collaboration for developing engineering techniques.
  • 💡 Opportunities presented by flexible operation and maintenance deferral.
  • 🔬 Detailed technical methods to assess and optimize turbine parts.
  • 📊 Insights into understanding corrosion fatigue interactions.
  • 📚 Strong academic ties enhancing research and development.
  • 🚀 Fraser Nash's diversified engineering services across industries like aerospace and industrial applications.

Timeline

  • 00:00:00 - 00:05:00

    John Douglas welcomes participants to a webinar about sustainable operational gas turbines, explaining the broad subject will focus on physics-based models for asset management. Laurie Brooking introduces himself with a relevant background in high-temperature materials and research on gas turbine blade materials.

  • 00:05:00 - 00:10:00

    John shares his professional journey starting with a PhD in gas turbine materials and his current role at Fraser Nash overseeing accounts related to turbines. The presentation will highlight Fraser Nash's capabilities, specifically the Rotating Machinery Center and the use of physics-based models for digital asset management.

  • 00:10:00 - 00:15:00

    Fraser Nash has grown significantly, now with offices in the UK and Australia, employing around 900 staff. They focus on broad capabilities and collaboration with academic institutions to solve complex problems for clients often returning due to their comprehensive approach to challenging issues.

  • 00:15:00 - 00:20:00

    The Rotating Machinery Center at Fraser Nash specializes in design, digital assets, and engineering services. They cover a spectrum from gas turbines to other rotating machinery. Their approach emphasizes developing robust, physics-based digital asset management systems rather than just data analytics.

  • 00:20:00 - 00:25:00

    Using physics-based methods, Fraser Nash believes they get better results in managing complex machinery. Their hybrid approach integrates data with physics to extend machine operating life, boost performance, and optimize part placement. They often help clients defer maintenance based on comprehensive assessments.

  • 00:25:00 - 00:30:00

    The discussion delves into the technical methodologies, where simple models are initially used and refined based on emerging data. This is framed within a probabilistic approach to manage uncertainties from sensor data and model accuracy, ultimately supporting risk-based maintenance decisions.

  • 00:30:00 - 00:35:00

    Highlighting examples of engineering services, they have developed efficient ventilation standards for gas turbine enclosures, vital for health and safety, especially considering new standards like ATEX. Investigations into hydrogen's effects are ongoing.

  • 00:35:00 - 00:40:00

    Asset management plays a crucial role. The approach allows understanding of machine's damage status in real time, guided by physics-based assessments. This is valuable for operational decisions, data interpretation, and risk assessments under varying load conditions.

  • 00:40:00 - 00:45:00

    Focus shifts to Laurie Brooking's analysis of corrosion-fatigue interactions in single crystal alloys, showing propagation patterns in high-temperature environments. This segment underscores the complexity of predicting material behavior and the importance of understanding these interactions for reliability.

  • 00:45:00 - 00:50:00

    Tests reveal how corrosion fatigue emerges through intricate mechanisms. Laurie observes how different elements and deposits influence crack initiation and propagation, indicating these materials' response varies significantly with conditions impacting a turbine's component lifespan.

  • 00:50:00 - 00:55:00

    Laurie presents how physics models like Paris curves are adapted to monitor crack growth in fatigue tests. Different rates of corrosion-fatigue interactions are mapped out, showing variance in material responses and the benefits of incorporating this understanding into turbine operation.

  • 00:55:00 - 01:03:16

    Q&A addresses standstill corrosion and uncertainty in sensor data, emphasizing physics models' adaptiveness and reliability compared to pure data analytics. The session closes with John acknowledging Rizvan's contributions and promising further discussions on probabilistics and risk management.

Show more

Mind Map

Video Q&A

  • Who are the main speakers of the webinar?

    John Douglas and Laurie Brooking.

  • What company do the speakers represent?

    Fraser Nash.

  • What is the main focus of the webinar?

    Addressing the challenges to sustainable operational gas turbines using physics-based models.

  • What is Laurie Brooking's area of expertise?

    High temperature materials and gas turbine blade materials, with a focus on degradation and super alloys.

  • What company is mentioned as having a growing international presence?

    Fraser Nash, with offices in the UK and Australia.

  • What are some applications of the discussed physics-based models?

    Applications include asset management and optimizing the operational integrity of gas turbines.

  • How does physics-based modeling benefit gas turbines?

    It helps manage damage accumulation, optimize operations, and extend the life of turbine components.

  • Can the methods discussed help with part-load operation of gas turbines?

    Yes, they can optimize and extend part-load operation conditions.

