Awesome Jobs: Meet Kevin Hardy, Submarine Builder

By Erin Biba

Kevin Hardy was responsible for building the Deep Ocean Vehicles that accompanied James Cameron on his submarine trip to the Mariana Trench. We chatted with him to talk about how he makes machines to send to the abyss, and how scientists discover new life in the deep.

We know almost nothing about who and what is lurking in the very deepest parts of the ocean. The good news is we have ocean engineers whose job is to battle buoyancy, pressure, and the communications challenges that would attempt to thwart our ability to learn more.

Kevin Hardy, an engineer at Scripps Institution of Oceanography in La Jolla, California, is on the leading edge of the science that is, literally, probing the deep. You may have heard his name before. Hardy was responsible for building the Deep Ocean Vehicles that accompanied James Cameron on his submarine trip to the Mariana Trench. Just off the coast of Guam, the Mariana Trench is the deepest spot on earth (about 11 Kilometers or 7 miles).

The landers, as Hardy calls them, built for the Mariana dive are the result of more than a decade of research, but still just the beginning. Hardy is constantly updating his technology and his goal is to one day create a plug-and-play style vehicle that anyone can use.

“For me it’s the idea of building a pickup truck and the bed is open. It depends on the end user, what he puts in there. We’re trying to make it easy for any scientist, not just the big guys. There will be mechanical interfaces that work on small boats so you don’t have to need the big giant ships. We’re trying to come up with criteria that says, if it doesn’t weigh and more than this and fits in this space you’re good.”

In other words, Hardy wants to use his technology to make the deepest, most inaccessible place on the earth accessible to anyone that has an interest. He chatted with us about what it’s like to make machines and send them into the abyss.

What made you want to get into building machines to study the surface of the ocean?

Years ago, when I was just a little kid, I was reading Diver Dan. In the 50s, when I was a preteen, we had Mike Nelson and Sea Hunt. And in the 60s Cousteau and the Sealab project. I was intrigued by outer space, but the chances to go into orbit were a lot lower then going to the bottom of the ocean.

For me, I really enjoy science and math. Engineering is really applied science. Something like astronomy, meteorology, oceanography, all those are observational science. The way to make observations is tools. Ocean science requires submarines and all the toys.

It’s about moving outside the comfort zone. In oceanography I look as far outside the comfort zone as I can get with the chance to safely make it back. We look for ports and ships of opportunities and going to the deepest ocean depths. That just calls to you. You can’t ignore it.

Science largely goes down to 6 kilometers, but the ocean goes down to 11. So many of the basic questions of chemistry and biology and geophysics are yet to be answered. With vehicles we can go down and discover this stuff.

Is that why we study the bottom?

Besides basic knowledge and pushing the technology, we learn about the earth. Tsunamis, earthquakes, and volcanoes all come from the plates down in the trenches.

There’s also different types of biology that lives down there. Even though it’s high pressure, it doesn’t change. There aren’t seasons really. The only effect is detritus falling from above. It’s actually fairly benign and the chemistry doesn’t change. Things have had the opportunity to stabilize. We think, from the site that we went to at the Mariana Deep, postulating based on our knowledge, that life may have started down there. We had this whole series of postulations but had no evidence and it was unbelievable the things we saw down there.

The reason I consider it important is the massive forces that come out of there. The beginning of life is potentially down there. We actually found some microbes down there that will help us fight against cancer and Alzheimer’s and improving life in general.

It’s crazy to think about how little we know about the deep ocean--and that it’s really only very recently that we’ve started to actually explore what’s down there...

They used to lower things down from ships like a trawl wire to drag along on the bottom and scoop up rocks and sediments and just dump it on the deck.

Over the years we’ve been looking at what’s keeping us out of the deep ocean. First, it was real expensive. They used to lower things down from ships like a trawl wire and they would drag a thing along on the bottom and scoop up rocks and sediments and dump it on the deck and say: “Here’s what the bottom looks like.”

It turns out there’s a lot of biology down there and it’s really fragile.

So you take a seven-mile-long, big piece of wire that’s made out of iron, and it weighs a lot so it took a big winch to lower it up and lay it down, which took a big hydraulic power system or diesel engine. That meant you needed a big boat, a big crew, and a big port. It was difficult to get down there.

