Episode: 6 Cliff Notes on 3dscan
Podcast Show Notes
Steve Fletcher, Business Development, Physical Digital. The big change is the use of Digital Manufacturing, collecting and using data intelligently through the manufacturing process and the products life cycle. Standardisation of systems and workflows especially within the Design, measurement and RE space.
This episode of Cliff Notes: Lead manufacturing Podcast. Tristan Bailey talks to Steve Fletcher at Physical Digital, about his work in Aerospace with Rolls Royce and working at the highest level of quality control. Now with Physical Digital the specialise on 3D scanning parts for reverse engineering and quality control. Testing wear and tear post use and for batch samples from medical sizes to boat hulls. Also to rebuild legacy parts when 2D drawings have been lost or may have changed in the years since first fabrication.
Bringing these services into the production cycle and onto production lines as scanners get faster, changes the game for digital manufacturing to full circle service not just for testing for rejects.
... Our business really is service on 3d optical scanning for measurement and reverse engineering but it's more than that you know, there's all the aspects of how acceptable the data is, how accurate the data is how useful the data is ... -- Steve Fletcher
... The first question we ask is why do you want to use this information and they come up with an answer for that it just makes that you know we we have we ave to engage them on how that data can be used within their manufacturing process not necessarily just to satisfy their customers acceptance criteria ... -- Steve Fletcher
... The beauty of the scanning for legacy components you know is that the ability that predominantly the drug in aerospace particularly the drawing is he master some of these components were designed in the sixties and still still being used in the field still being used on in the air so the ability to act to the manufactured some of these components ow using modern technologies requires a historical what a digital model. So reverse engineering the 2D drawing is not necessary that the right approach because the drawing is probably out of date because the component itself has gone through some changes, some concession and those are not reflected on the drawing. Then from there we can reverse engineer the 3D model into a fully compliant 3d representation of that component and then from them we can actually then build and make 2D drawings ... -- Steve Fletcher
... Archiving and storage of all this data that you're working on. Storage of data is a problem yeah you know insecurity of data. Everybody talks about posting information to the cloud that if you're looking at it very sensitive components the ability to actually put it to a storage that is except accessible for everybody is a challenge we have our own data storage facilities and we we archive data sometimes we have as a part of the project we have to remove that data completely from out from our storage facilities because of the sensitivity to know and generally most of the components we get to scan you can talk about. -- Steve Fletcher
Welcome to Cliff Notes where we ask leader, the way.
Steve, maybe you could tell us a little bit about your, your background and how you've got to physical digital today.
Okay, well my background is Rolls Royce apprentice of toymaker moving into product development. I'm five to four engine development moved really from Rolls Royce in to see measurement we using to how it always is a as a global process owner for dimensional measurement. I joined physical digital really from working with them in a number of cases, both with in previous lives with soft software provision and also working with them services, providing to Rolls Royce, so it was an ideal opportunities for me to to join them again and working with them and help develop them their, their business opportunities in the aerospace market where I've a lot of experience great is it sort of measurement QA sorts of areas to stutters define your background in a few words or or actual manufacturing and wealth go from tool make I move down into solely measurement inspection.
So my background is CRM inspection programming quality inspection and things like that. So, you know, my background really is around the quality might my role in Rolls Royce was a global process in a four dimensional measurement for the group. So he was understanding of how the technology could be used and how his us within the organization of Rolls Royce show so that abilities clover, and I'm using that now within physical digital and then then moving into physical digital it felt more like someone you were familiar already and you could you could bring that that sort of connection and that network of other
I think from a smaller companies perspective as well. It's they are very, very easy and the, you know, to adapt quite quickly. You know, they don't have to go through the the sort of the trolls of bureaucracy that large companies put themselves on that in some respects, you know, we can move very quickly. You know what, we don't have is the the resources that we would probably like to have to support some of these activities straight away. But I think we're, you know, we're developing good solutions good sound solutions that can be adopted and that's one of the things that we're trying to promote through our business, you know, our business really is service provision on 3d optical scanning for measurement and reverse engineering, but it's more than that, you know there's there's all aspects of how acceptable.
