 Good afternoon, everybody, and welcome to the Celebrating Faculty Career series. The series started in 2013, and it was really intended as an outcome of our college's last strategic plan as well as the Faculty of 2020, which had a focus on professional development of faculty through the different stages of their career. So this series in particular focuses on full professors who've been ranked seven years or more and provides an opportunity for them both in part reflective on their career as well as a future look into what they would like to do as a means to really convene a faculty and look at that, a chance for us to maybe look at it in longer term, what things have been done, which where they're going. And it's then followed by meeting with the head of the school and the dean of the college. Today it is my distinct privilege, actually, honor to introduce Dr. Steve Schneider, who is a professor at the School of Aeronautics and Astronautics. Steve did his bachelor, master's, and PhD, all three at Caltech, and he was a research associate there and joined Purdue in 1989. And as I said today, he's a full professor of aeronautics and astronautics, and his research covers all aspects ranging from hypersonic, supersonic, laminar, perturbant transition, low disturbance of wind tunnels, and associated instrumentation. And of course, for the rest of it, I'm going to pass the, I guess he's got a mic on, pass the stage to Steve and look forward to hearing more about this. Thanks. So thanks to everybody for coming. This is an unusual kind of a talk for me. I've never done this before. It's nominally a general audience at least, not specialists, not hypersonics people, not boundary layer transition people. And I've got 50 minutes to summarize three decades of work. So I'm not going to put any equations in. I'm going to skip all the technical details. And I'm going to spend more time talking about the kind of career decisions that I had to make along the way since maybe that's more general useful for a more general audience. Also, I got a little bit of a problem with my left lung today. If I grimace every once in a while, it's not the audience. So how did I get into what I do? An important thing for me was the two years I spent working for the Navy for Mike Reichman here in the middle and Dave Nelson over on the left after I got to my bachelor's before I went back to graduate school. And it started to give me some idea of what I wanted to do for my career, which maybe is a little different from the perspective a lot of people have. I saw the Navy's need to develop vehicles via some cost-effective method that involved engineering science. They paid Dave Nelson for 15 years to study lifting line theory and counter-rotating torpedo propeller design and feed paper tape and IBM 701 machines to figure out better ways to program them. They were willing to pay for engineering science. They had a long-term outlook, but it had to show that it was going to be useful for building vehicles someday. And you could also do a lot of engineering science that could take a long time and may never be useful for vehicles and they weren't interested in that and it's hard to know where to draw the line. I also met a lot of people who wanted to build systems by doing lots of full-scale tests, which for torpedoes was maybe a couple million dollars a test. And that way you have full-scale tests with everything in it, no assumptions, no simplifications, nothing academic. But boy, you could blow up a lot of $2 million torpedoes and still not know what was wrong and 10 years later still be trying to figure out how to make the system better. So it became obvious that the third bullet there is what you really want to do for real vehicle development and since this is an aerospace engineering department, I'm going to talk about vehicles kind of in general. You need some engineering science, you need some testing in different kinds of ways and you've got to buy down the risk for building a better system and that's always been something that I wanted to contribute from the beginning. How do we get better simulations that better approximate flow physics so that we can build things better? And maybe it's only 20 years down the line, down the road, or in my case I've been doing this 30 years and maybe I'm starting to contribute, but still you want to build things better. And Greg I see in the middle there, Greg and I are both working for Dave and Mike back then. So I ended up working on laminar turbulent transition and a lot of you may not know much about what it is, so I thought I'd put a few pictures in. This is a picture from Van Dyke's picture book of laminar turbulent transition of a round jet. There's a circular jet here with a cylinder of flow coming in the stagnant fluid. There's a shear layer where the velocity changes on the edge. That shear layer is unstable. You can see the shear layer starting to develop waves. The waves roll up into a vortex. The vortex develops a secondary instability. These wiggly rings around the vortex. The schwidnall instability that John Sullivan, and I think is not here, measured for his PhD thesis. These instabilities develop and get bigger and then they break down and it flows all irregular, unsteady and turbulent. So this is a classic problem laminar turbulent transition that's been around for 100, 150 years. It's important for a lot of applications. It's hard to figure out, but there's a lot of nice science in it. And if you could figure out how to do it better, then it really matters for a lot of stuff. So that's what sounded like a good thing that I ended up doing for my PhD thesis. This is another picture of this for sort of a more like a vehicle shape. This is from the Notre Dame tunnel up the road. Smoke is very carefully let in way upstream into an o-drive cylinder shape which sort of looks like some kind of a missile shape. The thing is spinning. If you go downstream, as the smoke goes downstream, you can kind of see these ring-like shapes that develop in the smoke. You can actually see the tomein schlichting instabilities that develop in a low speed boundary layer that's effectively two-dimensional axisymmetric here. Since it's spinning, you also get a cross-flow instability. I won't go into any of these in detail. And that gives you these smoke things here at a 45-degree angle. So there are several different instabilities in this case. Two can be seen in one picture of that instabilities that the laminar boundary layer develops that form vertical waves of various cones and then break down. And if you look far enough downstream, it looks turbulent again. So if we want to predict where a transition