3DRC Innovation Committee Update on Fabric Physics | PI Apparel

Okay, my name is Michelle Greenhouse. I'm here for the 3DRC Innovation Committee, and I'm here with my two colleagues, Giulietta and Christina. We are going to share a little bit about our Digital Fabric Physics Interoperability Committee and what we are working on now.

Oops, I think I hit the wrong one.

Okay, so just the 3 of us are here, but the committee consists of a bunch of other people as well who are not here. We have 5 different brands, and I'm not going to read them all. You can see them on the slide. And we have 1 academic consultant with us too, Dr. Jordan Kyosef, who's been very helpful with us on this committee. We have Calypso, who's been helping our project management for this group, and we're really excited to share what we have done. So today we're going to talk to you about our updated SOP, our standard operating procedure. Some of you may have heard about it. We launched it about 2 years ago. And we're also going to share an exciting online conversion tool that we built over the last year. And lastly, we're going to talk about fabric physics validation. It's a brief not comprehensive study of different validations that are out in the market. And then at the end, I think there'll be an opportunity for questions through Slido. And with that, I'm going to kick it off to my colleague, Christina.

Hi, everyone. I'm Christina Rappa from WL Gore and Associates. So, as Michelle mentioned, we've been updating our 3DRC SOP, and this is on data interoperability between different 3D software. So here's a QR code. You scan it, save it in your bookmarks, go to it. You can download the full SOP. What it does is it outlines all of the testing requirements for the different 3D software. So we include Style3D, Browzwear, AssistVideo, Gerber, Optitex, and Clo. So over the last few years, we've talked to all of the 3D software to try to understand what the testing requirements are for physics to be input into their software. So this is a comprehensive guide for that. And the next thing I'm going to talk about is our online conversion tool. So this is a beta version, so it's still in the beginning stages. It's being hosted by IEEE. Because we're an open collaborative team, this is open source code, so, you know, it's open open to everyone, but we are utilizing those equations that the 3D software companies have provided us and looking at the industry standard test methods. And what the online conversion tool does is it helps us convert the units. So grams to ounces and, um, dyne-centimeters to micronewton-centimeters. So that's what it does. Okay, user case. Why do we do this? I'm a material supplier, and what we've heard from you over the course of the last couple of days is that brands are asking their material suppliers to provide 3D material files in whatever format that they're using. But as a material supplier, it's challenging to test multiple times and provide those files in 3, 4, 5 different formats. So what we hope to do with this online tool is to test once on classical test equipment and then convert it into the units that your 3D software requires to consume. And then you can also work with your external labs if you're already sending out your materials and getting those material specs back with standard test equipment. Or maybe you're a bigger brand and you have your own QA lab in-house. So majority of the— well, I should say all of the 3D software requires 4 basic pieces of physics for the materials. So weight, if you have a scale, you can easily follow the ASTM test method to weigh your fabric. Thickness, again, it's a standard thickness gauge. Bending, so this is the calculation for your flexural rigidity, which helps with the drape of the material in the software. And there's a couple of different test, equipment for bending. The one on the bottom left there is the standard cantilever device. And the ASTM method is D1388, which our SOP outlines all the industry standard test methods, ISO, ASTM, some British standards as well. Some labs have the Kawabata machine, which is another way of measuring bending, pure bending, but not everyone has that. So we try to use the cantilever as much as possible. And then stretch. So the Instron machine or a Zwick or any dynamometer where there is a constant rate of extension can give us the raw data for stretch and elongation. And we have a test method here. If you are testing your materials for tensile strength, the labs are doing this test method already. So you can build it into your regular test protocol and have the lab send you their raw data. And that's what we're going to be using to feed into the online tool. OK, so this is what our online tool looks like. And as I mentioned, it's hosted by IEEE. We have two different components to it. One is a static calculation tool, and then the other is the conversion tool. Now, disclaimer here, this is really a tool that is converting the units. So every 3D software has their own proprietary algorithms. We don't assume any of their algorithm. We don't look into their software. We don't know how they manipulate that data. That's left up to them. What we're doing is just providing the test results in the correct units that they can then input. So—

all right.

So now we're going to take a look at the static calculations. So the first one is weight. And if you're a material developer like me, this is going to be coming in really handy for you. So you can weigh your fabric using any size swatch, a square, rectangle, circle, and we've just built in the calculations to convert it to grams per square meter. You can convert quickly from ounces per square yard to GSM. Very easy on-the-fly calculations. The next one is thickness. And so typically in a lab, you're measuring the thickness of your fabric in multiple places. So we allow you to put in up to 3 measurements. And then this will just calculate the average in either millimeters or centimeters, which is what the 3D software requires. Next is bending. So this is our cantilever device. And the traditional test method asks you to input the overhang length of the fabric as it touches a 41.5-degree angle on the device. So you put in your overhang length and then the weight of the fabric in GSM, hit the calculate bending button, and then it'll give you the units in, for Browser Optitex, Vidya, and Gerber in one go. So we've used the calculations that the software companies have provided us to give you dyn-centimeters, micronewton-centimeters, and gram-centimeters. And next we're gonna take a look at the online conversion tool. So right now we have this set up for CLO and Gerber with additional software to come. But I'm going to show you a brief video of how you would input your information.