The data is how accurate the data is how useful the data is and who else can use that information and how did they want to use that information and I think more importantly, why do they want to use that information. There's a lot of work that people are doing where they just say, well, we'll just scan it will take some digital information and that will be and that'll solve our problems, but the first question we ask is why do you want to use this information and if they could come up with a with a with an answer for that. It just makes it you know we we have to engage them on how that data can be used with us the manufacturing process not not necessarily just to satisfy a their customers acceptance criteria. You know, it's the use of the data and the continued use of the data is really important and the bear with the ability to go back to that data without having to re inspect it because you forgotten to capture information on it or you also want to use that information to reverse engineer or to design some holding fixtures or some sort of masking fixtures, or even one a 3d printer, you know, those are the.
Those are the air the questions that we we start to raise and pop to sort of some ideas into our customers head. It's not just about capturing information and having a wonderful picture that you can spend ran on a on a laptop, you know, it's more to it than I did as a practice that I like to follow to and I mean, is it in the sense that it's not just a finished product or some some sort of presentation rendering that you're doing it's it's adding more more cogs to their wheel and putting more processes available to fit into other places on their on their workflow.
Well, I mean, one of the one of the beauty. Beauty of for scanning for legacy components you know is that the ability that predominantly the draw in aerospace, particularly the drawing is a master some of these components were designed in the 60s and still still still being used in the field still being used on on on the in the in the air. So the ability to actually remanufactured some of these components.
Now using modern technology is requires a physical while a digital model so reverse engineering a 2d drawing is not necessarily that the right approach because the drawing is probably out of date because it's gone through the park component itself has gone through some varying changes. Some concession changes and those are not reflected on the drawing and also so so what we do is we we can we can reverse engineer the 2d adoring. We can also compare that to the scan model that we've captured and look at the deviation or the differences of those two and then from there we can reverse engineer the 3d model into a fully compliant 3d representation of that component and then from them. We can actually then build make 2d drawings. That's comparable with the original drawings. So we've we've taken a close the loop in that process. But we've also got some elements of control in there that we put into them about you know How good was the original data. How good was the the scanning information coming from that.
So we control all that and when it comes to things like blade where there's multiple instances of that component reverse engineering that as a taking a scan from a physical component Murphy's law saying the ones you're going to pick out the box is not necessarily the ones that is the correct one. So what we do is a data averaging have a number of components and then reverse engineer the average part and then again we can then compare that back to a 2d converted drawing into 3d model and all. And then we also have the ability that the 3d model we create from the scan data we can compare that through all the different stages. So we've got some basically confidence in the fact that what we deliver out is what the customers expecting to see that there's an interesting point that you've just touched on, and so could you just give us an idea of to sort of frame it here is what is the process with with them with your digital production. I mean, what, what would an average sort of product and maybe in, like you said, an error space blade coming in what what process would that go through where with yourself. Well, first of all, we need to look at the questions are about the drawing you know what what elements of the component from an inspection point of view, do they want to capture from a reverse engineering point of view what elements are they looking to reverse engineer, because sometimes they want to change either the root form or different for the different materials capable, you know, this being used in the in the blade in some cases the material is not available any more the internals of changed regarding the core so they want to ask a lot of we have to go through a lot of due diligence and questions on how what they want to receive and then what we then do is we we either reverse engineer the 2d drawing.
So if there isn't a model, we will convert that model that the drawing into a 3d representation into our CAD system. So we use both space claim annex as well space claim is quite easy to use for lots of freeform surface information, but generally when it comes to aerospace and high end sort of automotive and x orchids here is their cat of choice. So we put that into those packages. We then then scan the component. So we use gum equipment and we use triple scans. We have a number of them because some of their work we do is out on site. So we have mobile solutions, but we also have a scam box in our Birmingham facility. When we do repeat and multiple parts and production parts in there because it's obviously a very contained environment from a temperature and stability and accuracy.
So we then scan the component and then we then look at saying, Well, if it's multiple part what how do we fixed your eyes these, you know, so we we will build up either 3d printed fixtures. So helps us in the receding presenting of the component to the scanner scanner to the component.
So the components is not being manhandle all the time when we scan that and then then what we then do is we do a first off comparison against the using the SPL scan data to the model, we've created and then if we then sometimes we have to then go and create a fully 3d compliant model from scan data and then come pair models tomatoes. So that's the process and then we then go through a whole lot of inspection and give a dimensional report on how far we are out away from the original design because obviously when you start surfacing there's some element of cleaning up simplifying or whatever. So we then do that and do a comparison to give us a quality check or sanity check in some respects, that the data that we are giving back is has representative of the part that the customers expecting and where there is various in changes from the original drawing how to how we explain those away you know through simplified. See you know how this how the service regenerated how the features regenerated and, you know, and they can have that dialogue with with with the customer, then.