Once we got rid of the winch and the wire, now all you have to do is send your vehicle to the bottom. But then you had to solve problems with pressure housings, flotations, fail-safe systems, batteries, cameras. Every time you solve a problem there’s one more to go.

How did you solve those problems to get down to the Mariana Trench?

For our dive to the Trench, I worked with a company to build glass spheres that make the system buoyant and we could put equipment inside them. They’re made out of Pyrex. We’re still making incremental improvements as we go along -- manufacturing and quality control. You solve the first order problem and then you start finessing.

Once we had a substantially resolved solution we had to work out some other problems.

The through-hole connectors, for example. Once you have things inside the glass spheres, how do you communicate with things outside? Glass is useful because you can use infrared to communicate through it, but sometimes you need to have a physical connection. The connections for metal housings was first tried in the glass, but it didn’t work that well. There was a subtle problem that had to do with the loading of the glass. We figured out we needed to create a different metallic interface that was basically a large flat washer that made the surface area resting on the glass much bigger.

The company that was making the spheres was making them for 6.5 kilometers. You look at the constraint and ask: “Why you can’t take it deeper? How can i solve the problem?” It turned out we needed to modify the manufacturers tooling to create a slightly thicker glass wall to go deeper.

Why use glass spheres to control buoyancy?

The first order problem in making a lander is: can we get it down there and can we get it back. You have to be able to control the buoyancy and you have to control the stability.

We can get anything to the bottom, you just make it heavy and tie an anchor on it. It’s getting it back that’s the problem

In 1960, they used gasoline for buoyancy. You know that if you pour oil in water, the oil floats. So they took a whole lot of oil and discovered that some types of oil are lighter than others -- gasoline is lighter than oil. That’s why fire can spread across the surface of water. What they originally figured out is they could create a balloon full of gasoline and tie a heavy instrument onto it. The anchor was heavier then the balloon full of gasoline was buoyant. When you release the anchor the balloon will raise it to the surface.

Why not just use air? Gasoline sounds...kind of bananas.

The problem when you go to depth is compressibility. You’re going from 15 PSI to greater than 16,500. When you compress air into a scuba tank it shrinks down and its volume is less. So Archimedes principle is displacement: if your air is compressed you have less displacement so you get less lift.

When we make the glass it’s very strong, so the pressure is compensated. Fluid like gasoline doesn’t compress much. The glass shell we use is 7-to-8-inches thick. Even though you have that thick shell it displaces more water than the weight of the glass so it’s buoyant.

When they first tried it they used a rubber balloon. But then they moved on and made a shell out of steel. It was basically a 55-gallon steel drum, full of gasoline. You push it over the side into the water and it floats.

That must have led to accidents.

Steel does rust. There are stories of how they used wooden plugs to jam them into rust holes where the gas was leaking out. But they’d drop several tanks so if one sprung a leak they’d still have enough to come back up.

The glass the spheres go back to Corning. They’ve been around since the 60s. Corning made them for the Navy. They just couldn’t go as deep. I wanted to take what we knew and extrapolate and go other places with them.

So what kind of equipment do you load inside the glass spheres? What, exactly, is the lander doing once it hits bottom?

There are three things you might do in the deep ocean with these. One is collecting samples like water, sediment, or animals and then bring them back up. But there’s different criteria to do that. You want to keep temperature constant or keep pressure and temperature constant -- each sample has a different requirement. You might want to just collect data like temperature, salinity, or sonar charts. The last thing would be a navigational transponder. That would be useful to tell where the lander is going based on beacons that you drop down first.

If nobody’s been down there before how do you know where to drop the beacons?

They surveyed most of the trenches already with acoustic systems so we have a good idea of what they look like. We know where the crevasses are or, say, a mud volcano that we might want to land in the middle of its caldera.

How long does it take for the lander to descend?

It goes down about a meter a second. The trench is 11,000 meters so it took 11,000 seconds, that’s about three hours.

How do you follow the lander as it travels?

We don’t have the ability with the systems I’ve made to get live video back. We can communicate with it and know how deep it is, but usually what it does is you send a signal to it and it pings back. Sound moves at 1,500 meters per second. So you can calculate how far down it is based on how long it takes the signal to come back.