View more video summaries

Get instant access to free YouTube video summaries powered by AI!
Subtitles
en
Auto Scroll:
  • 00:00:08
    ah
  • 00:00:09
    good morning good afternoon everybody um
  • 00:00:12
    right on cue my builder has just got his
  • 00:00:13
    tools out and started creating noise i
  • 00:00:15
    hope you can't hear that too badly
  • 00:00:17
    um it's john douglas i'm just um i'm
  • 00:00:19
    just going to give it another minute or
  • 00:00:21
    so
  • 00:00:21
    to to let some more people
  • 00:00:24
    uh sign into this and then we'll get
  • 00:00:28
    started
  • 00:01:42
    okay so good afternoon everybody um i
  • 00:01:45
    think most of the people who are going
  • 00:01:46
    to join are probably joined already now
  • 00:01:48
    so
  • 00:01:49
    i'm going to get started um thank you to
  • 00:01:52
    everyone for
  • 00:01:53
    for taking an hour out of their time to
  • 00:01:55
    to listen to this webinar
  • 00:01:57
    on addressing the challenges to
  • 00:01:59
    sustainable operational gas turbines
  • 00:02:01
    my name is john douglas i've got an
  • 00:02:03
    introduction slide coming up
  • 00:02:05
    i'm going to talk through obviously this
  • 00:02:06
    is a very big subject area um
  • 00:02:08
    i'll talk briefly about some of the work
  • 00:02:11
    that we at phrase and ash have been
  • 00:02:12
    doing
  • 00:02:13
    in this very big subject area and i
  • 00:02:16
    obviously can't cover all of what we're
  • 00:02:17
    doing and all of what happens in this
  • 00:02:19
    area but hopefully
  • 00:02:20
    hopefully you're going to find what's
  • 00:02:21
    what's presented today of interest
  • 00:02:24
    um so
  • 00:02:28
    next slide um hopefully laurie you can
  • 00:02:30
    see that
  • 00:02:31
    i don't know if you want to give us a
  • 00:02:32
    brief introduction to yourself before we
  • 00:02:34
    get
  • 00:02:34
    started yeah thanks john yeah so
  • 00:02:37
    i'm laurie brooking and um yeah my
  • 00:02:40
    interest in high temperature materials
  • 00:02:42
    sort of started with my phd at
  • 00:02:44
    cranfield university where i was looking
  • 00:02:46
    primarily at
  • 00:02:47
    degradation and gas turbine materials
  • 00:02:50
    gas turbine blade materials so single
  • 00:02:51
    crystal
  • 00:02:52
    uh super alloys and super alloys um and
  • 00:02:55
    and after kind of being in in the
  • 00:02:57
    research environment there i then moved
  • 00:02:59
    over to
  • 00:03:00
    consultancy and and started at fraser
  • 00:03:02
    nation
  • 00:03:03
    sort of since since then have been
  • 00:03:05
    continuing to work in in largely high
  • 00:03:07
    temperature materials and after
  • 00:03:08
    integrity so
  • 00:03:09
    so looking at how how we can help our
  • 00:03:11
    clients to
  • 00:03:12
    improve the integrity of their assets
  • 00:03:14
    and longevity and
  • 00:03:16
    get the best cost benefits so yeah
  • 00:03:18
    that's me
  • 00:03:20
    thanks very much laurie um so i guess my
  • 00:03:23
    my interest started very similarly to
  • 00:03:25
    lorries with with a phd
  • 00:03:27
    in uh in gaster turbine materials as
  • 00:03:29
    well i'm a mechanical engineer but i was
  • 00:03:31
    looking at
  • 00:03:31
    single crystal alloys from a creep
  • 00:03:34
    fatigue perspective rather than from a
  • 00:03:36
    degradation
  • 00:03:36
    oxidation type point of view and over
  • 00:03:39
    the years i've been with fraser nash for
  • 00:03:42
    over 17 years now in various roles more
  • 00:03:45
    recently
  • 00:03:47
    on the commercial side looking after the
  • 00:03:50
    the account for
  • 00:03:51
    um turbines and i've recently set up the
  • 00:03:55
    rotating machinery center which which
  • 00:03:56
    i'll talk a little bit more about
  • 00:03:59
    as we go through the slides um
  • 00:04:02
    in terms of the presentation today
  • 00:04:05
    hopefully the content slide has come up
  • 00:04:07
    um there's opportunity obviously to
  • 00:04:09
    advertise fraser nash so a quick slide
  • 00:04:11
    on
  • 00:04:12
    praising ash without making it too much
  • 00:04:14
    of an advertisement but really just a
  • 00:04:15
    bit
  • 00:04:16
    brief of a brief on who we are and
  • 00:04:19
    um what we look like today moving on
  • 00:04:22
    from that
  • 00:04:23
    my little corner of fraser nash which is
  • 00:04:24
    the rotating machinery center
  • 00:04:26
    and some of the work that we're doing
  • 00:04:28
    within within
  • 00:04:29
    within within within that capability
  • 00:04:31
    area um
  • 00:04:34
    and really the presentation today um is
  • 00:04:37
    is is about the use of physics-based
  • 00:04:39
    models for
  • 00:04:40
    asset management for one of a better
  • 00:04:42
    word digital is one of those words
  • 00:04:44
    that's used quite a lot at the moment
  • 00:04:46
    what we're doing does fit within the
  • 00:04:48
    description of digital
  • 00:04:49
    uh one way or another so we we call it
  • 00:04:52
    digital assets
  • 00:04:54
    there's a lot of talk out there about
  • 00:04:55
    data and analytics um
  • 00:04:57
    and there is a place for that but with
  • 00:04:59
    complex machinery like this
  • 00:05:01
    we find that the use of physics-based
  • 00:05:03
    methods or mechanical engineering
  • 00:05:05
    uh gives us a better result in some in
  • 00:05:08
    some instances
  • 00:05:09
    and so there's various parts of that
  • 00:05:11
    physics uh i'll get into
  • 00:05:12
    some of the detail on that but really um
  • 00:05:16
    once i've given an overview of that
  • 00:05:17
    approach it's over to laurie to
  • 00:05:19
    to to a presentation on on what is what
  • 00:05:22
    is what he's looking at in terms of
  • 00:05:24
    corrosion fatigue interactions
  • 00:05:26
    you can see with crystal alloys um a
  • 00:05:28
    very particular corner of those physics
  • 00:05:30
    models
  • 00:05:32
    the next slide oops i think i've gone
  • 00:05:36
    too
  • 00:05:37
    so a little bit about frasier nash i've
  • 00:05:40
    been here for 17 years we were less than
  • 00:05:42
    100 people when i joined
  • 00:05:43
    we've had quite a growth trajectory uh
  • 00:05:45
    we've moved out of the uk and we're now
  • 00:05:47
    in australia as well so
  • 00:05:49
    we have a significant number of offices
  • 00:05:50
    in the uk we've got four officers in
  • 00:05:52
    australia
  • 00:05:53
    um i hope i'm not misrepresenting the
  • 00:05:55
    australia officers
  • 00:05:57
    um in terms of they've been there around
  • 00:06:00
    about 10 years
  • 00:06:01
    and all right there's over 100 people in
  • 00:06:03
    those offices now or if it's getting
  • 00:06:05
    closer to 200
  • 00:06:06
    people but it's that sort of order of
  • 00:06:07
    magnitude in total we've got 900
  • 00:06:10
    ish permanent staff in our offices in
  • 00:06:13
    the uk
  • 00:06:14
    and in australia we have a number of
  • 00:06:16
    other people positioned around the globe
  • 00:06:17
    in austria
  • 00:06:18
    in in america and europe on the site
  • 00:06:20
    with clients doing some interesting work
  • 00:06:22
    as well
  • 00:06:23
    um if we haven't got an expert within
  • 00:06:26
    the company that can help
  • 00:06:27
    then we have access to a network of over
  • 00:06:30
    three thousand other
  • 00:06:31
    host associates who we can call on for
  • 00:06:33
    expertise
  • 00:06:34
    um so we have quite a broad network
  • 00:06:37
    very very broad capability across
  • 00:06:40
    three to four thousand people and if we
  • 00:06:43
    still can't find someone that can help
  • 00:06:45
    with a problem
  • 00:06:46
    then we have very strong collaborations
  • 00:06:48
    with with the universities
  • 00:06:50
    the vast majority well a large
  • 00:06:52
    proportion of the people at frozen ash
  • 00:06:54
    have done some important post-doctoral
  • 00:06:56
    study or
  • 00:06:56
    sorry postgraduate study either masters
  • 00:06:59
    phds
  • 00:07:01
    postdoctoral and so very strong links to
  • 00:07:03
    academia
  • 00:07:04
    and and if and if we can't solve it if
  • 00:07:07
    our associates can't solve it then we
  • 00:07:09
    can call on
  • 00:07:10
    those collaborations to to get experts
  • 00:07:12
    in to help with whatever problems
  • 00:07:14
    um we're also getting something right a
  • 00:07:16
    lot of our customers keep coming back
  • 00:07:18
    um for for more and more help um
  • 00:07:21
    and that's great there's various reasons
  • 00:07:23
    for that but i think i think the biggest
  • 00:07:25
    reason for that
  • 00:07:25
    is the fact that um we don't just we
  • 00:07:27
    don't just put people on site
  • 00:07:29
    and sell people for man hours if you
  • 00:07:31
    like we
  • 00:07:32
    we like to own our customers problems
  • 00:07:35
    and we like to provide solutions to
  • 00:07:37
    those problems
  • 00:07:37
    and i think it's that it's that approach
  • 00:07:40
    which gives us an
  • 00:07:41
    edge in the marketplace so
  • 00:07:44
    that's that's i guess that's the
  • 00:07:46
    advertisement over for fraser nash
  • 00:07:47
    um if you want to find out more you can
  • 00:07:49
    call me or you can get in touch by the
  • 00:07:51
    website
  • 00:07:52
    we might be able to help you moving on
  • 00:07:55
    um my little corner of frasier nash is
  • 00:07:58
    is is the rotating machinery center
  • 00:08:00
    uh recently set up just before christmas
  • 00:08:03
    um and we've launched a
  • 00:08:04
    a part of the website which is fnc.co uk
  • 00:08:08
    forward slash rmse and
  • 00:08:11
    broadly speak about that that gives you
  • 00:08:13
    a brief overview of
  • 00:08:14
    some of the skill sets that we've got in
  • 00:08:16
    in the rotating machinery
  • 00:08:18
    field uh the vast majority of it is
  • 00:08:20
    around gas turbines
  • 00:08:21
    steam turbines pumps that type of stuff
  • 00:08:23
    but actually we do a lot of work on on
  • 00:08:25
    other rotating machinery
  • 00:08:27
    including energy storage devices and uh
  • 00:08:30
    rockets and various other gadgets so
  • 00:08:33
    we've got we've got a vast
  • 00:08:34
    experience uh we even do stuff on you
  • 00:08:37
    know
  • 00:08:37
    fans for kitchen appliances and that
  • 00:08:39
    type of stuff so so
  • 00:08:42
    we've got a broad experience there and
  • 00:08:43
    if you go to the website you'll find out
  • 00:08:46
    uh broadly speaking we've got three
  • 00:08:47
    service areas
  • 00:08:49
    machinery design a lot of our a lot of
  • 00:08:51
    our engineers
  • 00:08:52
    have had careers at original equipment
  • 00:08:56
    manufacturing companies and brought a
  • 00:08:59
    lot of that design experience with them
  • 00:09:01
    others have spent time on site with us
  • 00:09:03
    with those customers and
  • 00:09:05
    and gained a lot of experience in terms
  • 00:09:06
    of design and so we
  • 00:09:08
    you know whether it's a component design
  • 00:09:10
    for a blade or a disc or a
  • 00:09:12
    combustor or whether it's a full system
  • 00:09:15
    or a stage or a turbine we can pick up
  • 00:09:18
    that kind of work
  • 00:09:19
    and depending on the type of the type of
  • 00:09:21
    machine that we're looking at we can
  • 00:09:22
    actually design a full machine
  • 00:09:24
    and give you a set of drawings and we
  • 00:09:25
    can work with you to get that
  • 00:09:26
    manufactured so
  • 00:09:27
    we have a significant significant
  • 00:09:29
    machinery design
  • 00:09:30
    capability there and and some of that is
  • 00:09:33
    is outlined on the website um
  • 00:09:36
    the the next part of our service
  • 00:09:38
    capability there is digital assets and
  • 00:09:40
    that's what we're going to be talking
  • 00:09:41
    about today so i won't go into too much
  • 00:09:42
    detail here
  • 00:09:43
    um and then engineering services if you
  • 00:09:46
    know we've got a broad range of services
  • 00:09:48
    we've got a lot of engineers doing a lot
  • 00:09:49
    of different things
  • 00:09:50
    in a lot of different teams um if it's
  • 00:09:52
    not covered by
  • 00:09:53
    machinery design or digital assets then
  • 00:09:55
    there are a list of services there
  • 00:09:57
    and there's a brief list here which
  • 00:09:59
    covers thermodynamic performance