Okay. And I mean, does this end up being a very much a team process or a whole company process or do you have like one sort of QA or director or something that you are working. No, really, there's a team of people. We have people.
I mean, our engineers are very, very competent in regarding to the scanning capabilities, but certain people have key skills when it comes to CAD designing so we have to cat designers in house that we use that are compliant with the NX solution which is you know it's not a package that you can just jump on and and provide quality models from a scanning point of view, everybody's capable of capturing information and data and we we train our guys upon how to manually capture data because that gives us the other man I better idea on our they can then use that into fully automated solution you know how to present the part how the how the scanner works.
So it is a this is a whole team. And then we have quality assurance guys that look at, you know, and that generally is people like Dan, who is our engineering manager and Tim, who was a managing director. That will give they are the quality the voice of of quality really to the company goes out. We don't want to be doing is producing just a rough scan and providing a model that we're not fully compliant or very satisfied satisfied with because at the end of the day will come back, we'll have to we'll have to change it, you know, so we'd rather do it right, first time.
Yeah, so I mean you people always have sort of more physical industry experience or used to working with the sort of physical spaces well as sort of turning this stuff into India and that's some of our problems really is to getting people with the right levels of experience. We can we can go to universities and we can pull people off the very, very, very efficient when it comes to dealing with 3d data, you know, manipulating components on the screen but trying to get the right people that understand that how a manufacturing works is a bit of a challenge and that that comes with experience in some respects, and you have to have experience of the industry, you know there's a big void between aerospace and automotive in respect of the levels of people there is, you know, just because of the governance that's around aerospace, you know that need say certain level of understanding that is not learned taught at university or you have to be exposed to it.
I mean with with these industries that are sort of more regulated or maybe in the sort of motorsport industry where it's it's just such a fine grain of accuracy and improvement and define that that sort of bringing these people through or just working people up to to these levels helps and and this this physicals this not so physical and slightly more rapid process can help save them time and save them work. I mean,
I mean it is all about time you know time is the biggest sort of enemy from Forever for for a manufacturer, particularly the Formula One it's needs to be done quickly correctly, you know, the first time because that's Time is money but it's it is training people to understand the processes and it just take a lot out you know from from a small business point of view to, you know, train people on there in the right manner in the right way. Just take time. It's not a it is not an overnight process. It's an understanding of both the technology that we're using, you know, using
structured light solutions isn't a light using a digital camera is not pointing click though a lot of people sell it on the back of it being very easy to capture data quite quickly. Yes, you have got a nice image but how good is that image. How accurate is that image how complete is that image and how useful is that data capture from they're going to be used for for say inspection or reverse engineering and you know they're there is a there's a bet there's an understanding that is required on how to capture that information in the right way using the right methods and world flows and they stays change and they they will change depending if you're doing a whole body of a Formula One car or you're doing the turbine blade or your doings a ceramic core for turbine blade changes and there's lots of synergies between the two industries. When you're coming to composites, you know, in those areas that we have some expertise from Formula One days and automotive days that the aerospace industry are trying to trying to learn and understand a little bit more so we can help our customers by, you know, exposing them to the technology and the use of the technology in a better way.
And, when you speaking about these industries. I mean, what, what sort of sizes of piece. Do you work with them any working with parts or hold shells, or what's your Skype. We do a lot of benchmarking for the automotive industry. So we have four cars and a benchmarking exercise is a complete body doors open door short monotone bonnet. Sure. Part of the engine underneath and everything. So, and also we have done work for shipbuilding companies so you know the whole whole other component. And again, it's very, very hands on customers requirements on the application and then we will learn go down to small injection plastic parts or metal parts that go into the pharmaceutical industry.
In fact, we're currently doing some scanning on some 3d printed head implants. So, you know, they're looking at the quality of the 3d printed or additive manufacturing, should I say components you know how good is this additive manufacturing looks cosmetically bright but does it fit because what I don't want to do is to being held to modify that at the side of the of the patient. So there's all that element, things like that. So it's it's more of an understanding that the size of the components range quite dramatically really intrinsically there's no particular limit on on sort of types of material or size of product where there's always limits. I mean that the environment is a limit sometimes, you know, depending on the accuracy of the data. The, the size of the component that we measure is, you know, if we were looking for surfaces, then we can capture the the the data in a specific way but if we're looking for large surfaces that's got lots of curvature in detail, then we have to then look at that, that project and a slightly different way and use different methods to capture both the large scale data and the actual detailed information that's probably more relevant to what they're looking to do so it's it's it isn't one size fits all. I'm afraid it's it can.