Okay.

So the first thing is we're converting information from our classical test equipment into CLO, for example. You're going to put in your basic information such as a fabric name, supplier info, knit, woven, then your grams per square meter and your thickness in millimeters. Next, in the bending section, section. Um, in our SOP, there's a description of how to modify your cantilever for the clo bending, but here you'd put in your contact distance and your moving distance in each direction. And now is the stretch. So you're asking your lab to provide you with the raw data. We actually have a format that you can download. What we want is all of the millimeters and the kilograms force. So this is your typical stress-strain data that every CRE machine will provide. And then you upload it in both the warp direction, the weft direction, and then the bias direction. And the software will pick out 5 data points, both the elongation and the force. So you don't have to manually do it yourself. And then you're going to export the CLO file into a CSV format. Okay, and this is what it looks like, and this is what CLO has provided us for the format that they need to consume into their software. All right, so that's CLO. Now, if you're used to using the Flow Fabric Kit and writing down the information on a piece of paper, this will hopefully save you some time. If you're doing the testing already in your lab, just have them send the raw data, upload it into the tool, and then create your CSV file. OK, next I'm going to show you how to import that CSV file into the CLO software to provide the physics. So if you're familiar with creating material files in CLO, you go to the Fabric Emulator. And instead of inputting manually, you click on the folder icon, find your CSV file that you've just created in the online tool, There it is. Open up. Okay. And now it is importing the raw data into Qlo. So this is real-time video here. So it is taking a little bit of time to import it. But I suspect versus manually inputting the data, it's probably about the same amount of time or maybe faster to do it this way. OK. Now it's saved your Fabric file as a ZFAB. And this is just the physics. So your surface imaging is not in there yet. So now you want to take a look at that ZFAB file that's created and make sure that all the data you imported is correct. So you've done the exact same steps. You've uploaded it. Oops, that's the data that was imported in from that CSV file. And now you're clicking Compute. And now your CLO software is doing what they do with their algorithms to consume that raw test data and creating the ZFAB file. And then the very next step after this is to import your texture maps, save it, and then that's your finished file. [NOISE] OK, there's the texture maps. So that's how you create a clone material using the online tool from classical test equipment. And like I said, we also have it set up for Gerber, and then OptiSex and NVIDIA and Style3D will be next. Here's a QR code again to test out the online tool. If you're using any of these softwares, especially the— Gerber, Optitex, Vidya, we would love for you to test it out. We know what the calculations are, but not all of us have access to those software. So it's really a collaborative effort between brands to make sure that it's working right. And then if you have questions and comments or suggestions on how to change or improve or you find bugs, please let us know. Our email address is posted up here. All right. Now, next, I'm going to send it back over to Michelle to talk about Digital Validation.

Thank you, Tina.

How about that? An online conversion tool for physics. Right? Like, I could read that SOP kind of, but it was so technical. I thought, can't we just build something that will just take the data and give it to me how I need it in my software? And she did it. So well done. Okay, so now onto another topic. But at the end, we can talk a little bit more. There'll be questions and answers to talk about the online tool and the SOP. So I'm here to talk a little bit about fabric, digital fabric validation. So there's oftentimes some concerns about is your digital fabric matching your physical fabric? One way to do that, to look at that, is to do a validation. Now, just like in interoperability with physics, different softwares have different methodologies for validating your material. So, once again, we're stuck with an industry problem that we're trying to solve. And I didn't mention this at the beginning, but the 3DRC, our group in particular, is a group of brands and retailers who've come together to solve an industry problem in a noncompetitive way. So, this is our next industry problem. So, let's see, some of our softwares validate their material on a scale, on a sphere, some on a cone, some validate with other certain criteria, others with different criteria. So we are just putting together a non-comprehensive data set of what's out there, and our goal is to build confidence and credibility and trust in 3D. And we are hoping that this opens up a conversation for us to talk about digital Validation. So, as I said, validation is a methodology. It's not necessarily a how-to, but it's really a type or a way that you would validate a material. And what do we mean by validation? There are two different kind of ways that you can validate a material. One is quantitatively with data, actual data, and another one is qualitatively, which is more visual. And essentially it's a way for you to compare your digital material to your physical material. Quantitatively, some examples are you may count the number of flares that are draped over a sphere or a cone. You may measure the elongation or you may measure the volume Those are some examples of quantitative data. Qualitative data may be you visually look at how a material is draped in 3 times and then take an average. This is a brief summary about different softwares who are out in the market. Like I said, this is not comprehensive. We did not look at every single software, just a few, and we sort of put together who is who's doing their validations quantitatively and who's doing them qualitatively. What we learned is that all of the softwares that we evaluated are doing a matching physical environment to a digital environment, meaning they have the same sphere in their digital environment and their physical environment, or cone, and they're all doing a visual comparison. One of the softwares is actually using a quantitative measurement of counting the number of flares on the validation, or on the cone, and the other one, they're also measuring the elongation. One of the softwares we evaluated is actually measuring volume, and they're taking all criteria together and lumping it together and measuring everything by volume, or how the material drapes over the cone. or the sphere. And then lastly, one of our softwares we evaluated is using a machine called a QSEC machine, and that is measuring the bend, and we'll get into that a little bit later. Now I'm going to turn it over to Giulietta, and she's going to talk a little bit about some user cases.