is, we want to predict these instabilities because that's the mechanism for transition. So when I got to Purdue, then I had to figure out something to work on and Skip may remember this since he hired me back then and I see him in the back there. I got here in 89 and I had a bunch of different ideas for things that I could work on. It was an unusual time. There had been lots of money in engineering science for decades, but the Cold War was ending and the budgets were in free fall to quote my sponsor at AFOSR. I had several different ideas, some of which looked like I could really get them done and I was pretty confident, but they weren't very interesting because the sponsors didn't really see a lot of impact. I had one idea that looked really hard and I wasn't sure I could even do it and it might well fail, but that was the one that got funding because it mattered. And that was trying to move up from low-speed transition to high-speed transition which I really knew nothing at all about. I avoided hypersonic because it sounded like a really exotic specialized area that didn't have very many applications. But it was clear to lots of people including me that you wanted a low-noise hypersonic tunnel with low-noise comparable to flight just like people had developed at low-speed in the 40s and hardly anybody had that at high-speed except for NASA Langley was starting to make that to work. So that's my fourth bullet there. This is a really hard area. It's always including now if you've talked to my students about our tunnels really not running very well right now. It's really hard. It's always uncertain whether we can be successful. We may well fail. Success has been painful and slow, but the sponsors really supported it because the sponsors really saw the need. This was a problem really needed to be solved. So we have to not be afraid to work on really hard problems and not be afraid to fail because often those really hard problems are the ones that really matter. And you've got to be careful that how you go about doing this and a lot of times I wonder whether I was really doing the right thing but I had a lot of really distinguished senior colleagues, Eli Roshatko, Jim Kendall, Ken Stetson, you could run through a long list who all said no, no, no, this is a really good thing. It's going to matter again someday. Well, someday it turned out to be 30 years later, but it did matter. And you know, I don't know about you, but for me at least it's been a big attitude adjustment. I was pretty successful as an undergrad. I was pretty successful as a graduate student. I was used to being pretty successful. I was used to tackling things that I could be pretty sure that I could be successful at. And this was a problem I might well fail at. And still we might well fail at, although we've done okay in the last 30 years. So what is this problem we work on? I'm going to talk about the problem a little bit and show some pictures and I'm going to talk about the sort of career dynamics a little bit. I'm going to try not to run long. I talk too slow then I'll skip some stuff. So this is a picture of a cone flying down a ballistic range at Mach 4, just a part of the cone. It's a shadow graph shows density gradients. You can see the bow shock from the cone on the top and bottom. Near the lower surface here you can see the boundary layer where the frictional effects show up in a fluid flow and this is a spark shadow graph. So you can see the eddies here and the turbulence is all these variations in the density. When the supersonic flow behind the bleak shock goes over those eddies there's compression and expansion waves that are radiated out in the flow and you can see those there. If you look in the upper surface boundary layer you see a thin white line in place that's a laminar boundary layer. There's some little patches of turbulence in the upper surface boundary layer which is different because of maybe a small angle of attack. And there's bigger waves on the front of those patches of turbulence and on the end of those patches of turbulence we'll come back to that later. So again we're looking at where the flow goes from smooth and laminar to unsteady and turbulent and you can see very nicely in this picture that if there's a turbulent boundary layer it's going to radiate a lot of noise and almost every hypersonic wind tunnel in the world has a turbulent boundary layer like that and so a lot of noise which isn't going to be there in flight. So another way to look at the problem is heat transfer. At low speed we don't care very much about heat transfer but quite frequently at high speed we do this non-dimensional heat transfer on the vertical axis versus non-dimensional distance down a cone on the horizontal axis in the ADC tunnel B at Mach 8 and you see some data here in the squares which is the heat transfer going downstream and it decreases at first following the solid curve which is a laminar calculation. And then it starts to rise and it rises actually above the dash curve which is a calculation that's turbulent coming from the nose and that's because we are forming a relatively thin turbulent boundary layer that only starts part way down the cone and there's this transitional region in between so you see the heat transfer here rises by roughly a factor of three. That's a big deal. Hypersonic vehicles get hot. If anybody saw the space shuttle stuff which has been retired for a while you know it gets really hot it's a big deal you don't want it to get too hot and you don't want your thermal protection system to get too heavy or else there's no room left for your payload. This is another example of the same thing. This is from the reentry F-flight of a 13 foot long beryllium cone in 68 for NASA. Non-dimensional heat or dimensional heat transfer on the vertical axis against non-dimensional distance on the horizontal axis. The solid squares are now the flight data. You see the heat transfer rises by maybe something like a factor of six going through transition. It follows laminar at first, it follows turbulent later on. If you pick the onset of transition in the calculation to match the flight data it's pretty close. There's an uncertainty in either the laminar calculation or the turbulent calculation which we could talk a lot about how to estimate but it's not too big but I'll argue that the uncertainty in this transition location is a factor of three and that can dominate your performance estimates depending on your conditions that you want to fly at. Another way to look at this is from this plot which was a plot that Walt Williamson got released from a classified flight and so this is from a public report and all the