  • 00:10:01
    road dynamics materials that type stuff
  • 00:10:05
    and before we get into the talk on
  • 00:10:06
    digital assets then
  • 00:10:08
    and then into laurie's presentation just
  • 00:10:10
    a quick example
  • 00:10:11
    of a a service area
  • 00:10:14
    under the engineering services so
  • 00:10:17
    hopefully the sliders refreshed
  • 00:10:19
    there's a picture there of a gas cloud
  • 00:10:21
    within a within a
  • 00:10:22
    gas turbine enclosure our team of fluid
  • 00:10:25
    engineers have been working with
  • 00:10:27
    customers been working with health and
  • 00:10:30
    safety laboratories to develop
  • 00:10:32
    standards around atex and dasia
  • 00:10:35
    in terms of ventilation of enclosures
  • 00:10:38
    and
  • 00:10:39
    a lot of that work over the last 20
  • 00:10:40
    years has been focused primarily on
  • 00:10:42
    methane gas clouds
  • 00:10:44
    where do they form how do they form are
  • 00:10:46
    they ventilated and if they're not
  • 00:10:48
    ventilated where do we need to put
  • 00:10:49
    sensors
  • 00:10:50
    in that type of work like so the team
  • 00:10:52
    have been working on this for a very
  • 00:10:53
    long time
  • 00:10:54
    they've developed some very efficient
  • 00:10:55
    fluid dynamics methods which which cover
  • 00:10:58
    which cover the basics of how to how to
  • 00:11:00
    get that done
  • 00:11:02
    and how to get that done really quickly
  • 00:11:03
    really efficiently so if you have that
  • 00:11:05
    kind of problem you really should be
  • 00:11:06
    talking to our team
  • 00:11:08
    uh they're central to the development of
  • 00:11:09
    these standards and understand how to
  • 00:11:11
    solve these problems
  • 00:11:12
    and understand how to get you certified
  • 00:11:14
    in those in those sorts of problems
  • 00:11:17
    more recently we've been that team have
  • 00:11:19
    been working on
  • 00:11:20
    um how those standards can be modified
  • 00:11:23
    how they're affected by the introduction
  • 00:11:25
    of hydrogen
  • 00:11:26
    and whether that's hydrogen at one
  • 00:11:28
    percent or one hundred percent and
  • 00:11:30
    anything in between
  • 00:11:31
    uh those simulations are underway and we
  • 00:11:33
    are working with
  • 00:11:34
    we are working with uh various partners
  • 00:11:36
    to develop that standard and to actually
  • 00:11:38
    look at
  • 00:11:39
    uh installations within europe um
  • 00:11:42
    european facilities at the moment so
  • 00:11:46
    it's a little bit it's a little bit left
  • 00:11:48
    field but that's that's just one of the
  • 00:11:50
    services that we can offer
  • 00:11:51
    um it's worth having a look at the
  • 00:11:52
    website to see which other services
  • 00:11:55
    we offer it's still under development so
  • 00:11:57
    not everything that we offer is on there
  • 00:11:59
    and if you want to find out more again
  • 00:12:00
    please call me
  • 00:12:01
    or get in touch by the website
  • 00:12:05
    getting into what are we doing on the
  • 00:12:08
    on on the asset side um i guess a lot of
  • 00:12:11
    people on the call would be aware
  • 00:12:13
    that the the you know the oems um
  • 00:12:16
    power stations um owners operators
  • 00:12:20
    um have got a lot of telemetry on on gas
  • 00:12:23
    turbines tomorrow that's a
  • 00:12:24
    small unit pumping gas along an oil and
  • 00:12:27
    gas pipeline or whether that's a
  • 00:12:29
    large power station there's for a long
  • 00:12:31
    time a lot of data has been coming off
  • 00:12:33
    there for
  • 00:12:34
    the purpose of vibration monitoring
  • 00:12:35
    performance monitoring
  • 00:12:38
    we've figured out over the last five or
  • 00:12:40
    ten years how to make best use of that
  • 00:12:42
    data
  • 00:12:43
    in terms of understanding what's
  • 00:12:44
    happening with the machine and
  • 00:12:47
    even just at a simple level am i running
  • 00:12:50
    a full load steady state or am i
  • 00:12:52
    running at part load 50 um
  • 00:12:55
    we've got some real-time algorithms
  • 00:12:57
    which look at the data
  • 00:12:58
    understand what level the machine's
  • 00:13:01
    operating at
  • 00:13:02
    and then we've got lots of physics
  • 00:13:04
    models we've got lots of algorithms
  • 00:13:06
    which which will get into the detail of
  • 00:13:08
    what's happening with metal temperature
  • 00:13:11
    and metal stress within the components
  • 00:13:13
    that are of interest
  • 00:13:14
    um so if you're running at full load
  • 00:13:16
    steady state you're expecting various
  • 00:13:18
    failure modes within blades or discs or
  • 00:13:20
    nozzles
  • 00:13:21
    uh we're monitoring the stresses we're
  • 00:13:23
    moderating monitoring
  • 00:13:25
    the temperatures in those components
  • 00:13:27
    real time
  • 00:13:28
    and we're able to understand from that
  • 00:13:30
    how much damage is accumulating
  • 00:13:32
    within within within those components
  • 00:13:35
    and
  • 00:13:35
    you know you can integrate that
  • 00:13:36
    throughout the life of that that turbine
  • 00:13:41
    we do that with various physics based
  • 00:13:42
    approaches including creep
  • 00:13:44
    fatigue corrosion and oxidation my
  • 00:13:47
    interests are around greek fatigue
  • 00:13:49
    lori is very interested in the corrosion
  • 00:13:51
    and oxidation end of things
  • 00:13:52
    but we bring all of those models
  • 00:13:53
    together to understand the condition of
  • 00:13:56
    your machine
  • 00:13:57
    at any given point in time and you know
  • 00:13:59
    there's various advantages or various
  • 00:14:01
    opportunities
  • 00:14:02
    coming out of that um not least flexible
  • 00:14:04
    operation so
  • 00:14:06
    can i can i run my machine at 50 power
  • 00:14:09
    for
  • 00:14:09
    for longer than the usual maintenance
  • 00:14:11
    window the answer is yes you can
  • 00:14:12
    generally speaking
  • 00:14:14
    um part load power is generally um
  • 00:14:18
    less damaging than full well steady
  • 00:14:20
    state or than the design case
  • 00:14:21
    um but not always and that's where the
  • 00:14:23
    physics comes in so data analytics on
  • 00:14:26
    its own
  • 00:14:27
    um can't tell you when you're going to
  • 00:14:29
    bake the back end of the gas turbine for
  • 00:14:30
    instance
  • 00:14:31
    um and so so running a part load
  • 00:14:33
    generally it's less damaging but there
  • 00:14:35
    are instances where
  • 00:14:36
    you will cause more damage by running at
  • 00:14:38
    reduced load and it's the physics models
  • 00:14:40
    that really get into the nitty-gritty of
  • 00:14:41
    that
  • 00:14:41
    that tell you to tell you why and and
  • 00:14:44
    when
  • 00:14:45
    those things are happening um so
  • 00:14:47
    flexible operation
  • 00:14:48
    can i extend the life well the physics
  • 00:14:50
    will tell me whether i can or not
  • 00:14:53
    what if what if i want to run a power
  • 00:14:55
    boost what if i want to do 120
  • 00:14:57
    for a couple of hours a month what does
  • 00:14:59
    that do to my maintenance window what
  • 00:15:00
    does that do to the damage within the
  • 00:15:02
    engine
  • 00:15:02
    and that's that's where that's where the
  • 00:15:04
    value of these sorts of
  • 00:15:05
    algorithms come in um can i optimize
  • 00:15:09
    part placement to recover pop
  • 00:15:11
    performance of
  • 00:15:12
    the gas turbine yeah you can so we can
  • 00:15:14
    monitor the engine we can use the
  • 00:15:16
    physics
  • 00:15:16
    to understand when you might want to do
  • 00:15:19
    a wash on the on the on the on the
  • 00:15:20
    compressor
  • 00:15:21
    or when you might need to replace some
  • 00:15:23
    nozzles which have which will start to
  • 00:15:25
    build up uh product on them
  • 00:15:27
    uh or or or or durability issues um
  • 00:15:30
    you know so when to replace parts when's
  • 00:15:34
    a good time to replace parts with
  • 00:15:36
    that sort of stuff falls out of physics
  • 00:15:37
    and then um
  • 00:15:39
    maintenance deferring um i know my
  • 00:15:41
    engine
  • 00:15:42
    is coming up for maintenance within the
  • 00:15:44
    next three months but actually
  • 00:15:46
    i want to run it for six months what was
  • 00:15:48
    my previous what was my previous
  • 00:15:50
    use case done to to the the condition of
  • 00:15:53
    those parts and can i can i defer my
  • 00:15:55
    maintenance while i get through this
  • 00:15:56
    busy period for the next six months
  • 00:15:58
    and all sorts of questions and that's
  • 00:16:00
    and that's the kind of stuff that we're
  • 00:16:01
    working with power stations
  • 00:16:02
    working with oems to to solve and using
  • 00:16:05
    these
  • 00:16:06
    methods to to actually do so a little
  • 00:16:09
    bit
  • 00:16:10
    a little bit more detail on on without
  • 00:16:13
    getting into the physics a little bit
  • 00:16:15
    more detail on the technical approach
  • 00:16:17
    um we start simple and and we build from
  • 00:16:20
    there so
  • 00:16:21
    we've got some very very simple models
  • 00:16:24
    that
  • 00:16:24
    that that tell you about what's going on
  • 00:16:27
    in the physics of crete fatigue
  • 00:16:29
    corrosion oxidation there's no point in
  • 00:16:32
    us developing
  • 00:16:33
    highly complex models for your engine uh
  • 00:16:37
    on all on all corners without the best
  • 00:16:39
    the best approaches to is to start
  • 00:16:41
    with simple models and then start
  • 00:16:42
    looking at the results that are coming
  • 00:16:44
    off
  • 00:16:44
    and we'll find out very quickly whether
  • 00:16:46
    your engine engines
  • 00:16:48
    are a creep dominated fatigue dominated
  • 00:16:51
    or if they've got oxidation and
  • 00:16:52
    corrosion problems
  • 00:16:54
    um and from that we can then start to
  • 00:16:57
    build in
  • 00:16:58
    more and more detail on on the physics
  • 00:17:00
    and uh and
  • 00:17:02
    and uh and and take it from there um
  • 00:17:04
    there's no point in building a highly
  • 00:17:06
    complex fatigue model if you've got a
  • 00:17:08
    predominate machine
  • 00:17:09
    um that's the first point i guess the
  • 00:17:12
    next point is that
  • 00:17:13
    this is all about risk um we
  • 00:17:18
    we obviously need to understand what's
  • 00:17:19
    going on with the creep and the fatigue
  • 00:17:20
    for various reasons but actually
  • 00:17:24
    the sensors the data that's coming up
  • 00:17:26
    with sensors isn't 100 accurate
  • 00:17:28
    what's going on with thermocouple drift
  • 00:17:30
    is that thermocouple accurate
  • 00:17:32
    is it even working those sorts of
  • 00:17:34
    questions come into your risk based
  • 00:17:35
    model
  • 00:17:36
    what about my life and model my creep
  • 00:17:38
    model if i'm starting with a basic creep
  • 00:17:40
    model
  • 00:17:40
    then then there's obviously a lot better
  • 00:17:42
    certainty on the accuracy of that model
  • 00:17:45
    versus versus a mob spell model which is
  • 00:17:47
    being developed
  • 00:17:48
    very very neatly for your application
  • 00:17:50
    and so
  • 00:17:51
    understanding the uncertainty
  • 00:17:53
    understanding
  • 00:17:54
    the risk that comes with that
  • 00:17:56
    uncertainty is is a big part of what
  • 00:17:58
    we're doing
  • 00:17:59
    and all of what we're doing here sits
  • 00:18:00
    within a probabilistic wrapper
  • 00:18:04
    those methods have been developed from
  • 00:18:05
    monte carlo simulations of graphite
  • 00:18:07
    cores on nuclear reactors so
  • 00:18:09
    we're pretty confident that those
  • 00:18:11
    methods work and we're pretty confident
  • 00:18:13
    with them with the answers that come out
  • 00:18:14
    of that
  • 00:18:16
    um and i guess the the last point on
  • 00:18:18
    this slide really is about the the fact
  • 00:18:20
    that we're not
  • 00:18:21
    you know we're not hungry for more and
  • 00:18:23
    more data um
  • 00:18:24
    i mean it's obviously always quite nice
  • 00:18:26
    to have more data but
  • 00:18:28
    the more data you have the bigger the
  • 00:18:29
    problem you've got processing it um
  • 00:18:31
    there's a lot of machinery out there
  • 00:18:32
    with with with
  • 00:18:34
    with sensors what we try and do is make
