Roughness Revealed Pt. 1 - What is roughness?, professor christopher brown at imts 2012

Well thanks everybody for coming. This is a real honor and pleasure for me to be able to come do this. and a different kind of talk than I usually give. So I spent some time putting this series together and the first part is, "What is Roughness?"" then we talk about measurements and then calculations. Alright, so I'm Chris Brown and I'm from WPI and we're going to start off with a couple little ads for WPI In particular, in just a month we're going to have our second seminar on surface metrology for the americas. to continue a series that we started on international conference of surface metrology and the emphasis is going to be on education with two days of tutorials and a day of workshop err... uh, or a day of standards meetings and there will be triple sessions of the tutorials. We have stuff that will be uh... or material for all levels of users from uh people just starting with surface metrology to advanced. WPI is home to one of the three labs of its kind in the United States academic labs that is dedicated to fundamental work on and standards on surface metrology. We're looking for people to partner with, um.. and we're very happy to have Olympus as one of our partners. So in this talk, I want to discuss the advantages and importance of understanding roughness and have some look at uh... current and future perspectives. So, surfaces cover everything. Roughness is important and manufacturing is often about getting the surface you want with properties you want roughness being one of those important properties, um... and so that's what we're doing in manufacturing. One of the questions though that is worth looking at is, "What is Roughness?" and who decides? So when a product is designed the designer can specify the roughness of the part and when it goes to manufacturing, you have to try to manufacture that and then it goes to quality assurance to just kinda measure it and verify it. So why is there roughness? Why don't we just make everything smooth? Well part of it is cost limitation. To drive the surface smooth services down to uh... to lower levels cost more money but lots of times we don't know what we want in a surface either. and it's always had the roughness and so we're not sure what's going to happen if we eliminate that. So how can roughness be valuable? Wll, it helps with things like adhesion friction, lubrication aesthetics, cleanability, seals it's important here's a list that I've have been putting together and it's by no means comprehensive of all the things that can be influenced by roughness some of these we've done research on some we'd be delighted to have partners to research uh... these things with. If you've got some some additions to the list, I'd love to hear them. I'm sure there's more. Design is uh... primarily done with form and its usually done with smooth, what we call, Euclidean shapes because if we magnify them enough we find there's spheres or cylinders or, um... flat surfaces. But the surface texture or the surface roughness also need to be designed and now one of the groups that I work with, ASNE, has got designations for design engineers are supposed to indicate roughness on surfaces. The most used kind of roughness, over ninety percent of the industry runs on Ra. the average roughness and we'll be looking at that in the next few slides. One of the things to keep in mind though is this cutoff or sampling lengths and that specifies the scale which the roughness will be evaluating. And that's an important concept and that will be a theme in this talk as well as the next too. So here's the average roughness and if nobody says anything and they just tell you the roughness is a... tenth of a thousandth or something they're talking about average roughness in microinches of course. Now, the form is the larger scale and those are the spheres and things that we looked at a moment ago. The finest scale is the roughness and waviness is is what's in between. The other thing that we talked about in ASME B46 is the uh... the lay and the lay is the directionality on the surface or the anisotropy and how its oriented so this is a machined surface we can clearly see there's the direction the tools passed and there's a certain lay to the surface and that can be important for many things. So, in surface metrology as I said, we talk about the three parts to the surface. The largest one is the form so that might be cylindrical. Finest is the roughness, and intermediate is the waviness. And that's the way it can get broken down. Now for the designer they would like to know the relationship between the performance and the roughness and so there might be some sort of curve that looks like this and some sort of an equation you could work where the performance is some function of the roughness Now there's lots of different formss that this relationship might take so here it looks like the rougher the better in terms of the performance. Quite often it's the other way around. We think the smoother, the better for the performance, but often times it's much more complicated and we start thinking about the smoother, the better for the roughness and then there's some tricky things like friction where if it gets too smooth, it's not going to be as good. So smoother is not always better and this is why we see... we're seeing double-sided tolerances on roughness. So here's an example of a rough surface and this has been finished by lapping and one of the interesting things you can see on this is it looks like they can see individual grains on this surface and this is a surface that has been measured with an Olympus confocal microscope and it's a valve seat andone of the wuestions you have is.. "Well how is this going to work for sealing?" and what other kinds of properties and what other kinds of uh... finishing might we use on this? so almost any kind of manufacturing you do is going to uh... is going to influence the roughness. And so, trying to understand these relationships and how they will impact the surface as well as how the surface impacts the performance is important part of being able to design the things. Now quite often when we take a look at some manufacturing process vairable like feed or grit size or something there's arelationship between adding the resulting roughness And often it takes form that look something like this to see how that's generated. Now the stuff we see in all the textbooks and that's the peak to valley roughness function of the feed and the tool nose radius so that can give us a look at the first indication about the roughnesses and now the first question is, "What is the limit of the domain of the applicability on that?" So the development of this is actually pretty simple and it's an approximation. But we're really dealing still with the Euclidean kinds of things and not with something that we'd always intuitively call roughness because that looks like a very predictable shape and in fact, the tool nose might actually be useful here's an example of this with a four millimeters radius tool and I forgot what the feed is on this, but this is what we