details are missing and this is common in hypersonics. So we have on the left hand axis the temperature on some thermocouples and degrees rank in. The right hand axis is the same temperature in Kelvin if you like those units. It's some sort of carbon phenolic heat shield which is probably some kind of sphere cone and these are some thermocouples that are halfway down some vehicle that we're not going to see the details for. The horizontal axis is the time from launch in seconds, 1740 up to 1770. Can anybody read that from the back? Is that big enough? And there's also altitude here in thousands of feet. So in about 10 seconds you drop about 10,000 feet. If you look at the temperatures here in the circles at first there are about 400 Kelvin or so, 300 Kelvin a little bit above room temperature and you get to about 300 kilofit and the temperature of the near surface thermocouple 12,000 of an inch deep about a third of a millimeter starts to rise. The deeper thermocouples at 90,000 and 250,000 rise more slowly because the time it takes for the heat to diffuse into the body. You can see this thermocouple starts to rise and the near surface thermocouple rises up to about 2,500 rank in and then all of a sudden there's a change in the rate at which it rises. That's when we get boundary layer transition. The heating rate goes up much faster. It is much larger. The temperature goes up much faster. It goes up to about 4,200 rank in or something and then the thermocouple starts to fail and you get some oddball points up here that are basically garbage. So if you want this vehicle to survive, you see it's getting really, really hot. You don't want it to go turbulent any sooner than you need to. You want to know when that is. Another way to look at laminar turbulence transition where it has an effect is that it has a big effect of separation. And we tell all our undergrads and lots of people probably have heard that the golf ball has dimples on it more or less because the dimples trip the boundary layer to turbulent. The turbulent boundary layer resists separation on the back side of the golf ball going around the backside stays attached longer and the drags lower. So this is a very well-known effect. The mixing and the turbulence makes the flow stay attached longer. These are some pictures I found at Mach 3. There's a mercury capsule shot from right to left in this case down a ballistic range at NASA Ames. It's again a spark shadow graph of the density variations. You see the bow shock in front of this mercury capsule. In the left-hand side, the boundary layer on the face is laminar. The laminar boundary layer is unable to turn this very large angle around the corner of the blunt face. It separates and we have all this turbulence in this big way. If you look at the right-hand case now, the surface is rough. The boundary layer goes turbulent on the blunt face. It successfully turns that very large angle around the corner and stays attached down the back face. There's then a oblique shock here where the attached boundary layer encounters this compression corner going on to the cylinder and the heating on the beginning of the cylinder here behind the oblique shock is almost as big as the heating it is to the blunt face, whereas on the left-hand side, it's an order of magnitude smaller. Also, you can see or imagine that the forces and moments are different and actually the aerodynamic stability of this is vastly different. So transition has a big effect on separation. Another example of that is this plot from the X-33 vehicle. So these are pictures of the X-33, which was kind of like a shuttle vehicle. There's a body flap on the back of it. This upper picture, the colors are proportional. The heat transfer, the boundary layer comes in laminar. The heating goes down. It's blue here. It comes into this flap, which is deflected into the flow. You get very high heating on this flap where it's deflected into the flow. The lower surface case, there's a roughness here that trips the boundary layer turbulent before the flap and the heating on the flap is smaller. If you look at the plot, that's the heating non-dimensional divided by the distance along the flap only. And for a laminar incoming boundary layer, the heating is 60% higher or 50% higher than it is for a turbulent incoming boundary layer. Also, the forces on the flap are different. The pressure is different, and so the control authority of the flap changes, which is a big deal. Frequently, not always. So when I started doing this, I had to bet my career that this application was really going to play out. And there had to be enough different uses for hypersonic transition that even if some of them went away, the space shuttle got retired. It didn't get replaced. That was not predictable 30 years ago. But there had to be enough applications that some of them would still be around and would need this particular technical area. And so there's a bunch of different applications here for lifting reentry vehicles for military and civil purposes, ballistic reentry vehicles for the same things. I'm not going to try to read all those because I'm already 23 minutes in. Cruise vehicles, which could be scramjet powered, which all depends on whether the scramjet guys can be successful or not. But if they are, they've got to build a vehicle. They can't fly an engine by itself. And there was the national aerospace plan. You could try to go single-stage to orbit. That was too big of a step, but you could try to go multi-stage and maybe you'd get cheaper into orbit. And missile defense over the years has turned out to be a big deal. From the beginning, it was clear there were lots of military applications. There were going to be lots of security concerns, the space shuttles, the civil vehicle, but it's all export controlled. But it was pretty clear 30 years ago that Purdue was the kind of place where you could work on stuff that was military sensitive with some security issues, and Purdue would be okay with that. And that's been also things that make this work. So why do we care about laminar turbulent transition? This is a summary slide. I'm not going to go into all the details again, but it increases heating. It increases skin friction and therefore often drag, which is often important. Transition can be asymmetric. If it's asymmetric, you get yawing moments from it, which can be a big deal as it was for the shuttle. There's the effect on separation, which affects control authority and inlet operability for a scramjet floor body and optical distortion. And if you have a really high speed flow, then the high speed and the bow shock will create some electrons by dissociating and