  • 00:18:38
    best use of that sensor data
  • 00:18:40
    without actually needing to go in and
  • 00:18:42
    refit new sensors
  • 00:18:43
    and and new telemetry and just try and
  • 00:18:46
    make use of what we've got
  • 00:18:47
    occasionally we're coming to a situation
  • 00:18:49
    where we just don't have enough data
  • 00:18:51
    and it is sensible to to put more
  • 00:18:53
    sensors where we need them
  • 00:18:55
    but most of the time we we get away with
  • 00:18:59
    with just the sense of data that exists
  • 00:19:03
    so it's not a big expensive job to refit
  • 00:19:06
    your fleet
  • 00:19:09
    and then just turning some of that into
  • 00:19:11
    it into a picture quickly
  • 00:19:12
    um on the left-hand side we've got the
  • 00:19:15
    data platform
  • 00:19:17
    on the right-hand side we've got a
  • 00:19:18
    risk-based decision making tool
  • 00:19:21
    uh and in between we've got our
  • 00:19:22
    physics-based tool set
  • 00:19:24
    um just just to you know on the first
  • 00:19:27
    level we've got some very simple models
  • 00:19:29
    um we will implement them very quickly
  • 00:19:32
    very cheaply to start producing results
  • 00:19:35
    and
  • 00:19:35
    and once once we start to get results
  • 00:19:37
    from that kind of approach that's when
  • 00:19:39
    we can start to think about do we need a
  • 00:19:41
    creep fatigue interaction model do we
  • 00:19:43
    need a more complex corrosion
  • 00:19:45
    or oxidation model uh and from that
  • 00:19:48
    do we then need to get into the various
  • 00:19:50
    specialist models that
  • 00:19:51
    that can be developed on a needs must
  • 00:19:55
    basis
  • 00:19:57
    hopefully hopefully that's given you
  • 00:19:59
    quite a nice overview of our
  • 00:20:01
    approach to managing your assets using
  • 00:20:04
    using sort of digital
  • 00:20:08
    techniques coupled with with
  • 00:20:10
    physics-based approaches
  • 00:20:14
    i can't talk in detail about all of that
  • 00:20:16
    i'm not going to talk in detail
  • 00:20:18
    about all of that i think at this stage
  • 00:20:20
    really that's a little bit
  • 00:20:21
    that's that's enough of an understanding
  • 00:20:23
    to understand our approach
  • 00:20:25
    um i think i'm just going to hand over
  • 00:20:27
    to lori now so that he can talk a little
  • 00:20:29
    bit about his corrosion fatigue
  • 00:20:30
    interactions
  • 00:20:31
    and what he's seeing on on single
  • 00:20:33
    crystal alloys
  • 00:20:38
    great so um hello everyone um yeah my
  • 00:20:41
    name is
  • 00:20:41
    laurie brooking and uh i'll be taking a
  • 00:20:44
    quick look at corrosion fatigue
  • 00:20:46
    interactions in
  • 00:20:47
    single crystal super alloys
  • 00:20:51
    so this is a common issue and it's seen
  • 00:20:53
    both in industrial and aerospace
  • 00:20:56
    gas turbines and it's
  • 00:20:59
    a cracking morphology often similar to
  • 00:21:02
    fatigue
  • 00:21:03
    and that's observed typically in hot gas
  • 00:21:05
    pump components
  • 00:21:06
    you know earlier and sooner than
  • 00:21:08
    predicted so we'll be talking about what
  • 00:21:10
    causes this issue
  • 00:21:12
    how do these cracks propagate and how we
  • 00:21:14
    can look to predict
  • 00:21:16
    and mitigate against against this
  • 00:21:18
    problem so on
  • 00:21:20
    um just just just at the right here of
  • 00:21:23
    the bullet points you can see there's a
  • 00:21:24
    couple of images of uh micrographs
  • 00:21:27
    uh cmsx4 single crystal super if you're
  • 00:21:29
    not familiar so i was just going to kind
  • 00:21:31
    of explain what is a single crystal c
  • 00:21:32
    probably
  • 00:21:33
    why they're different to to a lot of
  • 00:21:35
    other allies so you can see we've got
  • 00:21:37
    that
  • 00:21:38
    cuboid cuboidal type micro structures
  • 00:21:40
    camera gamma prime microstructures for
  • 00:21:42
    31
  • 00:21:43
    and the gamma prime is is the kind of
  • 00:21:45
    the cuboidal feature the square feature
  • 00:21:47
    there
  • 00:21:49
    and the matrix is the bit in between
  • 00:21:52
    those cuboidal features
  • 00:21:53
    so they're both face centered cubic um
  • 00:21:56
    and
  • 00:21:56
    the gamma prime is an l one two ordered
  • 00:21:59
    face
  • 00:21:59
    centered cubic and just next to that the
  • 00:22:01
    really high mag
  • 00:22:02
    five nanometer image is showing you the
  • 00:22:05
    interface between that camera and the
  • 00:22:06
    camera prime so really this is
  • 00:22:08
    this is what makes these materials a bit
  • 00:22:09
    special so whilst you've got two phases
  • 00:22:12
    because the unit cell size is so
  • 00:22:14
    comparable uh we actually
  • 00:22:16
    don't really see a phase boundary you
  • 00:22:18
    can see that the lattice coherence is
  • 00:22:20
    really very good
  • 00:22:21
    and in in these materials certainly at
  • 00:22:24
    this temperature obviously things change
  • 00:22:25
    around a bit with different temperatures
  • 00:22:26
    but
  • 00:22:26
    we don't really have a grain boundary
  • 00:22:28
    hence why we call it a single crystal
  • 00:22:30
    super light and that comes with numerous
  • 00:22:32
    advantages for high temperature
  • 00:22:34
    material properties creep fatigue things
  • 00:22:37
    like that
  • 00:22:38
    along the bottom i've got some images of
  • 00:22:40
    what typical corrosion fatigue damage
  • 00:22:42
    looks like so what we're talking about
  • 00:22:43
    in this presentation the type of damages
  • 00:22:45
    we we
  • 00:22:46
    would see so the image on the bottom
  • 00:22:48
    left there
  • 00:22:49
    is of an industrial gas turbine and we
  • 00:22:51
    can see
  • 00:22:52
    a corrosion damage on the platform where
  • 00:22:54
    the arrows are pointing to and
  • 00:22:56
    actually bits of material has broken off
  • 00:22:58
    there the middle image
  • 00:23:00
    is undecration platform and again this
  • 00:23:02
    is quite typical of what we would see
  • 00:23:04
    um so it's a highly stressed area of the
  • 00:23:06
    blade but actually lower temperatures
  • 00:23:08
    than a lot of other other parts of the
  • 00:23:09
    blade or some other parts of the blade
  • 00:23:11
    but we see colonies of cracks
  • 00:23:13
    propagating through the platform outside
  • 00:23:15
    through
  • 00:23:16
    through the root region that highly
  • 00:23:18
    stressed region there
  • 00:23:19
    and then the image on the bottom right
  • 00:23:21
    is actually an image of
  • 00:23:22
    what can happen if if this goes
  • 00:23:24
    undetected you know obviously this
  • 00:23:26
    happens
  • 00:23:27
    a lot sooner other sort of said than
  • 00:23:29
    expected in many cases
  • 00:23:30
    and if it's not noticed the image on the
  • 00:23:33
    right there is is some blades which have
  • 00:23:34
    been ejected at the back of
  • 00:23:36
    an aircraft engine actually upon takeoff
  • 00:23:38
    so certainly not
  • 00:23:39
    something that you want to be happening
  • 00:23:41
    and
  • 00:23:46
    a bit of background then into why this
  • 00:23:49
    is happening
  • 00:23:50
    and we believe the cracking occurs as a
  • 00:23:52
    result of simultaneous deposit induced
  • 00:23:54
    hot corrosion i'll talk a bit more about
  • 00:23:56
    what that is uh in a bit combined with
  • 00:23:59
    loading
  • 00:23:59
    or stress and and the cracking can occur
  • 00:24:03
    as the static loads as well as as as
  • 00:24:06
    fatigue cycling so it's it's kind of a
  • 00:24:08
    bit of a stress corrosion cracking type
  • 00:24:10
    mechanism similar to what we would see
  • 00:24:12
    you know in aqueous uh stress corrosion
  • 00:24:15
    cracking at
  • 00:24:15
    obviously much lower temperatures
  • 00:24:17
    typically and i think a very
  • 00:24:19
    very good question is why is this
  • 00:24:20
    mechanism become prevalent in
  • 00:24:22
    recent gas turbine designs and i think
  • 00:24:25
    there's a couple of
  • 00:24:26
    a couple of reasons for that the first
  • 00:24:28
    is increases
  • 00:24:29
    in turbine efficiencies you're driving
  • 00:24:32
    up
  • 00:24:33
    operating temperatures due to your car
  • 00:24:35
    not cycle
  • 00:24:36
    and we see the extended effect of hot
  • 00:24:38
    corrosion so
  • 00:24:39
    really we're pushing the boundaries and
  • 00:24:41
    trying to improve efficiencies of these
  • 00:24:43
    turbines or running some different fuels
  • 00:24:46
    hydrogen etcetera which might need us to
  • 00:24:48
    look at running at higher temperatures
  • 00:24:49
    and
  • 00:24:50
    and that can really extend the effects
  • 00:24:52
    of hot corrosion into areas where we
  • 00:24:54
    wouldn't typically see it
  • 00:24:55
    and the second use is sort of related so
  • 00:24:58
    because again we want to achieve higher
  • 00:25:01
    temperatures to improve efficiencies
  • 00:25:03
    we're looking at using materials um
  • 00:25:05
    which have higher gamma prime factors
  • 00:25:07
    factors fractions sorry due to
  • 00:25:10
    um their you know sorry they've got high
  • 00:25:13
    gamma prime fractions in order to
  • 00:25:15
    achieve uh better high temperature
  • 00:25:17
    properties so better creek properties
  • 00:25:19
    and things like that
  • 00:25:19
    but by increasing the gamma prime
  • 00:25:21
    fraction we actually
  • 00:25:23
    reduce uh the the refractory refractory
  • 00:25:26
    element
  • 00:25:26
    um concentration uh weight percentage
  • 00:25:29
    and those refractory elements are really
  • 00:25:31
    very good for building up protective
  • 00:25:33
    oxides so
  • 00:25:34
    we typically see with a lot of the
  • 00:25:36
    really good high temperature super
  • 00:25:37
    alloys
  • 00:25:38
    they're a bit worse particularly at
  • 00:25:39
    lower temperature corrosion and so
  • 00:25:42
    there's a bit of a trade-off there and
  • 00:25:43
    that's another reason i think we're
  • 00:25:44
    seeing this mechanism become more
  • 00:25:47
    more of an issue as we use those real
  • 00:25:49
    high temperature
  • 00:25:50
    high gamma prime fraction superb
  • 00:25:56
    so i was just going to introduce high
  • 00:25:59
    temperature deposit
  • 00:26:00
    induced corrosion or hot corrosion
  • 00:26:02
    because it's not necessarily that
  • 00:26:04
    that well known about so it's typically
  • 00:26:06
    defined by
  • 00:26:07
    two mechanisms low temperature or type 2
  • 00:26:10
    hot corrosion and high temperature or
  • 00:26:12
    type 1 upgrade and you can see there
  • 00:26:14
    on the figure on the right we've got a
  • 00:26:16
    fairly linear relationship between
  • 00:26:18
    temperature
  • 00:26:19
    and oxidation rate however you have
  • 00:26:21
    these two humps
  • 00:26:22
    that accelerated damage um for type two
  • 00:26:25
    and type one with type 2 occurring at
  • 00:26:27
    slightly low
  • 00:26:28
    low temperature and the reason we get
  • 00:26:30
    these two hunts is
  • 00:26:32
    largely due to chemistry so hot
  • 00:26:34
    corrosion occurs as a result of
  • 00:26:35
    corrosive species
  • 00:26:37
    forming a liquid melt or eutectic on the
  • 00:26:39
    surface of the blade
  • 00:26:41
    and it typically um is is forming
  • 00:26:44
    due to gaseous conditions and and also
  • 00:26:47
    the deposition of
  • 00:26:49
    of those species so depending on your
  • 00:26:52
    your alloy
  • 00:26:53
    composition your coating composition
  • 00:26:55
    your deposit composition
  • 00:26:57
    you can see the temperature of these
  • 00:26:58
    these mechanisms move around
  • 00:27:00
    and actually historically that that
  • 00:27:02
    craft there on the right is
  • 00:27:04
    is often thought of as as a kind of
  • 00:27:06
    baseline but i think in reality
  • 00:27:08
    the chemistry is quite a lot more
  • 00:27:09
    complex and we see a bit more of a
  • 00:27:11
    probably complex picture in in reality
  • 00:27:15
    um so it's typically thought that um
  • 00:27:18
    operation is electrochemical so that
  • 00:27:21
    liquid deposit or liquid mixture forming
  • 00:27:24
    on the blade is is a
  • 00:27:25
    is a an ion transport and um
  • 00:27:28
    is enabling uh acidic dissolution
  • 00:27:32
    to happen kind of you know localized
  • 00:27:34
    locations and cause
  • 00:27:35