Hello.

Sorry, I'm shorter than Platos. So now for a user case, one of our examples is digital fit. So in a digital fit session, you need to ensure that your fabric drape in 3D is accurate to your fabric drape physically. And part of that is when determining pattern corrections. So if you see the image on the right, it shows drag lines in an area that maybe you would wanna make some sort of correction in, but in reality, the physical sample is not portraying that. So it comes into question how the fabric was actually tested. Is it incorrect, the direction, or did something go wrong in the physics testing process? Which brings us to our next slide of other considerations. There's many variables in fabric physics testing, and we've bucketed 4. So environment, material type, simulation, and end use. So for the environment, one of the questions that gets brought up is, do we need a humidity-controlled environment in order to test a fabric? For material type, do we need validation criteria per fabric type, whether something's a heavyweight woven or a drapey knit fabric? And for the fab— for the end use, do we need to validate on a fabric swatch, or should we just do it on a garment?

Then—

and these are just a few in regards to validation— but our real question is, can the industry align on an avatar and a specimen size? Just based on what we shared earlier, they do vary from software to software. And lastly, how will this be defined a percentage of volume or a flare count. So now I'm going to pass the floor back to Michelle, and she'll expand on some more of these examples.

Thank you, Julieta. It's a group effort. We like to share the wealth for everybody. So yeah, those were just a smattering of the thumb. We had many, many variables on the slide, but it was just too many, too much text, but you know, you can imagine, you know, all the different variables are at play here and how difficult it is to come together with all the different softwares. So we have a big problem ahead of us, we think, but I think we can solve it. So on this slide here, I have some questions some examples of what it means to count a fabric flare, fabric by the flare count. That's the image on the left-hand side. The image also is reflecting measuring the elongation from the tip to the ground. So those are more of the quantitative measurements. The image on the right-hand side is an example of how you might measure volume where you are draping the material over a sphere, and then using a— some sort of a slider to measure, has this increased by 10%, 5%, and you might drape multiple different fabrics over it to get an aggregate. The example here is the QSEC test, which has an ISO standard, and what this test does is actually measures the drape coefficient, and this is used particularly for bend, where, like I said, in our analysis we did have— look at one company who was using the drape meter, or I think it's also called the QSEC machine. How you receive that test information back is also varied, and perhaps we could align on one way for that to come back to us. But we noticed that there were way to see the data come back to you visually, like the image on the left, or in a more numerical format, like the image on the right. And then we're going to leave you today with some conversation starters. So, first, just maybe a show of hands, how many people in the room think that fabric physics validation would help build confidence and trust in 3D?

Oh, wow.

Yes, okay. How many people think that this validation could inform a potential digital tolerance? Okay, getting there. And then the big question would be, does the industry even want a digital tolerance?

Yes.

Yes.

You know, physical fabric has a tolerance, so why can't a digital fabric have a tolerance as well? I think that could really help build the trust in 3D, especially if it's standardized across the board for all softwares. So, we can't talk about every single thing today, but we have a few bullets here on the right-hand side if you want we can talk about later, or I think we also I also want to leave you with, would the industry like the 3DRC Innovation Committee to continue this discovery? I have one yes, yes, yes.

Good job.

Okay. Okay, but we're also going to leave you with this most amazing online tool, QR code, and our SOP. So— [FOREIGN LANGUAGE] Yes. I don't mean to rush this It's okay. I'm Sandy Eda.

Sorry.