ionizing the fluid, and transition will have an effect on that, so it's going to affect signatures and communications, even though we don't say much more than that in public. The other issue with getting into a field, well, can you make anything better? What's the best thing they got? Are you going to be able to make anything better? And this is what we've had for 50 years. So this is a plot, a momentum thickness Reynolds number divided by edge Mach number against edge Mach number. Berkowitz and Marta Lucci collected a whole bunch of classified flight data way back in the Cold War when there was lots of money, and they picked out just blunt sphere cones near zero angle of attack, and they wrote this public AIAA paper by analyzing it and comparing it to a lot of different correlations. They looked at more than 50 correlations. This is just a sample plot for a popular correlation. So we're trying to take the transition location and correlate it with the mean properties in the boundary layer at the point where transition onset is as measured, and they calculated all these flights with the same inviscid boundary layer one-dimensional heat conduction codes, and they came up with this scatter magnitude in momentum thickness Reynolds number, which means two orders of magnitude in arc length Reynolds number, which is a big uncertainty. That's where my factor of 3 plus or minus comes from, which is probably conservative. And these are only blunt sphere cones near zero angle of attack. So the problem is if you're trying to correlate a thing with really complicated physics of which you've seen a little bit against some really simple parameters, it's probably not going to work very well. In fact, it doesn't work very well. And now that we have all these big computers, we ought to be able to do better. Why is an experimental guy talking about computers? So people have said transition prediction is hopeless. We can't do it. And a famous Cold War manager's name I won't mention said, look, it's a black hole. We just kept pouring money and those academic types didn't come up with anything. So then you design around the problem. That's what engineers always do. Or you give up the project. It's too hard. Or they say, oh, we got the secret sauce. We got the secret sauce. Give us all the money. It'll be great. It'll be great. Our company has a secret sauce. We can't tell you what it is because that would be proprietary. But give us all the money. It'll be great. It's too risky. And you run into that a lot. Or you run into not being able to talk about it because people don't want to admit what they don't know. I think after 40 years have a whole bunch of really smart people with a lot of money flying, a lot of vehicles trying a lot of empirical techniques. I don't think you're going to do any better than what they did. We got to put more physics in. And now we have at least big computers. So we can put a lot more of the physics in. And in fact, at lower speeds a lot more physics got put in and gets used by Boeing and Airbus to make predictions on lower speed vehicles. So why can't we do that at high speeds? You got to turn this around a little bit and not just look at what's easy. If something is really uncertain and really hard and it has a big effect then if you can make some progress on it you'll really make a difference. So this has been a strategy more or less and it took a long time to be able to articulate it. It turns out in hypersonics it's been known for a long time that no single ground test facility can simulate everything. Although a lot of times people don't want to admit it because they're trying to make a sale. You list some of the things up there. There are many different aspects of hypersonic flight and you just can't get them all simulated on the ground. Here we've worked to develop quiet tunnels with no low noise like flight. That's good but they're only cold flow at lower hypersonic Mach numbers and Reynolds numbers. And then you can say we'll simulate it and we have big computers now we can simulate it but you still have to make assumptions in the simulations. The conventional tunnels will still be around we have conventional tunnels that cost hundreds of millions of dollars that have a lot of capability but they're not going to match the noise. And so what we really need and what everybody's really been trying to do for a long time in fact this is more or less what engineers do is we take theories maybe expressed in computer codes now we compare them to tests and we try to improve the theories and almost all the theories have approximations to figure out how to take that combination of things and how to make it better so we can predict things better which sort of goes back to my second slide and we're going to have to focus on the things that really look like they're going to matter because we're not going to be able to understand everything on a finite budget on a finite time. If you go into working on a vehicle design program somebody then asks you where's the transition location and Stan Bauslog worked on the shuttle for a long time he went off to a certain vendor that I won't mention in a recorded video and the day he started working for the certain vendor on the X-33 program he walked in the door he's a new hire and he says tell us where the transition location is we got to know he said well I don't know anything yet we haven't done any calculations we haven't done any measurements yet but I got all these people working on designing this vehicle the structure for the vehicle the thermal protection system for the vehicle give us an estimate we'll fix it later because all these people have to know whether it's laminar or turbulent and wet and a crystal ball doesn't work so great another side of this business is one that's maybe particular to this business but I don't think so there's other businesses like this and it's that it's a cyclical business and this is a very old cartoon half a century old which I got from Ken Stetson and you can see this picture of a guy on the park bench saying that I worked in ballistic systems research and hardened reentry vehicles and then basically I got laid off because peace kept breaking out and that's 1964 in the middle of the Cold War right? so the budgets go up the budgets go down and you have to figure out how to deal with the cyclical nature of a field this slide comes from the Wyon wither hypersonics report from 2000 there's kind of a nominal plot here of personnel and hypersonics against time from 1960 to 2000 can you read that in the back? That's too small you know every time there's a big program the budget goes up and then people come into the field