    cause this is cracking but just a
  • 00:27:38
    general note
  • 00:27:38
    um on on alloys and and how they tend to
  • 00:27:41
    respond to this mechanism
  • 00:27:43
    generally speaking chromia forming alloy
  • 00:27:45
    solids which form a
  • 00:27:46
    chromium oxide they're more resistant to
  • 00:27:48
    type 2 and lower temperature oxidation
  • 00:27:50
    and corrosion
  • 00:27:51
    and your alumina forming alloys are more
  • 00:27:53
    protective for type 1 and higher
  • 00:27:54
    temperature
  • 00:27:55
    oxidation and corrosion
  • 00:27:59
    i also i think it's it's um
  • 00:28:02
    sometimes debated a little bit as as to
  • 00:28:05
    whether on or not hot corrosion
  • 00:28:07
    reactions are deposit regulated
  • 00:28:09
    or whether they're regulated by gaseous
  • 00:28:11
    composition or
  • 00:28:12
    partial pressure of sox or so3 and i
  • 00:28:15
    suppose in my view
  • 00:28:16
    again i think it's uh there's several
  • 00:28:18
    computing mechanisms and
  • 00:28:19
    it's it's quite a complex picture
  • 00:28:21
    however i suppose
  • 00:28:22
    that notes in industrial gas turbines um
  • 00:28:25
    you often have
  • 00:28:26
    uh filtration um which is limiting the
  • 00:28:29
    amount of
  • 00:28:30
    solid salt or deposit species or species
  • 00:28:34
    which can be
  • 00:28:35
    just through the engine of the positive
  • 00:28:36
    so i think and a lot of the time we may
  • 00:28:38
    see more of a
  • 00:28:40
    kind of a factor of your gaseous in
  • 00:28:43
    terms of
  • 00:28:44
    limiting uh your hot corrosion um
  • 00:28:47
    however in aerospace engines we don't we
  • 00:28:48
    don't have that same filtration
  • 00:28:50
    um so i think a lot of the time
  • 00:28:53
    composition
  • 00:28:53
    and deposit composition can play a
  • 00:28:55
    bigger part
  • 00:28:57
    in an aerospace application so again i
  • 00:28:59
    think it can vary a little bit and
  • 00:29:01
    there's there's quite a lot of complex
  • 00:29:03
    chemistry going on
  • 00:29:04
    as well
  • 00:29:08
    so if we take a look then at some
  • 00:29:09
    experimental test results here
  • 00:29:11
    um these are a couple of images of
  • 00:29:14
    statically loaded specimens so there's
  • 00:29:15
    no cycling here these are cracks which
  • 00:29:17
    have been generated
  • 00:29:18
    uh you know in line with kind of stress
  • 00:29:21
    corroding cracking rather
  • 00:29:22
    rather than fatigue and what we see is
  • 00:29:25
    multiple initiation sites
  • 00:29:27
    and so the top image there is all the
  • 00:29:30
    plain cylindrical
  • 00:29:31
    specimen again statically loaded at
  • 00:29:32
    quite high stretch there 911 pascals
  • 00:29:35
    but we see uh multiple crack initiate
  • 00:29:38
    cracks initiate and propagate um
  • 00:29:41
    and uh obviously it's it's very much
  • 00:29:45
    dependent on on the corrosion so if you
  • 00:29:47
    take away that corrosion you obviously
  • 00:29:48
    don't expect these materials to
  • 00:29:50
    to initiate propagate cracks at these
  • 00:29:52
    temperatures it's it's it really is
  • 00:29:54
    corrosion
  • 00:29:55
    driven and the the layer image is
  • 00:29:57
    showing something similar
  • 00:29:58
    it's an optical image rather than a um
  • 00:30:01
    electron
  • 00:30:02
    microscope image and and that's showing
  • 00:30:04
    a similar thing on the c ring at the
  • 00:30:05
    slightly lower stress there 500
  • 00:30:07
    megapascals
  • 00:30:08
    but we can see again multiple crack
  • 00:30:10
    initiation sites um
  • 00:30:12
    and uh environmental kind of marking as
  • 00:30:15
    well going on so showing the
  • 00:30:17
    significance of the environment to this
  • 00:30:20
    propagation
  • 00:30:22
    if we have a look at a few
  • 00:30:23
    cross-sectional images um
  • 00:30:25
    now then um we can start to see how this
  • 00:30:27
    mechanism is interacting a bit more with
  • 00:30:29
    the microstructure so
  • 00:30:31
    image a there up on the the top left is
  • 00:30:34
    jane what looks to be
  • 00:30:35
    a bit like a corrosion pit and we've got
  • 00:30:37
    a crack which is
  • 00:30:38
    initiated and propagated out at the
  • 00:30:40
    bottom of that pit we obviously can't
  • 00:30:41
    tell what came first a bit on the crack
  • 00:30:43
    um but we we also see the corrosion
  • 00:30:47
    interacting preferentially with the
  • 00:30:49
    gamma prime and this is kind of backed
  • 00:30:51
    up by the the following edges so we can
  • 00:30:52
    see
  • 00:30:53
    really at the crack tip there in image b
  • 00:30:55
    again it's interacting with the gamma
  • 00:30:57
    prime
  • 00:30:58
    propagating on orthogonal q flames
  • 00:31:00
    through the material
  • 00:31:01
    and even some bridging through the the
  • 00:31:04
    gamut and
  • 00:31:05
    so the gamma prime is is typically the
  • 00:31:07
    strengthening
  • 00:31:08
    phase or precipitate so
  • 00:31:11
    preferentially oxidizing that is
  • 00:31:13
    obviously locally weakening
  • 00:31:15
    the material in a table like this crack
  • 00:31:17
    propagation
  • 00:31:21
    if we have a look then what we've got
  • 00:31:23
    here are pd
  • 00:31:24
    um plots so these represent the crack
  • 00:31:27
    growth or propagation
  • 00:31:28
    rate throughout the test so the way we
  • 00:31:31
    do this
  • 00:31:32
    and again it's quite a common technique
  • 00:31:35
    that's used in fatigue
  • 00:31:36
    um less so in stress corrosion cracking
  • 00:31:39
    and again less so
  • 00:31:40
    high temperature with deposits to
  • 00:31:42
    creation um
  • 00:31:43
    but it's quite a useful method for us to
  • 00:31:45
    actually look at a bit more detail at
  • 00:31:47
    what's going on so what we do is we pass
  • 00:31:49
    a large current through the specimen and
  • 00:31:50
    we have a couple of threads
  • 00:31:52
    either side of where we expect cracking
  • 00:31:54
    to occur and as that crack propagates
  • 00:31:56
    and we lose
  • 00:31:57
    specimen cross-sectional area we see an
  • 00:31:59
    increase in potential drop or resistance
  • 00:32:01
    which we can measure
  • 00:32:02
    and that corresponds to to our correct
  • 00:32:04
    size so
  • 00:32:06
    there's two tests here they have uh
  • 00:32:09
    comparable test conditions and
  • 00:32:11
    other than they they're using different
  • 00:32:13
    deposits so
  • 00:32:14
    so a different composition of the salts
  • 00:32:17
    so the one on the left
  • 00:32:18
    is uh sodium potassium sulfate based
  • 00:32:20
    salt we can see
  • 00:32:21
    an incubation period so the crack hasn't
  • 00:32:24
    started straight away
  • 00:32:25
    but then it initiates and it's kind of
  • 00:32:28
    continuing on then
  • 00:32:29
    at a very linear rate um not really
  • 00:32:32
    showing much
  • 00:32:33
    correlation to to the crack size as we
  • 00:32:35
    would expect to see with fatigue
  • 00:32:37
    fairly linear rates trembling along if
  • 00:32:40
    we have a look at the
  • 00:32:41
    chloride salt we see something similar
  • 00:32:42
    except we've not got that incubation
  • 00:32:43
    time we do have an overall
  • 00:32:45
    linear growth rate there's a little bit
  • 00:32:47
    of jumping we can
  • 00:32:48
    see going on there but i think the key
  • 00:32:50
    thing to note here
  • 00:32:51
    is that with these two salts we see a
  • 00:32:53
    very different rate of propagation
  • 00:32:55
    so obviously these aren't calibrated
  • 00:32:57
    we've just got potential difference here
  • 00:32:59
    but it does
  • 00:33:00
    it does correspond to the the crack the
  • 00:33:02
    crack size so we can see that the rate
  • 00:33:04
    of increase of
  • 00:33:04
    potential drop or the rate which central
  • 00:33:06
    drops increasing is twice
  • 00:33:08
    as high for sea salts so that chloride
  • 00:33:11
    based so you know these temperatures on
  • 00:33:14
    on these alloys
  • 00:33:14
    so it's kind of highlighting the
  • 00:33:16
    significance of of what the deposit is
  • 00:33:18
    actually made of
  • 00:33:19
    and or what the composition of the
  • 00:33:21
    deposit is
  • 00:33:22
    you know for this mechanism
  • 00:33:25
    so starting to introduce fatigue into
  • 00:33:28
    the equation then so
  • 00:33:29
    now we've got stress cycling going on
  • 00:33:31
    this is quite a simplistic view of
  • 00:33:33
    of how corrosion or this type of
  • 00:33:36
    corrosion affects
  • 00:33:37
    the life of these alloys and again it
  • 00:33:39
    doesn't tell the whole story but it does
  • 00:33:41
    just start to help us unpick a little
  • 00:33:42
    bit what's going on
  • 00:33:44
    so we've got three um curves on this on
  • 00:33:47
    this stretch cycles to fairly aggressive
  • 00:33:49
    an sn curve
  • 00:33:50
    at the first two or the square in the
  • 00:33:53
    triangle
  • 00:33:53
    are five microgram um fluxes so that's
  • 00:33:57
    the rate of deposition
  • 00:33:58
    of our corrosive species and the circle
  • 00:34:02
    is 1.2 micrograms centimeter
  • 00:34:06
    squared per hour rate of of flux
  • 00:34:09
    deposition
  • 00:34:10
    so if we compare the two five microgram
  • 00:34:12
    results to start with
  • 00:34:13
    one is a one second dwell so this is a
  • 00:34:15
    trapezoidal test
  • 00:34:16
    so we're ramping up the stress and we're
  • 00:34:19
    holding it that dwell time represents
  • 00:34:22
    the holding the stress at maximum value
  • 00:34:25
    and then we're ramping it back down and
  • 00:34:28
    so we've got a one second dwell in the
  • 00:34:30
    60 seconds so the 60 seconds we'll test
  • 00:34:32
    the triangle uh
  • 00:34:34
    contains or has a lot more corrosion
  • 00:34:36
    occurring per cycle
  • 00:34:38
    than the one second twelve so we can see
  • 00:34:40
    more corrosion
  • 00:34:41
    in that case worse for t life fairly
  • 00:34:44
    simple
  • 00:34:44
    fairly simple story and if we compare
  • 00:34:47
    the two one-second dwell tests
  • 00:34:48
    so the square in the circle again the
  • 00:34:50
    square
  • 00:34:51
    has a higher flux and more corrosion
  • 00:34:54
    than the circle
  • 00:34:55
    and we can see again that drop in
  • 00:34:56
    fatigue life corresponding to that
  • 00:34:59
    increased amount of corrosion so that's
  • 00:35:02
    all makes all makes sense but again this
  • 00:35:04
    is quite a simplistic view so we'll
  • 00:35:06
    we'll start to dig into a bit more what
  • 00:35:08
    what's going on because i don't think sn
  • 00:35:10
    curves are a great way to
  • 00:35:11
    to look at look at what's going on
  • 00:35:15
    so what we've got here then is uh is
  • 00:35:17
    again using that potential drop
  • 00:35:18
    technique but this time on
  • 00:35:20
    fatigue specimens uh 550
  • 00:35:23
    degrees c and these are not fatigue
  • 00:35:24
    specimens um
  • 00:35:26
    we've got some plots of the crack
  • 00:35:29
    propagation
  • 00:35:30
    um so showing how that crack depth is
  • 00:35:32
    varying over
  • 00:35:33
    both cycle and uh and time and
  • 00:35:37
    and uh this sort of starts to give us a
  • 00:35:39
    bit more information on what's going on
  • 00:35:42
    so what we can see is uh if we if we
  • 00:35:45
    start by looking at the cracked
  • 00:35:47
    cycles so that put on the left on the
  • 00:35:49
    left uh we've got
  • 00:35:50
    three three tests notched one knots two
  • 00:35:52
    and not just three so
  • 00:35:54
    knots two and notch three uh are a
  • 00:35:56
    longer dwell again
  • 00:35:57
    than notched one so knots two and notch
  • 00:35:59
    three have more corrosion going on per
  • 00:36:00
    cycle and again we can see that
  • 00:36:03
    reflected in
  • 00:36:04
    in the fact that they take fewer number
  • 00:36:06
    of cycles to fail
  • 00:36:07
    uh than notch one so more corrosion
  • 00:36:09
    fewer number of cycles
  • 00:36:10
    fail again that same story coming
  • 00:36:12
    through we can also see though
  • 00:36:14
    that we get these crack jumps occurring
  • 00:36:17
    uh
  • 00:36:18
    as we resolve the specimen so the
  • 00:36:19
    crosses there on on those curves
  • 00:36:22
    represent the point at which we've
  • 00:36:23
    resulted the
  • 00:36:25
    the test so as our corrosive species or
  • 00:36:28
    deposit