I just want to say that you have witnessed some of the most amazing work that our 3DRC group has brought to bear. This came out of a PI event years ago when people said we want an online tool where we can standardize fabric physics, and this committee has delivered it. What a fantastic representation.

So thank you. Really, no, thank you.

And if any of you have team members that are interested and want to be part of this work, please, please, please, it is all volunteer. They met every week during the pandemic, and it has been just a multi-year effort to get here. And I just honestly cannot say thank you enough. You are transforming our industry.

So that's it.

Questions?

I think there's a microphone that was going to go around and then the Slido.

Hi, thanks so much. That was really cool. I'm so glad you guys did that. My question is kind of a stupid question, but I'm curious about it, which is you guys had to go through so much work to kind of standardize this so we can input these units into these different tools. Why— did you guys have a chance to ask these software providers why we can't just input the standard measurements into their software?

We did actually work a lot with all of the different softwares.

Yeah.

I'll let you answer.

I would say every 3D software feels like their version of the material physics is the best. So it's really challenging to have them collaborate themselves and come up with that industry standard. So I think what we've done with the online tool is we understand everyone is, all the software is unique, but we don't wanna test 6 times. We just wanna test once. So we respect what they're doing and just input it into the units that's required. So it's tough. I don't know if the 3D software companies out here would agree with that or not. Like, they feel like theirs is the proprietary, the best. So— I see a Style3D guy right here. I was in your presentation yesterday.

Yes?

You—

it might have been by accident—

you mentioned all the 3D software for this, but you didn't mention Browzwear. Ooh, yeah.

Yeah.

OK, so— so as you know, Browzwear has their Fab Analyzer, their own test equipment. Which they feel like is the best representation of their test data. We have one of them at Gore as well. We utilize it as well as our industry standard test equipment. We actually use the Visu conversion tool as well to convert some Fab Analyzer test results into CLO. So they've built that already. But what we wanted to do is really focus on classical test equipment. And for browseware, the weight, the bending, and the thickness, it's pretty straightforward. It's industry standard test methods. But the stretch elongation, they're using dynamic testing. So multiple iterations and then their software is pulling out values. We don't want to like dig into that. We don't want to reverse engineer it. We want to respect what they're doing. So that's why we have an we haven't included it in the online tool yet, but I would encourage you to check out Vizzu's conversion tool too.

There's another question here. Oh. The main softwares have different simulation engines. How do they align to create trust regarding digital materials?

You want to take this one?

Sure.

Okay.

So, that's again one of the many variables that we have out here, and that is their secret sauce, the simulation. And so, we don't want to do anything that's that's going to interfere with that. And like you said, we respect that. We can really just work on the data that's there. And that's kind of why we want to standardize that. Like, I don't know how much more I can say about that.

But yeah.

Any other—

oh, yes.

Hi.

I'm from the— have all the technical, the collaborative, like that hard facts, right? Those, those hard facts. Because the other side, when I saw the, uh, the sick families, so do you also take into consideration, um, soulmate and sexual fertility and how to pattern? Yeah, that's how Patrick said. that sort of thing. The defence, it's gone from preserving the US old-fashioned banknotes to get those things out, not print the zone. Yeah, there's many, but also, yeah, I'm at the shop at Gresington, you know, stuff like Sorareto Monsignor, and they just come in and out. Yeah, so that strikes they get the scarf back.

Yeah, that's, that's a really great question. Um, well, that's why we want a tolerance, right? Um, but no, we did not. We, I mean, we know that that's a variable. Oh, okay, great. Um, yeah, so that's why we want to align. Can we do it on a fabric square, or do we do it on a fabric circle? Do we do it on a garment, or do we do it on a skirt. You know, if we do it on a garment, is it a simple garment or is it a complex garment? So that's kind of what we need the industry to align on. There is different simulations in every software. There's seam allowances. There is 100 different variables that we could talk about. So I think that's kind of what the committee needs to explore and figure out how we can align on a standard. They've done it for physical fabric, so I don't see why we can't [FOREIGN LANGUAGE] Yeah. But that's a really great question. Other questions?

What are some practical steps brands and manufacturers can take towards testing in a more standardized interoperable way?

I have a good answer to that. One thing that I've learned is just kind of remove as many variables as possible. So have maybe, if it's possible, have one person do the testing in in the same location, in the same place every time. And make sure that that testing is as close to a lab testing environment as possible, meaning the table isn't, you know, a card table that could move around, that your equipment doesn't move, you know, from day to day, that the humidity is pretty much the same as possible in that room. And then you take at least that variable out, like Jim didn't test it today and then Mike I tested it yesterday and Sarah tested it Friday and they had different temperatures in different regions and different whatever. So at least that variable is gone. But yeah, I think that means we're wrapping up, right? Yep, that's it. Thank you.