and there's a whole bunch of people and then the budget gets cut for some reason and then after the budget got cut at the end of Apollo even before we were done landing men on the moon they were laying people off right because a lot of money and they'd gotten down and then you can repeat that for other cycles where there was a lot of interest and the interest went out and so it's a cyclical business it's not the only one there's lots of them oil exploration is famous for this I mean it's just famous for this and so can you, you have to decide if you're going to be in a business like this if you're going to ride out the down cycles and the idea is if you can sustain what the financial people or the Wall Street people would call a counter cyclical investment if you can sustain it through the down cycles then you're there and ready to go with the technology and the ability when the cycle comes up and you beat everybody else to the up cycle but you've got to be able to sustain it through the down cycle and that's not easy because everybody's saying it's the down cycle is going to last forever right and then when you're in the up cycle everybody's saying the up cycle is going to last forever and neither one of them is true any more than it is in Wall Street with bull markets in bear markets right ok so it's the same kind of thing and then what you have is the personnel then have gone with the cycles and so there were a lot of people close to retirement when I got into this and somebody needed to know this field and a lot of those people were in their sixties and they were all saying yeah yeah yeah do this do this do this will help you will help you somebody needs to know this later on and they give me a lot of help because it's not enough to have a room full of reports somebody asked to know what's in them ok I got 17 minutes I think I'm ok so what do we do with the wind tunnels here well besides that picture I showed you which is really most of it this is a measurement of mass flow fluctuations on the center line of the old JPL tunnel against Mach number from 1 up to 5 this is a paper that's almost 60 years old now and as you look at the mass flow fluctuations and you move the adjustable wall nozzle the mass flow fluctuations went up from like 0.05 percent at a little over Mach 1 and a half to something like 1 percent at Mach 5 and they were on the way up at something like Mach number squared and so what happens is that noise radiated from the tunnel wall keeps going up rapidly with Mach number and there really was no way to get rid of that noise without making that boundary layer laminar so now you have a science project to build the wind tunnel to do the science by measuring on the model and the wind tunnel it's hard and when I first started doing this I didn't know anything and I got invited to go spend the summer at NASA Langley working with Ivan Beckwith here in this picture and some other people and I was a little bit worried you know how's this going to work and Jim Kendall said oh Ivan Beckwith good Kansas farm boy you'll have a good time working with him you won't have any problem he'll treat you well so you know you have also all the people that you work with and Ivan was really great to work with and he's an example of a lot of people whose which I am becoming one of them the old guys who know they don't have that much longer and they don't want to see what they spent their career doing go to waste so they want younger people picking it up and running with it this is another picture of the people Greg was at a workshop like this I think it was the following one but I'm not really sure there's a couple of workshops at NASA Langley during the end of the national airspace plane program during the last boom Langley funded a whole bunch of people to go there for the summer and study things and work with each other you know you end up with a community that you work with that's the people in your area and hypersonics it wasn't a very big community for a long time and end up working with this people it's like a small town for a long period of time and you want to establish a good reputation in that community and everybody has a different style of working and you want to establish a good reputation that sort of fits your style and the other part of this is I've always worked with the mission agencies I've never had any success getting money from NSF and for the mission agencies they've got a mission to achieve you want to know your customer you want to know what their missions are you want to know the people that are going to make the decisions because those are the people that are going to be deciding whether you get funding those are the people that you have to work with to contribute to the mission whatever it is so then we started trying to build quiet tunnels at Purdue and skip remember maybe the first part of this Salomon's not here he'll remember a little bit more of that I got a few friends in the back remember all my cries about when the thing didn't work we started on a Mach 4 tunnel in 1990 it was a shoestring operation there was hardly any money it wasn't clear we were going to get anywhere or I would get tenure or anything one of the questions came up immediately was can you even build a quiet tunnel with graduate students and I got a bunch of graduate students here and so they can all be proud that so far we've been successful and hopefully we will screw up at this point you know we don't have any full-time technicians and there was no way we were going to be able to afford them and the argument always was well if you get good, careful graduate students their whole theses depend on it they can be careful just as careful as the technicians are and we have been successful with that in the end we had a lot of help from the machinists at the airspace lab and our electronics tech John Phillips now another thing that happened which was really unpredictable I mean could not have been predicted was Langley had a huge operation in 1990 you know maybe a dozen, two dozen people working on this millions of dollars a year and all went away it disappeared two-thirds of the aeronautics budget disappeared two-thirds of the air force aeronautics budget disappeared and that was really hard in a sense but in another sense the competition got eliminated and people looked at what we were doing when we were having a hard time and there wasn't anybody else so that's a big issue too who's the competition can you have an edge for it everybody else and then another big step which Skip probably knows more about than I do is Sullivan, Hafe-Hurst, Bateman and a bunch of others from Purdue got us a half million