It is very project specific the you find that definitely people are you having to work and adapt to each job and sort of set up the the individual individual voice. I mean, we've we've we've created our own generic workflows that we that we use that satisfy our quality procedures as well because obviously we don't want to be doing is reinventing the wheel. Every time we have a project. So there is different methods that we use and workflows that we use that are common in certain areas, but on the data capture side that can vary quite dramatically depending on the level of detail required the component, the environment that the component is captured in and you know if they want to re measure this at any time you know from a multiple point of view, how do we ensure that we would the we've got the best setup that we can go back to it without again a complete refresh because obviously we want to reduce the time on site and offer you know a service to the customer. That is not paying twice for now so you're not necessarily having to warehouse parts and things apart from your own equipment. But does this become an issue, sometimes for for both the archiving and sort of storage of all this data that you're working with storage of data is is is a
problem and security of data, everybody talks about posting information to the cloud. But if you're looking at very sensitive components. The ability to actually put it to a data storage that is acceptable accessible for everybody is is a challenge we we have data store. We have our own data storage facilities and we we archive data. Sometimes we have as a part of the project, we have to remove that data completely from our from our storage facilities, because of the sensitivity, you know, and generally most of the components we get to scan.
We can't talk about so it's shared sharing data over the cloud is not necessarily the way they want to go. I think there's there's work that we have been working on with people like the the MRC and the MTC where they are looking to develop secure cloud technology and there is there's no lots of things on the shelf that people using and selling but how does that apply to my business, you know, and how do I give access to how do I give them access to this information and how secure is that information that is being challenged and just to sort of sort of step, step, a bit larger scale of things a little bit more of a view.
I mean, have you seen this this industry of mean deep. Do you feel this industry is sort of a development of sort of measurement measurement and and sort of sort of QA, but the sort of digital digital manufacturing space changed in the last three years.
Oh dramatically. I mean, I think I think measurement has come of age. You know, I mean, I think we've come from in an economic tool from aerospace we've come from in an industry where we've had to inspect everything you know as a part of the of the process of confidence in the process because of the processes have not been that great but we're all we've done is control the processes to away where measurement is being the sort of the catch for everything. So it's the system scrap detector of the end of the process where now they're in trying to look at embedding measurements within the manufacturing process which then gives a lot more confidence in how the manufacturing process is working
machine tools manufacturing processes now are getting a lot more accurate lot more consistent consistent. But it's not just the machine tools. It's the the programs that get sent to the moon machine tool. Are they the right level or they're being used by the right people at the right time is all the fixtures right you know and measurement becomes a big part of that, not necessarily the end of the line, but as part of the manufacturing process and we're seeing and I've seen over the last three years where they are using measurement now more effectively within the manufacturing process. Unfortunately, there's a split in camps where manufacturing and quality, not necessarily sit around the same table and agree on the same things. And I think as soon as measurement becomes part of a manufacturing element. I think the world will be a better place for it. If you look at universities. We don't teach measurement or inspection. It's a separate module.
It should I believe it should be part of the manufacturing process engineering, you know, it's a big key is not examined as a cost center at that point it's an Adam and how are you seeing for for the next next three years is sort of a continued education or do you feel this sort of other sort of bigger movements that are helping bring this thing and I think the technology is there. I think there's a lot of technology around from data capture from machining from relatively a manufacturer from joining welding all of these processes are there and the technology is there and the companies are there there's providing the services. I think where there's there is big gaps is either the use of the technology, the understanding of that technology for particular applications. I think the level of competency of people to use this is is on the on the slant now where the people are getting to understand how this technology can benefit them and I think it's the I think there's a there's a big changes a lot of talk around industry 4.0 and the Internet of Things, which
I think it's there. You know, I think there's the I think there's a there's a thirst for industry to adopt it as well which is which is which is also more important. It's not just sort of a buzzword people like to talk at presentations to sell their companies. It's that it is a it is being used and actively being used and promoted within companies because of you know a lot of companies in the UK want to reassure their components, because of the quality aspects of the control of their qualities, rather than shipping it to, you know, third world countries or developing countries. It's they want to be able to have more control of that but also want to be able to ensure that their information if they are offshoring it is it the right level the right standard and delivered at the right time and data measurement form is a big part of that.