  • 00:36:29
    is is used up it evaporates or you know
  • 00:36:31
    it's it's becoming used in the corrosion
  • 00:36:33
    process
  • 00:36:34
    and we have to kind of replenish that so
  • 00:36:37
    we can see how the crack jumps and
  • 00:36:39
    interacts with with those resulting
  • 00:36:41
    processes if we have a look at how how
  • 00:36:44
    this works out on a time basis rather
  • 00:36:46
    than a cycle basis
  • 00:36:48
    um you can see the the story is a little
  • 00:36:50
    less clear
  • 00:36:51
    so notched 3 and notch two again have
  • 00:36:54
    some fairly big jumps
  • 00:36:55
    early on we've got more stress corrosion
  • 00:36:57
    cracking going on
  • 00:36:58
    um but they've not increased it
  • 00:37:01
    there's a higher rate because there's
  • 00:37:03
    less fatigue elements going on this is a
  • 00:37:05
    much one specimen so
  • 00:37:07
    there's no clear correlation in terms of
  • 00:37:09
    time to failure
  • 00:37:10
    in in this case because i think the
  • 00:37:12
    mechanism is a bit more complicated than
  • 00:37:14
    the sf curve sort of depicts
  • 00:37:20
    if we move on to have a quick look at
  • 00:37:22
    the fracture faces
  • 00:37:23
    of those kind of specimens um
  • 00:37:26
    we see lots of environmental markings
  • 00:37:29
    and things going on
  • 00:37:30
    as well which also represents um
  • 00:37:34
    those kind of start stop type behavior
  • 00:37:36
    of the crank propagation
  • 00:37:38
    mechanism i think again is it's fairly
  • 00:37:40
    complicated we've got multiple cracks
  • 00:37:42
    interacting with each other things
  • 00:37:43
    um but it's it's sort of useful to note
  • 00:37:46
    that
  • 00:37:46
    that start stop behavior is it's sort of
  • 00:37:50
    represented on the fracture phase as
  • 00:37:52
    well so
  • 00:37:54
    i think quite interestingly um if we
  • 00:37:56
    start to to look at
  • 00:37:57
    corrosion and fatigue interactions on on
  • 00:38:00
    a paris curve
  • 00:38:01
    we can we can understand a bit more
  • 00:38:03
    what's going on so i've included here
  • 00:38:05
    on the left a paris curve and if if
  • 00:38:08
    you're not familiar with it
  • 00:38:10
    i'll just quickly run over what what
  • 00:38:12
    paris curve is
  • 00:38:13
    um so we've got stress intensity factor
  • 00:38:16
    on the x-axis
  • 00:38:18
    so that's measured in megapascal's root
  • 00:38:20
    meter
  • 00:38:21
    so it's a kind of stress to geometrical
  • 00:38:25
    um parameter and it's it corresponds to
  • 00:38:29
    crackdown
  • 00:38:30
    so as our crack depth increases our
  • 00:38:32
    stress intensity factor
  • 00:38:34
    uh typically increases as well um on
  • 00:38:37
    the y-axis we've got the crack growth
  • 00:38:40
    rate so d
  • 00:38:40
    n d a sorry over d n uh where d a is
  • 00:38:44
    changing crack length
  • 00:38:45
    over the end is changing in cycles
  • 00:38:48
    and what we see for typical fatigue
  • 00:38:52
    if we look at the red line is we see
  • 00:38:54
    initiation so
  • 00:38:56
    rapid increase um
  • 00:39:00
    corresponding between stress potential
  • 00:39:02
    or or rapid increase between
  • 00:39:03
    uh relationships between stress
  • 00:39:05
    intensity factor and crack growth rate
  • 00:39:07
    and then that levels off so we get is
  • 00:39:09
    this flat
  • 00:39:10
    um line on the log log plot which
  • 00:39:13
    represents a paris curve and
  • 00:39:15
    we've got a constant and an exponent to
  • 00:39:18
    to help us
  • 00:39:19
    to model that uh so that's our kind of
  • 00:39:21
    typical fatigue area and we'll use those
  • 00:39:24
    constants
  • 00:39:24
    uh quite frequently to model fatigue
  • 00:39:28
    and then again just towards or just
  • 00:39:30
    towards the end we'll see
  • 00:39:31
    uh crack growth deviate from from that
  • 00:39:34
    paris line
  • 00:39:35
    um again so really that bit in the
  • 00:39:37
    middle that cm
  • 00:39:39
    straight line is our paris growth and
  • 00:39:42
    that's when we know
  • 00:39:43
    that uh our crack rate is is fairly in
  • 00:39:45
    line with a typical fatigue dominated
  • 00:39:47
    crack group so if we have a look there
  • 00:39:49
    our three results plotted on
  • 00:39:52
    paris curve so the plot on on the right
  • 00:39:54
    here
  • 00:39:55
    what we see if we start looking at
  • 00:39:57
    notched one so
  • 00:39:58
    to the left of what i put in there is a
  • 00:40:01
    threshold line
  • 00:40:02
    we see there's very little uh or no
  • 00:40:05
    relationship really
  • 00:40:06
    between stress intensity factor and the
  • 00:40:08
    adm um
  • 00:40:10
    early on in the correct growth of
  • 00:40:13
    our crack propagation is is driven
  • 00:40:15
    really more by
  • 00:40:16
    time dependent or corrosion uh driven
  • 00:40:18
    factors rather
  • 00:40:20
    rather than stress intensity factor
  • 00:40:22
    however there becomes a point at the
  • 00:40:24
    crack depth
  • 00:40:24
    at which the crack then starts to grow
  • 00:40:26
    more in line with fatigue
  • 00:40:28
    so when we look to the right of that
  • 00:40:30
    threshold line for the notch one
  • 00:40:31
    specimen we see
  • 00:40:32
    we actually start to get that straight
  • 00:40:34
    line so that paris curve
  • 00:40:36
    um is starting to come through there so
  • 00:40:38
    we can we can sort of hypothesize that
  • 00:40:40
    um at that point fatigue has become the
  • 00:40:43
    dominant factor so
  • 00:40:44
    to the left we've got stress corrosion
  • 00:40:46
    cracking or time dependent
  • 00:40:48
    uh um mechanisms dominating our crack
  • 00:40:51
    growth and
  • 00:40:52
    to the right we've got more fatigue
  • 00:40:53
    coming into play and
  • 00:40:55
    i guess what's quite interesting is to
  • 00:40:56
    compare um
  • 00:40:58
    the knox one to the notch three uh sorry
  • 00:41:01
    knots two and notch three
  • 00:41:02
    specimens so not too much three or a
  • 00:41:04
    longer dwell so there's more corrosion
  • 00:41:05
    going
  • 00:41:06
    on but we actually see the crack depth
  • 00:41:08
    at which they
  • 00:41:09
    start to behave more like fatigue or
  • 00:41:13
    they start to
  • 00:41:14
    correspond to fatigue crack craters is
  • 00:41:16
    higher so that's telling us that
  • 00:41:18
    corrosion in this case is actually
  • 00:41:20
    suppressing
  • 00:41:21
    fatigue and again we can hypothesize um
  • 00:41:24
    that that's due for
  • 00:41:25
    for could be due to a couple of reasons
  • 00:41:27
    uh so there's
  • 00:41:28
    correct closure uh which can occur so
  • 00:41:31
    that that's where your corrosion
  • 00:41:33
    product actually wedges open your crack
  • 00:41:36
    effectively reducing
  • 00:41:37
    um your delta k so your stress intensity
  • 00:41:40
    factor
  • 00:41:40
    factor range because there's less um
  • 00:41:42
    crack open
  • 00:41:43
    crack opening closure occurring um or
  • 00:41:47
    there's also
  • 00:41:47
    crack crack uh tick blunting effects
  • 00:41:49
    which which can also happen
  • 00:41:51
    um but i guess this relationship of
  • 00:41:53
    corrosion suppressing fatigue great
  • 00:41:55
    is perhaps slightly uh counter-intuitive
  • 00:41:58
    and
  • 00:41:59
    i guess it's important to to sort of
  • 00:42:01
    higher in fact it's suppressing fatigue
  • 00:42:03
    grades which is
  • 00:42:04
    certainly not suppressing the corrosion
  • 00:42:06
    uh or stress corrosion cracking good
  • 00:42:08
    driven uh propagation gain going on
  • 00:42:10
    earlier
  • 00:42:11
    um i suppose as well interesting to note
  • 00:42:14
    is um
  • 00:42:15
    this is uh this is also quite similar to
  • 00:42:18
    and the behavior of single crystal
  • 00:42:20
    surprise uh between
  • 00:42:21
    air and vacuum um so we will actually
  • 00:42:24
    see
  • 00:42:25
    um that crack fatigue propagation in
  • 00:42:28
    single crystallized
  • 00:42:29
    in air um is slightly
  • 00:42:32
    worse or there's less fatigue
  • 00:42:36
    propagation in air than there is in
  • 00:42:38
    vacuum at high temperatures
  • 00:42:40
    so i i guess there's some comparisons
  • 00:42:42
    that can be made there and
  • 00:42:44
    i suppose we see those effects more
  • 00:42:45
    maybe in single crystal always because
  • 00:42:46
    we don't have
  • 00:42:47
    grain boundary oxidation going on uh
  • 00:42:50
    forms of propagation occurring
  • 00:42:56
    so i was going to lastly talk a little
  • 00:42:58
    bit then about
  • 00:43:00
    why why does all of this happen and what
  • 00:43:02
    is that what is the mechanism behind it
  • 00:43:04
    so
  • 00:43:05
    i was going to start off here i've got a
  • 00:43:06
    few um sem images
  • 00:43:09
    um of a statically legislation again so
  • 00:43:11
    no fatigue
  • 00:43:12
    in these last few slides that i'm going
  • 00:43:15
    to present you
  • 00:43:16
    um so what we can see again in
  • 00:43:20
    echo is what i was talking about earlier
  • 00:43:21
    we can see the crack propagating
  • 00:43:24
    on q planes so 10801a
  • 00:43:28
    and we can see corrosion interacting or
  • 00:43:31
    prevalent preferentially attacking the
  • 00:43:32
    gamma prime
  • 00:43:33
    um and image c there's showing
  • 00:43:37
    a what looks to be an early corrosion
  • 00:43:39
    pit with again
  • 00:43:41
    corrosion interacting preferentially
  • 00:43:42
    with the gamma prime
  • 00:43:44
    image d um is actually showing slightly
  • 00:43:47
    the opposite so that's a corrosion pip
  • 00:43:48
    but it was taken on a compression side
  • 00:43:50
    and but we can actually see that in that
  • 00:43:52
    case the corrosion is interacting more
  • 00:43:54
    with the gamma matrix um potentially
  • 00:43:58
    suggesting
  • 00:43:58
    uh you know there's some stress factors
  • 00:44:02
    in this preferential attack um also just
  • 00:44:04
    highlighting
  • 00:44:05
    uh the complexity of this this corrosion
  • 00:44:08
    mechanism as well
  • 00:44:09
    um and certainly that echoes a lot of
  • 00:44:10
    what i've seen and
  • 00:44:12
    you can you can you can convince
  • 00:44:15
    yourself of one thing
  • 00:44:16
    and then you know several weeks later
  • 00:44:17
    you see something that's just completely
  • 00:44:19
    the opposite
  • 00:44:19
    really throws a spanner in the works
  • 00:44:23
    and so you're looking down at some
  • 00:44:24
    higher mag images so these are
  • 00:44:26
    transmission electron microscope
  • 00:44:28
    images of the crack tip
  • 00:44:32
    what we can see is the crack cutting
  • 00:44:34
    through and propagating through the
  • 00:44:35
    camera prime
  • 00:44:36
    quite interestingly because because
  • 00:44:37
    again just to highlight those those
  • 00:44:39
    gamma prime
  • 00:44:40
    uh phases are strengthening so they're
  • 00:44:43
    quite a lot stronger than the camera
  • 00:44:45
    and we can also see dislocations so
  • 00:44:47
    stacking faults
  • 00:44:48
    on the octahedral plane so one-on-one
  • 00:44:50
    stacking faults
  • 00:44:51
    ahead of the crack chip and again that's
  • 00:44:54
    not unexpected at these temperatures we
  • 00:44:55
    would expect to see slip occurring
  • 00:44:57
    on those planes and but
  • 00:45:00
    interestingly the crack doesn't really
  • 00:45:02
    appear to interact with it so i think if
  • 00:45:04
    this mechanism was fatigued driven or or
  • 00:45:07
    plasticity was particularly important we
  • 00:45:09
    perhaps see the crack
  • 00:45:10
    interacting and propagating a little bit
  • 00:45:12
    more on those planes
  • 00:45:13
    than we do and but we don't it seems to
  • 00:45:16
    just carry straight on ahead
  • 00:45:18
    on on the cube plates
  • 00:45:23
    so some edx then uh just to look at um
  • 00:45:26
    what's going on and again i've just
  • 00:45:29
    taken a little snapshot i've looked at
  • 00:45:31
    quite a few of these but what we can see
  • 00:45:34
    is um
  • 00:45:35
    firstly there's no convincing evidence
  • 00:45:37
    of diffusion or absorption
  • 00:45:39
    of corrosive species ahead of the crack
  • 00:45:41
    tip it all
  • 00:45:42
    appears to be happening at the
  • 00:45:44
    corrective there's not much going into
  • 00:45:45
    the material
  • 00:45:46
    and certainly not a lot of evidence i've