dollar gift to start to build the big tunnel which ended up bringing in several times that in air force money depending on how long you count over what period of years you count that and if we hadn't got money to start building a bigger tunnel where our effort would have been too academic, too sciencey not enough life to contribute to anything and it would have ended a long time ago so you end up depending on lots of people so what do we build we built this Mach 6 quiet tunnel here which is what we're running now they were making runs this morning we were showing it to people we have a 120 foot long driver tube, 18 inch in diameter a converging diverging nozzle 9 inch in diameter runs off so a vacuum tank in the end we got a laminar at high Reynolds number I'll show you just a little bit of that later on I'm skipping all the technical details there really wasn't any competition for a long time except that we weren't sure we could make it work but all this stuff came back on the high end side and now the Chinese actually have a bigger tunnel than we do although it doesn't seem to work better than ours from what we can tell so far we have good optical access so we took the last Langley design ended up moving to Texas A&M and they've been running it down there we took that last Langley design we made some improvements to it that we hoped would work and in the end worked we have optical access I'll show you a couple of pictures and we run this tunnel most of the year 20 or 30 runs maybe a week I don't know it was a long hard slog to build this thing this is my program manager at AFOSR when we started building this thing Steve Walker next to the shop that built the nozzle for us and this is the picture of the nozzle and it was kind of amazing and it still amazes me later on Steve Walker became the HDB2 program manager at DARPA and we were able to help him I'll show you just a little bit of that and now he's actually the DARPA director so I never figured I would ever know the DARPA director and actually I do now so this is just one slide for about eight or nine years we started on the tunnel design in the mid 90s we finished most of the design in the late 90s we had to modify the building Steve Walker took over from a previous program manager who wanted us to build a bigger tunnel Steve said that's way too much money and we ended up downsizing the plan there was a week there I was pretty near panicked because I thought the whole thing would get axed after I was halfway through it but it didn't we started the nozzle fabrication and then the cost doubled and then the cost doubled again and the schedule slipped and slipped and slipped and then we finally got the thing done and it was only quiet at 8psi which was a factor of what 15 lower than it was supposed to be it took us five years to figure out why it didn't work I had a very gracious program manager who was my former student who understood the importance of this but at one point John Schmisher came to me and said look this is the last grant I can get you if this grant you can't get the thing to work then I'm going to have to pull the plug on the budget and I told John look essentially you know the Air Force needs this you and I both know that we haven't figured out there's any reason why it can't work if you stop funding me you have to start somebody else from scratch which is going to take a lot longer and so we got to keep going there's no money then I'll just fail but I'm not going to quit because if the sponsor smells you're going to quit why should they give you any money and there were a lot of days going out in tears almost sometimes really trying to get this thing to work there's a little bit on the nozzle I'm skipping all the technical details my students have all seen this this is the aluminum throat I think actually in a picture this is I think the stainless steel throat this whole thing from beginning down through the throat where we remove the contraction wall boundary layer through a suction system and going out through the end of the nozzle this about 12-14 feet something like that several tons of stainless steel all precision machine that was half a million bucks in about five years or something four years if you look then at the throat region there's a little bit of a schematic of that the flow comes in on the contraction we're going to remove the contraction wall boundary layer we're not really sure we need to do that but Langley did it and we wanted to reduce risk and do everything that Langley had shown was necessary at least from their limited measurements because this was already hard so we wanted to remove that boundary layer here at the throat then we start a new boundary layer goes downstream into the contraction so if you look at that whole nozzle it's like 14 feet long including the contraction there's this little tiny bleed lip here that's three quarters of a millimeter thick that we separate the two boundary layers on and when we machined that it was misformed because we were so afraid of scratching it because it has an almost perfect mirror finish on it we were so afraid of scratching it when we measured it we didn't measure it which was a mistake and you have to make a lot of decisions with this in any program you have to make a lot of decisions with incomplete information because you can't wait till you know everything perfectly you never get done and you couldn't afford it so we misfabricated this like it was round when it wasn't round and essentially that and the shape being not quite right in the end got a little tiny separation bubble here which is like a tenth of a millimeter thick and maybe a millimeter long here in a whole thing that's like three meters long and that we think is what tripped the boundary layer and made it only quiet at a PSI so we made in the end I'm skipping all the details of this and I'm getting close to the end we machined this lip here which is only a millimeter across here and took a little tiny bit of nickel off this surface very, very, very carefully we didn't do it they did it down a micro craft when we ran the tunnel then after that it actually worked so this is pedo pressure fluctuations in black against time this Tom Giuliano's data from eleven years ago now the blue is the pressure at the entrance of the contraction that's stagnation pressure so the run starts at zero seconds here you break a diaphragm downstream of the nozzle an expansion wave goes up through the nozzle into the driver tube goes to the end reflects from the contraction you get a little pressure you see these little stair step pressure drops in the contraction entrance that's every time the expansion wave reflects from