And I see and I'm seeing lots and lots of instances where
companies like Rolls Royce and Pratt and Whitney and GE trying to provide a global network of service houses and manufacturing centers globally to support their customers.
So they're setting up in countries that are probably not as advanced as we are over in the West. So how do we get those plants up to speed, supported by local people and an infrastructure of companies around them that can that can support those activities. And that, again, it's not an overnight sort of thing and that's that's where I'm seeing a lot of a lot of growth in those areas as well. And we're seeing activities around measurement and supporting activities, even though we are predominantly a UK based organization we we do support activities in mainland Europe and in the States. So we'd like to a bit more of that. So from a knowledge sharing point of view it.
So this was just gonna come on to with Brexit obviously facing some some changes times in the UK and Europe as sort of adjustments politically and I mean, do you see this is sort of a sort of a challenging time for the market or an opportunity for yourselves and and and being able to reach out or be more individual I think when it comes to engineering.
I think we are very much on the we is like slight driving the Titanic is very we're not going to turn a big ship around overnight, you know, and change just because the political changes or influences. I think we'll see this long term, but I think there's a lot of collaborative work being this already been going off and already been established and I can't really see those companies wanting to change from that vision and I think from our point of view, we'd like to back some of that you know from an aerospace perspective you know those components so may globally. Anyway, they're made within main in your domain in the UK. There are assembled in different parts of your report the UK, I can't really see that changing unless some something becomes better for those companies do you want to change from efficiency point of view profitability point of view, I don't think any government is going to change. I think it will have an influence, but I don't think it's going to change overnight. I think from our point of view we we're we're seeing a lot of activity on companies wanting to reassure their, their component back in into the into Europe which is or back into the UK, which is which is a great thing.
But if we haven't got the infrastructures or companies or support of the government that allows that to happen.
That's where we see you know things are going to the change.
I mean, certainly seen for the last few years, the sort of movement of sort of great British manufacturing. So Greg bridge industry. There's so much expertise here, but it is accepted that it's no business tends to be in just one country that does move around and even if the parts are made here there for selling somewhere else or for assembling or whatever you want to sell a manufacturer where your customers are and
but you might met you. You're gonna work with whether the skills that teach and get that balance of things. So, I mean, what was the decision come back to closing what you see the the next year for sort of fiscal Did you and your clients and going this
well we're we're, we're gonna we were trying to promote more of the aerospace market capabilities we we've got some very large projects that we are working on or potentially working on in the new part in the first part of the year, supporting some aerospace very large aerospace companies on new components and the adoption. The technology. So we're working with both the technology and the end user client to provide complete turnkey solutions
because a lot of these companies don't have the manpower. Sure. So we're we're being those those that people as as an interim step, we want to promote more of the the use of of non contact measurement and the the ability of non contact measurement to be used in the manufacturing process as well and not necessarily the end of the line, but part of the line and how it can be adapted as a part of the manufacturing process. So we're seeing more of that we're working with other clients as well into things like the visual reality where scan data can be used to not from a nominal source, but the actual source. How does that actual money. Look, compared to what the nominal war and how can I use that to simulate to visualize what my component component looks like or assemble of the components and there's some work that the
Japanese man of car manufacturers are looking to use actual data to assemble components, rather than nominal data so that's that's quite interesting. And obviously you you need to be able to capture information on the component. So it's there's some new schools of thought on there, but it's also the use of data, you know, supporting of 3d printing or additive manufacturing there's now becoming more use within the mainstream manufacturing companies, the ability that you know doesn't have to be accurate, you know, as a component itself from a process but long as it's repeatable and can be effective within the manufacturing process
3d printing can be quite useful. We certainly use 3d printing for some of our fixed during so it's changing of the thinking as well. In some respects, and yes, the people are interested to get in contact with yourself to find out where you're doing it and physical digital and where's good to to find you and getting content. Well, we certainly have a website that is because we have a lot of case studies on there and understanding so physical digital dot com is a is a good point of initial source, we're going to we're supporting a number of consultancy exhibitions, where next week we're in the gas turbine user group at the Think Tank Museum in Birmingham and that is on legacy components or for power generation very similar to to the gas turbine aerospace stuff but you know it's the larger sim similar problems. So we're going to be there and you know generally the website is our is is our initial sort of point of contact.
Great. Well, thanks very much for your time today Steve, thank you.
Do you know someone who would make a good guest or have any comments on this episode?
Send me your suggestion, I record 1 - 2 months in advance so can book guests in the diary.