  • 00:45:48
    seen of that
  • 00:45:50
    we do see presence of electrolytes at
  • 00:45:52
    the crack tip so you can see we've got
  • 00:45:53
    sodium
  • 00:45:54
    uh there right down at the crack so top
  • 00:45:56
    left
  • 00:45:57
    uh if we have a look along we've got our
  • 00:45:59
    allying elements and you can see how
  • 00:46:01
    they segregate between the camera and
  • 00:46:02
    the camera prime
  • 00:46:03
    uh quite interesting nothing too
  • 00:46:05
    controversial there other than the
  • 00:46:06
    tungsten
  • 00:46:07
    uh in orange uh which we've found in the
  • 00:46:09
    gamma prime and i think that's
  • 00:46:12
    slightly debated but certainly we see it
  • 00:46:14
    we've seen it
  • 00:46:15
    more importantly uh we've also got
  • 00:46:18
    oxidation going on right down the crack
  • 00:46:20
    so
  • 00:46:20
    in green in the bottom left we can see
  • 00:46:23
    uh our oxidation is happening
  • 00:46:25
    right down right down at the crack
  • 00:46:30
    so the proposed kind of mechanism i've
  • 00:46:32
    caught with that is uh
  • 00:46:34
    sort of fairly similar to stress
  • 00:46:37
    corrosion cracking
  • 00:46:39
    it certainly has a few uh analogies
  • 00:46:42
    which could be made to stress corrosion
  • 00:46:44
    cracking aqueous stress growing and
  • 00:46:45
    cracking at lower temperatures but i
  • 00:46:47
    suppose there's a high temperature
  • 00:46:48
    uh mechanism and and while stress
  • 00:46:52
    corrosion cracking i think is is a bit
  • 00:46:53
    better understood
  • 00:46:54
    i think we're still really trying to
  • 00:46:56
    understand in some
  • 00:46:58
    detail the chemistry of what's going on
  • 00:46:59
    here so this is a
  • 00:47:01
    simplistic view of my my hypothesis
  • 00:47:04
    um but what we see is typical uh
  • 00:47:08
    hot corrosion type features so we've got
  • 00:47:10
    a nickel cable oxides
  • 00:47:11
    uh below that we've got dissolved oxides
  • 00:47:14
    and then below that we see kind of
  • 00:47:18
    an electrolyte um and i've put um
  • 00:47:21
    sodium sulfate there but uh i think
  • 00:47:24
    again
  • 00:47:24
    there's there's a number of compositions
  • 00:47:26
    that i like my electrolyte could be
  • 00:47:29
    so we then see formation of a pit or a
  • 00:47:31
    crack
  • 00:47:32
    uh kind of locally and again
  • 00:47:35
    i don't think there's any um definitive
  • 00:47:38
    way i don't think it's a pit then a
  • 00:47:39
    crack or a crack kind of tip
  • 00:47:40
    i think it kind of probably just depends
  • 00:47:42
    a lot of the time
  • 00:47:44
    and if we move down then to image b um
  • 00:47:46
    in terms of how this
  • 00:47:48
    sort of um electrochemical circuit gets
  • 00:47:51
    set up in the crack
  • 00:47:53
    so we see our electrolytes wicking down
  • 00:47:55
    the crack and we see oxidation happening
  • 00:47:57
    kind of around the crack the cracked tip
  • 00:48:00
    and we'll see a localized anodic region
  • 00:48:01
    region somewhere near or at the crack
  • 00:48:04
    tip but
  • 00:48:05
    really that's where we've got sort of
  • 00:48:07
    acidic conditions and dissolution of the
  • 00:48:09
    oxide
  • 00:48:09
    and the substrate going on and i think
  • 00:48:11
    it's that localized nature really that
  • 00:48:13
    drives
  • 00:48:14
    this this mechanism we don't see
  • 00:48:16
    broad-fronted attacks we see
  • 00:48:17
    very localized attacks which cause you
  • 00:48:19
    know colonies of cracks
  • 00:48:22
    and and then the image on the right i
  • 00:48:23
    suppose is is me trying to
  • 00:48:25
    theorize as to why we see corrosion
  • 00:48:29
    cracks or corrosion driven cracks
  • 00:48:31
    propagating on orthogonal
  • 00:48:32
    uh planes or the q-flames rather than
  • 00:48:35
    where we would typically see them on
  • 00:48:37
    uh you know the the the slip flames the
  • 00:48:39
    one one one of the hedral flames
  • 00:48:40
    and and i think that's because if you
  • 00:48:42
    take a point uh at the cracked tip
  • 00:48:44
    and where you have a low-priority region
  • 00:48:48
    and you grow that out you can see that
  • 00:48:50
    the the gamma prime features which is
  • 00:48:52
    kind of
  • 00:48:52
    or the gamma prime phase which is going
  • 00:48:54
    to preferentially interact with
  • 00:48:56
    uh from that point are all kind of
  • 00:48:59
    orthogonally located uh from from that
  • 00:49:01
    crack tip so
  • 00:49:02
    so that's kind of my theory as to one of
  • 00:49:04
    the reasons why we see this preferential
  • 00:49:07
    orthogonal growth propagation occurring
  • 00:49:12
    so i think with that hopefully i've not
  • 00:49:13
    bought you too much
  • 00:49:16
    but yeah thank you thank you very much
  • 00:49:18
    for listening
  • 00:49:19
    and welcome any questions
  • 00:49:31
    okay so thank you very much for that
  • 00:49:33
    laurie um
  • 00:49:34
    a really really fascinating uh walk
  • 00:49:37
    through some of some of the some of the
  • 00:49:38
    detail there
  • 00:49:39
    on on on what we're doing
  • 00:49:43
    in terms of physics modelling for for
  • 00:49:44
    corrosion fatigue
  • 00:49:46
    um there's a couple of questions come in
  • 00:49:49
    actually there's one particular question
  • 00:49:50
    that's coming
  • 00:49:51
    um about about um connection problems
  • 00:49:54
    and i don't know if
  • 00:49:55
    if that's been a very localized issue or
  • 00:49:57
    if that was something
  • 00:49:58
    that that a number of people have seen
  • 00:50:00
    there uh all i can do really is
  • 00:50:02
    apologize for that at the moment
  • 00:50:04
    um obviously the anarchy will get in
  • 00:50:06
    touch with yamaki
  • 00:50:07
    and talk to them and find out find out
  • 00:50:09
    whether or not there is an issue
  • 00:50:11
    with with the platform although it's
  • 00:50:13
    just a localized issue with that
  • 00:50:14
    particular individual but
  • 00:50:15
    like so if if there has been an issue
  • 00:50:17
    for you then then uh we're very sorry
  • 00:50:19
    about
  • 00:50:19
    it um and uh and i guess it's obviously
  • 00:50:22
    i'm a key to to investigate what's going
  • 00:50:23
    on with this with this
  • 00:50:24
    platform i think it's work cast is the
  • 00:50:26
    platform that we're using today
  • 00:50:28
    um so moving on from that we have got a
  • 00:50:31
    bit of time for some questions and a
  • 00:50:33
    number of questions have been coming in
  • 00:50:35
    um i'm gonna i'm just gonna
  • 00:50:38
    i've been i've been sifting through them
  • 00:50:40
    a little bit to trying to try and um
  • 00:50:42
    to try and get some essential ones out
  • 00:50:44
    but actually there's a load that'll just
  • 00:50:45
    come in in the last couple of minutes so
  • 00:50:46
    i'm going to kick off with a question uh
  • 00:50:49
    directed to you laurie and then i'll
  • 00:50:51
    then i'll have a look through the inbox
  • 00:50:52
    and see if there's any other interesting
  • 00:50:54
    stuff that's coming
  • 00:50:55
    so the first question from um arun
  • 00:50:59
    um can't work out where your from a roon
  • 00:51:02
    you're an engineering manager the email
  • 00:51:05
    address is msn.com
  • 00:51:07
    the question is how do you assess
  • 00:51:09
    standstill corrosion in a gas turbine
  • 00:51:12
    it's a very good question so lori i
  • 00:51:14
    don't know if you've got any thoughts on
  • 00:51:15
    that
  • 00:51:17
    yeah thanks john um i'm just standstill
  • 00:51:21
    corrosion isn't isn't
  • 00:51:22
    a term i'm necessarily familiar with but
  • 00:51:24
    i can sort of
  • 00:51:25
    imagine you're referring to corrosion
  • 00:51:29
    kind of you know not during operation i
  • 00:51:30
    don't know what your view is on that
  • 00:51:32
    john yeah that's that's exactly what it
  • 00:51:36
    is so
  • 00:51:37
    obviously yeah if you leave a if you
  • 00:51:38
    leave a machine uh standing still
  • 00:51:41
    uh especially uh near a marine
  • 00:51:44
    environment
  • 00:51:44
    um how do you protect that from from um
  • 00:51:48
    from corrosion and the answer is um you
  • 00:51:50
    protect it from corrosion by
  • 00:51:52
    by blocking the airways and keeping it
  • 00:51:54
    warm and keeping it ventilated and
  • 00:51:56
    keeping it dry
  • 00:51:57
    um that's not something that would
  • 00:51:58
    particularly cover
  • 00:52:00
    directly in what we're doing but you
  • 00:52:02
    know that's not to say that we can't
  • 00:52:04
    develop corrosion models
  • 00:52:06
    that are appropriate especially for the
  • 00:52:08
    compressor and all sorts of components
  • 00:52:10
    um the answer is we we can develop those
  • 00:52:13
    models and we can put that physics into
  • 00:52:15
    into into into these models but
  • 00:52:18
    primarily what we're doing is looking at
  • 00:52:20
    looking at what's happening during the
  • 00:52:21
    operation and and whether or not you can
  • 00:52:24
    get more life or get more use from your
  • 00:52:26
    engine rather than
  • 00:52:28
    get more standstill time out of your
  • 00:52:29
    engine um i think it is a short answer
  • 00:52:32
    but the answer the long answer is we we
  • 00:52:34
    can develop we can develop those methods
  • 00:52:36
    if you need us to
  • 00:52:39
    okay i don't know if you've got anything
  • 00:52:41
    to add to that laurie um
  • 00:52:42
    you might yeah we can't do that no i
  • 00:52:46
    don't think you're right that was that
  • 00:52:47
    was my kind of understanding as well i
  • 00:52:48
    guess
  • 00:52:48
    it's standstill corrections props
  • 00:52:50
    perhaps a bit more of an issue now
  • 00:52:51
    there's
  • 00:52:52
    lots of assets not being used due to
  • 00:52:54
    code
  • 00:52:55
    so uh so yeah probably quite quite
  • 00:52:57
    relevant question
  • 00:52:59
    okay all right so i'll tell you i mean
  • 00:53:00
    there's a number of questions that are
  • 00:53:01
    coming through on
  • 00:53:02
    the general and there's actually one two
  • 00:53:05
    that's been directed to me so i'm going
  • 00:53:06
    to start answering some of those
  • 00:53:07
    questions if you want to look through
  • 00:53:08
    the inbox and see if there's anything
  • 00:53:10
    that you want to answer there laurie
  • 00:53:11
    then then
  • 00:53:13
    and look at that and um i'll hand over
  • 00:53:14
    to you in a minute yeah
  • 00:53:16
    cheers yeah okay so there's there's one
  • 00:53:19
    here from
  • 00:53:20
    james maddock at rolls royce uh directed
  • 00:53:22
    to me
  • 00:53:24
    um you mentioned the use of predictive
  • 00:53:26
    physics-based approach
  • 00:53:28
    to potentially take credit for park
  • 00:53:30
    power operation to extend the life of
  • 00:53:32
    assets
  • 00:53:33
    the question is do you use this for
  • 00:53:35
    safety critical applications
  • 00:53:37
    is your tool set qualified for this i
  • 00:53:40
    guess the question from rolls royce a
  • 00:53:41
    safety critical operation
  • 00:53:43
    would be potentially an aero engine and
  • 00:53:46
    the the answer is we haven't developed
  • 00:53:48
    it for rolls-royce aero engines
  • 00:53:50
    um but we do believe the physics is
  • 00:53:52
    sound and we do believe that
  • 00:53:54
    that we can certainly support you in
  • 00:53:56
    your operations and i think actually
  • 00:53:58
    some of our engineers are supporting
  • 00:53:59
    your your your teams in terms of um
  • 00:54:03
    physics based assessments of your
  • 00:54:05
    components um
  • 00:54:06
    but primarily the work that we've done
  • 00:54:07
    is on the industrial gas turbine side
  • 00:54:10
    and that's that's that's anything from a
  • 00:54:13
    from a 10-ish megawatt machine uh
  • 00:54:16
    to a to a to a full-size frame engines
  • 00:54:19
    which are operating in power stations
  • 00:54:21
    around
  • 00:54:22
    uk and actually europe um and and
  • 00:54:25
    we i mean there's something there about
  • 00:54:27
    the the tool set you know
  • 00:54:28
    we have we have methods that we develop
  • 00:54:30
    but we don't have a
  • 00:54:31
    set of software that we just license
  • 00:54:33
    that you can plug your engine into
  • 00:54:34
    we we we developed bespoke solutions and
  • 00:54:38