the contraction the pressure is going down from 160 psi down to 95 or 90 over the length of like eight seconds this is the pedo pressure so we have a pressure sensor that's high frequency pointed upstream into the flow the first part of the run here is about a quarter of a second to get the run to start up there's a long nozzle we got to get all that air out of there and start up the flow the first part of the run we have big fluctuations here two or three percent of the mean in the end and that's typical for a conventional noise tunnel you have all that turbulence on the nozzle wall it's giving you a lot of fluctuations then the fluctuations actually go up here although if you spread out there part of the trace and look at more carefully there's spikes in there individual spikes in there that's when those turbulent patches go by that I showed you in that picture in the beginning if you remember back to that and then the boundary layer drops laminar and the noise drops two orders of magnitude and then mean pedo pressure goes down because the laminar boundary layer thinner the gas has expanded more coming out from the throat and then the Mach number is higher in front of the P.O. tube so the bow shock is stronger and the pressure behind that shock on the P.O. tube is lower and so we have a lower pressure over here and a slightly higher Mach number with a couple of five here little turbulent bursts coming by on the nozzle wall in the course of five seconds so you know we came back thrilled and you know it's a classic thing like the more you have to beat your head against the wall to get something to work the greater it is what it works so I'm getting close to the end I think I'm okay on time I got about five minutes I spent a long time trying to become a subject matter expert because that's part of this gig that's what I needed to to support the tunnel I spent a lot of time reading papers when people weren't very interested in it learning from the old guys that were going to retire accumulating a whole bunch of papers including lots of reports that aren't readily available taught a specialized course became a Sandia lab consultant that's part of my tagline in the title served a lot of programs as a technical expert did it even if I wasn't getting paid because I wanted to be inside the customer knowing what the customer really needs and you know you develop a level of trust in your community it's amazing to me how much of this business really runs on trust and not contracts how much do you really believe these people that they're going to do what they say to do when they say something it's a lot a lot of work getting things ready hoping that this is really going to pay off someday not really knowing I have a few slides left we worked on then on this Falcon HDV2 program which is not public except a little bit's been released it's some sort of a glide vehicle sort of like this that flew a boost glide trajectory so we made these measurements in the tunnel this slide was released so I'm going to talk a little bit about these this slide I'm going to show you a little bit on the next slide this thing boosted out of the hemisphere comes in does a glide and then it was supposed to go down and do an impact a quadrillion at all supposed to launch out of Annenberg and it was supposed to all be laminar above some altitude I'll show you on the next slide and then they decided well maybe we get or get some measurements in a wind tunnel to check that so they made some measurements at NASA Langley as some undetermined conditions here but at Mach 6 under conventional noise you see this picture the colors are proportional the heat transfer you have high heating in the thin laminar boundary layer by the nose then the heating goes down to the cooler colors of blue as the laminar boundary layer gets thicker then they go back to the warmer green colors when the boundary layer goes turbulent in this funny W shape which you might want to figure out why so you could figure out how to extrapolate it to flight and if you look at when this goes turbulent based on Reynolds number it was a lot sooner than they thought it was going to be well but you know that's noisy flow about how much risk do you want to take with your 50 million dollar flight can we keep the program alive or is it too scary so we went and put the model in our tunnel and under low noise conditions at a similar Mach number in Reynolds number it was all laminar and the tagline on the slide here is the transition Reynolds number was at least twice that's what's released this was Steve Walker's program and it was really amazing because he stayed with us for a long time when we were trying to build the tunnel he was worried whether it would actually ever really work then he went over to the HDD2 program he was the guy running the program when we were able to help him so that was really cool and then it's amazing that he became the DARPA director so this is a plot, altitude and velocity here's these boost glide trajectories this big glide is a long time at really high speed things get really hot if it's not going to be laminar your thermal protection system is going to get really heavy so you probably want it to be laminar they thought it would be all laminar in the white part here above some altitude but then the ground test data says maybe not that two order of magnitude uncertainty comes in where is it really going to be and so what we did as part of this and I'm skipping all the details is develop some improved mechanism based prediction methods which drove down the risk to some extent so this is now my last slide since I've skipped all the technical details and I did okay on the time we spent years and years and years working on this three decades actually more or less from the late 80s through now it's turned out that one full cycle of the interest is three decades which is a long time and now this is the aviation week cover page from a little more than a year ago you can see the title here from the big national blue ribbon committee as the US lost its lead that's a year ago as of a couple of weeks ago hypersonics is now quote the highest technical priority for the defense department according to Mike Griffin who some of you know he was the AIAA president he was the administrator of NASA for a while he's come out and visited us in our tunnel he's now the under secretary of defense for research and engineering so he's kind of like the highest guy for research and engineering and the vice chair of the joint chiefs which is pretty high up there too says that well we lost our advantage in hypersonics but we're not giving up yet we're going to try to get there so