looked at before and you can see the feed marks but inside the feed mark it isn't perfectly smooth and there's all kinds of things that are influencing it. Here's a a part that was polished and then put in a tumbler made by Bel Air and we can see the progression of time in these tumblers. These tumblers are intended to remove burrs and put on a non directional isotropic surface finish and... Here's another example of this One of the things too that when you put a part in a mass finisher like this, is that you can take corners off. And that is often what the intention is. So here's an example of what we did. This is also measured with an Olympus confocal microscope and one of the great things about this and the neat thing that we're seeing now in surface metrology is that we can image these edges and trace the evolution of an edge through a manufacturing process like this and so that is often very revealing when you think you've got a really nice edge, you can put it under there and you can see how much room there is for improvement. Let's go back for a moment to a turned surface and I'll be in the Measurement talk at 1pm. I'll be talking a little bit more about comparing profiling and uh... 3D measurements but uh... if we just take a look at a profile and these are the most common things that are done today and essentially there's the measured levels profile. Now, we can extract the waviness and that leaves the roughness. So uh... there are some other height parameters and we talked about Ra and we also talked about Rt which is peak to valley and that can be broken up so we can see the valleys and the peaks and that has some usefulness that I'll talk about in the third talk For right now I want to get back to this idea of scale inand I mentioned this cut offentions cut off and so it's the designer's job to put the cut off uh... that you're going to use for specified specifying the roughness on the print and here's something interesting that I think a lot of people aren't aware of. Here's the same profile with two different cut offs. One where the cutoff is 250 micrometers, in other words, 800. And you can see how dramatically that changes the roughness value same measurement different filtering and the RHA changes by forty one percent so jokingly I'll tell the engineers are working with..."If you're having trouble meeting a spec and it's time to go home... change the uh.. change the cut off if nobody's told you which one to use and maybe you can meet the spec." That's something to watch out for an often I've said said this is not specified. So how far can we take this and how far can we go with machining by chip removal and turning? So this is a part that was diamond turned near Keene New Hampsire that uh... at Nanotech Systems. So they faced it first and then they put on a series of grooves as a kind of demonstration project. One of the interesting things that you can see on this is different colors and these colors are not from pigments but these are from light interacting with a very precisely made surface and actually you can see some more of that green there and blue and those things. Let's see what those little marks look like. So here's a measurement, uh... a preliminary measurement we made right in the center of that with the uh... Olympus LEXT OLS4000. And these ridges are five hundred nanometers apart, which is pretty interesting when you think about it because the laser's only four hundred and nine nanometers and we're easily imaging ridges that are 500 apart. So here's a 3D perspective on that also from the Olympus LEXT. and in this is uh... I'll talk about this more in our Measurement talk but this is with a 100x lens uh... with the um... with the zoom at 8x and as i said these are five hundred nanometers apart, these ridges and these are machined in and these are machined in with a diamond tool. so it shows how far 1. we can go with manufacturing to uh... to to produce a surfaces and 2. how well we can inspect them. Now, one of the interesting things about manufacturing surfaces, is lots of times the very best surfaces we try to get are made with the abrasive processes so i thought to be worth it to take a look at some abrasives. And so one of the... one of the interesting things is the abrasives themselves are chaotic at fine scales err.. scales that we can sense, it depends on what size they are. And this is a very interesting measurement. Now that's the true color off the Olympus LEXT. Now this measurement itself is a ten by ten stitch so that we can show a significant part of the abrasive with several particles on there uh... and so each of these ten by ten's is a 124 x124 so over a million elevations a hundred times to make that image and uh... and there it is in true color and you can see some of the light interacting because this is an unused abrasive with some of the abrasive particles in an interesting way there. And i want credit for the students who've done that and I should mention that we were able to train graduate and undergraduate students to be making useful measurements in less than half an hour on the Olympus LEXT. Alright, to kinda wrap things up here So surface metrology, this is measurement and analysis of surfaces to better understand how to manufacture and how they're going to behave so one of the things that uh... makes this uh... particularly interesting is that we're trying to take a look at these surfaces and figure out how we could tell good services from bad and what's interesting is that if we go to sufficiently fine scales, the surfaces are chaotic in other words, they are a jumble of things and they're not easily described by Euclidean geometry and that's the thing that we're most used to talking about geometrically So one of the things that are lab has been doing is trying to find better ways to characterize these chaotic geometries. So what are we seeing in manufacturing surfaces? Well as I showed you the trend is towards greater control of the surface formation and try to push the form and euclidean shapes the finer scales like we saw those 500 nanometer grooves. And the other thing we are seeing is greater sophistication in measuring and analyzing surfaces. And so what are the needs here? Well, we need a better understanding of the processes and the tool workpiece interactions and the performance that can come from this greater sophistication. So, I want to acknowledge some of the uh... people have been helping the lab, especially Olympus, uh... Digital Surf who makes software for doing analysis that we use and some of the companies that are supporting some of the work in the lab like Supfina for super finishing Metso who makes valves and I'll talk uh... and in the third of this series I'll talk a little bit about the software that we've developed. So I thank you for your attention and I'd like to point out that uh... if you're interested and a possible graduate certificate program in surface metrology or abrasive processes, let me know about this because we're thinking about putting some of it together at WPI. Well thank you very much *applause* Yeah, is the levels right, the levels good? Alright, are we ready to go... Haha...