    those solutions can be for the oem who
  • 00:54:40
    runs a fleet of 15 000 engines
  • 00:54:42
    or it can be a bespoke solution for a
  • 00:54:45
    particular power station that has a
  • 00:54:46
    particular type of gas turbine
  • 00:54:48
    and um and we'll develop those methods
  • 00:54:51
    either way and and they have been
  • 00:54:53
    validated we work with the
  • 00:54:55
    power station we work with the insurer
  • 00:54:57
    and we work with the oem
  • 00:54:59
    to to to see the best way of of managing
  • 00:55:03
    that plant and and quite often we're
  • 00:55:04
    very successful in getting more hours of
  • 00:55:06
    operation
  • 00:55:07
    out of that plant before the before the
  • 00:55:10
    before the maintenance window is due
  • 00:55:13
    um hopefully hopefully that answers your
  • 00:55:16
    question james
  • 00:55:18
    um there's a couple of questions coming
  • 00:55:22
    on on the uncertainty modelling and
  • 00:55:25
    and the bayesian approaches if you like
  • 00:55:28
    um
  • 00:55:29
    one of them oh gosh something's happened
  • 00:55:30
    to the order of the questions while i've
  • 00:55:32
    been talking there
  • 00:55:33
    one of them comes from ian mcafee
  • 00:55:37
    you've asked how big a problem is
  • 00:55:40
    uncertainty
  • 00:55:40
    in measurement drift characteristics of
  • 00:55:43
    the thermocouples what is the effect of
  • 00:55:44
    this uncertainty
  • 00:55:46
    and then cam i think you're in hong kong
  • 00:55:50
    cam you've asked may i know more about
  • 00:55:53
    risk-based decision tool
  • 00:55:55
    um i think there's some more come in
  • 00:55:57
    whilst i've been talking
  • 00:55:58
    and whilst laurie was talking there on
  • 00:56:00
    uncertainty it's quite
  • 00:56:02
    it's quite an interesting area i think i
  • 00:56:03
    think you know what the the the short
  • 00:56:05
    answer is i think we need to do
  • 00:56:06
    we need to do a session um i'm gonna i'm
  • 00:56:08
    gonna try and organize a session
  • 00:56:10
    with the imac e and do another one of
  • 00:56:12
    these webinars where we can talk about
  • 00:56:13
    the risk-based methods that we've been
  • 00:56:14
    using
  • 00:56:16
    um getting into the detail the the
  • 00:56:19
    there is a massive problem with
  • 00:56:20
    uncertainty on the measurement side
  • 00:56:22
    um you know what what happens if a
  • 00:56:25
    thermocouple fails whilst whilst you're
  • 00:56:27
    sat there monitoring it how do the
  • 00:56:29
    algorithms deal with that
  • 00:56:31
    uh yes there's drift on thermal couples
  • 00:56:32
    and yes there is
  • 00:56:34
    there is inaccuracy in the measurement
  • 00:56:36
    you know you know if you just think of a
  • 00:56:38
    basic
  • 00:56:39
    sort of rule of thumb type thing then
  • 00:56:40
    you know a 10 degree difference
  • 00:56:43
    in in in temperature can give you
  • 00:56:46
    the order magazine of about 50
  • 00:56:48
    difference on a creep life
  • 00:56:49
    calculation so so you know that that
  • 00:56:52
    level of uncertainty on the measurement
  • 00:56:54
    can lead to quite quite a lot of
  • 00:56:56
    uncertainty in terms of
  • 00:56:58
    your prediction of creep damage and
  • 00:57:01
    therefore the the the amount of
  • 00:57:03
    damage that you've actually done to your
  • 00:57:04
    engine um and
  • 00:57:06
    a simple way of sort of adding all of
  • 00:57:09
    those uncertainties up
  • 00:57:10
    will generally if you just add those
  • 00:57:12
    uncertainties on top of each other
  • 00:57:14
    you will end up in a position where
  • 00:57:16
    you've got so much uncertainty
  • 00:57:17
    that you've got absolutely no no no no
  • 00:57:20
    no
  • 00:57:20
    clear view on how much damage you are
  • 00:57:23
    predicting with your engine
  • 00:57:24
    and so so this is this is where the the
  • 00:57:26
    probabilistic methods
  • 00:57:28
    uh using using um that we've been using
  • 00:57:31
    on on nuclear power stations and
  • 00:57:33
    actually adding up those uncertainties
  • 00:57:35
    in the correct way to give a realistic
  • 00:57:39
    view of your uncertainty that's that's
  • 00:57:40
    where the value really comes into this
  • 00:57:42
    and i think we do need to put a
  • 00:57:43
    presentation together on how we're doing
  • 00:57:44
    that because
  • 00:57:46
    because it's a little bit difficult for
  • 00:57:47
    me to answer in in
  • 00:57:49
    in in this session but um
  • 00:57:52
    i'm going to make that happen i'm going
  • 00:57:53
    to make that happen we are going to do a
  • 00:57:55
    session on that
  • 00:57:56
    um and hopefully you guys can watch that
  • 00:57:58
    and learn more about how we do that
  • 00:58:01
    in the meantime if you want to get in
  • 00:58:02
    touch i can put you in touch with those
  • 00:58:04
    teams and they can talk about that more
  • 00:58:06
    um hopefully
  • 00:58:08
    that's enough on that um pradeep
  • 00:58:12
    where are you from you're from the
  • 00:58:14
    ministry of
  • 00:58:15
    defense standards engineering um
  • 00:58:19
    that's uh that's interesting so you
  • 00:58:22
    asked an interesting question
  • 00:58:24
    do you use machine learning and
  • 00:58:25
    artificial intelligence in your data
  • 00:58:27
    analysis models
  • 00:58:29
    um not in this instance we do have that
  • 00:58:32
    capability within the company
  • 00:58:34
    and and we we can use that capability in
  • 00:58:37
    terms of trending
  • 00:58:38
    what's happening to fleet of engine uh
  • 00:58:41
    and
  • 00:58:42
    and that type of stuff um what we're
  • 00:58:45
    talking about here today is really about
  • 00:58:46
    our physics-based approach
  • 00:58:48
    to to understanding what's going on with
  • 00:58:50
    the damage accumulation within the
  • 00:58:51
    engine
  • 00:58:52
    um and there's a there's a time and a
  • 00:58:54
    place for for those sorts of things so
  • 00:58:56
    if you think about highly complex
  • 00:58:59
    systems
  • 00:59:00
    like the hot gas path within a gas
  • 00:59:02
    turbine it's very very physics driven
  • 00:59:04
    uh machine learning algorithms are not
  • 00:59:07
    particularly suited for that type of
  • 00:59:08
    stuff
  • 00:59:09
    if you think about a situation like
  • 00:59:12
    rotor dynamics
  • 00:59:13
    you can predominantly you're looking at
  • 00:59:16
    a an uncertainty model
  • 00:59:18
    you know there's so much uncertainty in
  • 00:59:20
    what's going on with that vibration
  • 00:59:21
    monitoring that the physics model gets a
  • 00:59:22
    little bit swamped
  • 00:59:24
    from my experience i'm sure there's
  • 00:59:25
    people that would disagree with me
  • 00:59:27
    on that but um and i think that's kind
  • 00:59:29
    of there are scenarios like that where
  • 00:59:31
    machine learning and artificial
  • 00:59:33
    intelligence can can actually be a lot
  • 00:59:35
    more helpful than the physics-based
  • 00:59:36
    approaches
  • 00:59:37
    um that would be my answer i don't know
  • 00:59:39
    if you've got any further views on that
  • 00:59:40
    lori
  • 00:59:42
    yeah i i'd agree with what you've said i
  • 00:59:44
    think as well though um
  • 00:59:46
    with with the with the physics models
  • 00:59:48
    we'll we'll understand them and
  • 00:59:49
    and you know generate them from testing
  • 00:59:52
    like i've shown or
  • 00:59:53
    um analysis um and
  • 00:59:57
    get the model but often will then fit it
  • 00:59:59
    to large data so
  • 01:00:00
    it's not a typical machine learning as
  • 01:00:02
    such but it is
  • 01:00:03
    kind of you know an automated fitting
  • 01:00:05
    process often i think
  • 01:00:06
    um where you have a lot of data and you
  • 01:00:08
    have a model but you need to look at
  • 01:00:11
    how that you know what what your
  • 01:00:13
    constants and parameters are
  • 01:00:14
    for your model you know considering your
  • 01:00:16
    data um so i think that's
  • 01:00:18
    kind of machine learning in my in my
  • 01:00:20
    eyes my mind
  • 01:00:21
    it's more fitting though i suppose yeah
  • 01:00:25
    okay i'm sure that that our answers the
  • 01:00:28
    question and their answers is probably
  • 01:00:29
    open to the whole
  • 01:00:31
    area of debate there but actually
  • 01:00:33
    looking at the clock
  • 01:00:34
    we're pretty much out of time i think i
  • 01:00:36
    mean i don't know if you've been looking
  • 01:00:37
    through the inbox
  • 01:00:38
    and whether or not you've spotted any
  • 01:00:39
    more questions that'll come in and if
  • 01:00:41
    you want to answer any of them before
  • 01:00:42
    before we close out
  • 01:00:46
    uh yeah i think you're right we're
  • 01:00:48
    running a bit low on timeout there were
  • 01:00:50
    a couple i'm just trying to see
  • 01:01:00
    there was one there was one by alfred on
  • 01:01:03
    single crystal blades being good for
  • 01:01:04
    high temperature but not for
  • 01:01:06
    h2s and co2 environments and i think i
  • 01:01:09
    made made mention to this that yeah as
  • 01:01:11
    we
  • 01:01:12
    try to sort of push the operational
  • 01:01:14
    temperatures of these we turn more and
  • 01:01:16
    more to single crystal blades for their
  • 01:01:17
    their creep properties i think in
  • 01:01:19
    particular um but often yeah
  • 01:01:21
    the detriment of the corrosion and um
  • 01:01:24
    you know oxidation
  • 01:01:25
    properties um so i think
  • 01:01:29
    there's we can obviously look to
  • 01:01:32
    different materials but i guess coatings
  • 01:01:33
    as well is another
  • 01:01:34
    another area and several surface
  • 01:01:36
    treatments and
  • 01:01:38
    offer possible solutions that you can
  • 01:01:40
    you can look to um
  • 01:01:41
    but yeah again it's obviously all all
  • 01:01:43
    quite uh sort of
  • 01:01:45
    complex stuff and needs individual
  • 01:01:47
    consideration i think perhaps a lot of
  • 01:01:49
    time
  • 01:01:50
    yeah i agree with that actually i mean
  • 01:01:51
    so that's a really good point by alfred
  • 01:01:53
    um you know that certainly the the newer
  • 01:01:56
    generation of single crystals which have
  • 01:01:58
    got
  • 01:01:58
    quite low chrome contents uh they tend
  • 01:02:01
    to
  • 01:02:01
    suffer more in these types of
  • 01:02:03
    environments than the earlier generation
  • 01:02:06
    and um and actually we we we've um
  • 01:02:09
    we we funded a phd at cambridge
  • 01:02:10
    university just look at that problem in
  • 01:02:12
    particular
  • 01:02:13
    and we never got to the techno economic
  • 01:02:16
    assessment for one of the better word
  • 01:02:18
    which which sort of tries to weigh up
  • 01:02:20
    the argument of
  • 01:02:21
    is is it better to put an expensive
  • 01:02:23
    coating on or is it better just to use a
  • 01:02:25
    a sort of an earlier generation of
  • 01:02:27
    single crystal which has got higher
  • 01:02:29
    chrome content
  • 01:02:30
    it's an interesting subject area and
  • 01:02:32
    i've never got that
  • 01:02:34
    deep in it but it's a really good point
  • 01:02:35
    by alfred i think
  • 01:02:37
    yeah definitely yeah okay
  • 01:02:40
    so i think i think i mean we're running
  • 01:02:42
    over the clock and people have probably
  • 01:02:43
    got
  • 01:02:43
    meetings to get to on teams and that
  • 01:02:45
    type of stuff so
  • 01:02:47
    i think i think we're going to pull
  • 01:02:48
    stumps there unless unless there's any
  • 01:02:50
    other burning questions that you want to
  • 01:02:51
    answer laurie
  • 01:02:53
    uh now i'm happy with that thanks john
  • 01:02:55
    all right well then i guess
  • 01:02:56
    i guess it's time just to say thank you
  • 01:02:58
    very much to everyone who's made it this
  • 01:03:00
    far
  • 01:03:00
    and i hope you found it interesting and
  • 01:03:02
    um like i said
  • 01:03:04
    i'll we'll try and get some more
  • 01:03:06
    sessions organized around the
  • 01:03:07
    probabilistics
  • 01:03:08
    and and that type stuff and hopefully
  • 01:03:10
    speak to you again soon
  • 01:03:14
    okay goodbye
Tags
  • gas turbines
  • sustainability
  • asset management
  • physics-based models
  • Fraser Nash
  • engineering consultancy
  • high temperature materials
  • single crystal super alloys
  • digital assets
  • creep and fatigue