now after all this time and all this investment through this really long cycle and lots of things where we almost failed we are now well positioned to aid the DOD in defending freedom so Mike grandson could grow up free like the rest of us in a world that at least mostly is free we hope thanks turn this off questions maybe not a question for this question but thank you Steve for sharing your wonderful story so thanks we always think of successful people just incredibly smart I know you are incredibly smart but it also shows tremendous amount of hard work tremendous dedication and passion so we really appreciate sharing this story well I have one head who I won't name tell me you got to get out of this field the money's all gone out but if you really decide you want to ride out the cycle and it's true in the stock market too right you don't sell at the bottom of the market that's when you don't give up that's when you're tempted but you don't give up if your strategy is to ride it out you ride it out you don't quit and if university people don't ride it out who will right isn't that almost our role if we can't ride out the cycles who will Rain so piece is takes a long time with the experience that nobody is talking about how do you know we have lost the back vision of this audience well you have to ask Mike Griffin right he's looking at the intelligence and he's not going to tell you right or me yeah Steve you talked about vetting your career on a problem you don't go so you can take credit for how you did that as an assistant professor with a seven year tenure cycle doing something that would take forever to put the equipment together and money that the department didn't get classified material that you couldn't publish in open journals I remember talking to you you worked very hard coming up with papers that you could publish so that you would get recognition for and I need to thank you and Purdue that I did get tenure and I didn't get kicked out problem in the context of university but you're successful and fortunately university is yeah it's a huge investment right a lot of people in a lot of risk but if we don't take risk here in the university who's going to take that kind of risk yes really interesting story of the bank on the journal how did you find it eventually how big was it is there manufacturing approaches now it's really really hard to make that because it's going to get polished like a mirror and we spent five years trying all kinds of things that didn't fix the problem and finally Gary Brown at Princeton I ran into at a meeting and gave me one of my many talks about why it's not quiet yet and my long list of things we were trying and he said it's got to be something in the throat and I said Gary I can't change the throat it took years and hundreds of thousand dollars to build the throat I don't have money to make another throat and he said make a surrogate throat oh I didn't think of that so I put the job into the shop queue in Arrow to make an aluminum throat took six months to get to the head of the shop queue Gary made us an aluminum throat that was cheap made out of aluminum nothing fancy no electrophorm nickel no exotic ground mandrel ran quiet at like 20 psi the first time we put it in the tunnel two and a half times better than our super fancy electrophorm throat the first thing we tried that done anything in like four years it's like ah then we had to figure out why and then after all we nailed it down so Steve what was it like when you had to give all those talks explaining why your tunnel wasn't quiet yeah that was painful it doesn't work yet it doesn't work yet but it wasn't too bad because of two things the sponsor and I followed Ivan Beckwith's suggestion at the very beginning he said you get money for this make sure the sponsor knows this is risky don't play games with them make sure the sponsor knows this is risky and we had the sponsors on board and the sponsors have been very supportive all along and the other party is all the technical people I knew they all knew this was really hard and they all knew it mattered I mean I had a meeting at Sandia some of the details of this comment but one of the high level program managers came in and said look I know you're giving this talk and I know you're frustrated that it doesn't work but don't quit we know this really matters this really matters we support you if you can get this to work it really matters don't quit and that really made a big difference it wasn't just me trying to do this and I'm crazy there was a lot of other people saying no no no you're not wasting your time you might get this to work we're behind you other comments looking ahead what what perhaps has changed in the drivers for hypersonic flight have sort of things changed entry with space is it what's making it change is it just the competition you mean in the budgets you know basically if you look at for example what the DOD has been saying publicly for a long time well a couple of years now the chief of naval operations made a public statement a couple of years ago with the fiscal 17 budget I think he said for the first time in 25 years we have great power competition with the Chinese and the Russians he named them and so for a long time you know we were the sole superpower we had lots of anti-terrorism concerns all the high tech stuff we were way ahead anyway we're going to let it all slide cold war is over right and now we got competition and now our competition for example the Russians went ahead and marched into the Crimea right and so now we got competition which is a little scary and spending a lot of money and maybe getting ahead of us and so it's competition that drives the budget and if you work for the defense department it's always the threats that drive the budget if there's a threat that we can't deal with and you have a technology that'll deal with that threat and somebody really says that's needed then that's what drives the budget does that make sense and you know we didn't know it was going to happen you know 10 years ago we would have said maybe low cost to orbit with some sort of scramjet stage was what would drive the budget but doesn't seem like it that seems like the rocket guys are maybe driving the cost per pound down into with another way of getting into orbit right and most things are way out of your control but you know if you're going to contribute to an enormous system you've got to pick a puzzle piece then push your puzzle piece you can't try to do everybody else's puzzle piece unless you're going to play the bandwagon game and try to jump on the bandwagon a little bit ahead of everybody else and that's never been my style