What kinds of animals have you guys been finding at that depth?

Who are the kings of the deep ocean? There are still mysteries.

Humans have calcium carbonate in our bones, most vertebrate animals do. But when you go down to depth it turns out at high pressure it forces that calcium carbonate into solution. It will just disappear after about 8 kilometers because of the pressure. Anything that doesn’t use calcium carbonate can survive down there so you won’t find any vertebrates. But who are the kings of the deep ocean? There are still mysteries. How do we get them to approach our vehicles when they’re in perpetual darkness and we’re making noise?

Most animal’s eyes have become accustomed to blue and green even though light disappears at 1,000 feet, it gets absorbed, but these animals still have eyes. Why? They don’t need them. It turns out they have light generating mechanisms, 90-some percent of animals in the deep ocean have bioluminescence.

So how DO you get them to approach? The landers must scare them away, no?

We land on the bottom and these things thump and land and then they’re quiet and just sort of sit there. We’ll have a baited trap there. We don’t really know how the animals communicate. The Russians have identified that they’ll migrate many kilometers to get to a dead animal fall. When we went out there with James Cameron we wanted to show the deep ocean life to people top side so we turned on white lights. But we would be dark for a time and then all of sudden turn the light on and start rolling the cameras.

We have another technique, too. Human sensitivity to light is that we can’t see ultraviolet or infrared. Our visible spectrum is a block and some animals have it a little wider. Fish are right in the middle, like us they see blue and green. But many fish can’t see red. We can turn on red lights and they’ll never know. It’s like walking into a room with infrared cameras, you can’t see it but the camera sees the room totally illuminated.

With that we can see behaviors undisturbed. If the animals are looking for bioluminescence that’s pretty dim. If you turn on a halogen bulb and shoot them right in the eye, that’s gonna hurt and you’re going to scare them away and change their behavior. You want to have blinds like photographers out in the Serengeti.

And then you scoop them up with a robot arm? How do you do that if you’re not communicating with the lander live?

Manned subs are limited by time because of air and personal endurance. But a lander can go down and stay there for an entire annual cycle. This is a whole different class.

Imagine a balloon on a string with a little package on the bottom. The lander floats about a meter off the bottom. But I want to reach the bottom and capture animals and grab the sediment. So I have to deploy something from the sides. I have an arm that’s restrained and released by a mechanism that I can use with a timer or an acoustic thing that triggers the arm. When the anchor releases the arm just swings under the vehicle.

At this time we don’t know what’s around it. So now we’re asking other questions -- what’s the next incremental step? Instead of the arm falling down, what if there was a ramp and a rover came out and it drove and sniffed around and picked things up?

Cameron and I talked at one point about what if you had a lander like a pack mule? It has to control its buoyancy. You have the lander sitting down there and it had drawers that the rover would pull out and it could put samples into the drawers. When lander is full of stuff it returns to the surface.

Right now we have one arm on each side for hydrodynamic balance. You want the wings to be tucked in as you’re going down and on the way back up you want them to be in the wake so it’s stable.

You have manned subs where people they can look out the window and pick things up. But you’re limited by time because of how much air and personal endurance. But a lander can go down and stay there for an entire annual cycle. This is a whole different class.

Speaking of incremental steps, now that your dive to the Trench is complete what are your working on next?

We have some new radio beacons that we’re trying out. Tomorrow we’re going to go out on the ocean, throw this over the side, ballast it so it’s floating, and see how it does.

Radio waves don’t transmit through water very well. Only about 18 inches. That’s the advantage the space guys have over the ocean science guys. They get great bandwidth. We have to use acoustics. When this thing gets to the surface, we get what we call “surface expression.” Once you’re out of the water you can talk to all kinds of things.

So from an engineering point of view, we want to know how much of the device do we actually have to have out of the water? I’m trying to keep it down to a minimum. During the last test in San Diego Bay I had it 2 inches out of the water and we were getting 4 miles of range.

But if the lander is just going straight down, and then coming straight back up, why does it need to send a signal when it surfaces?

We’re always trying to cheat Davy Jones. He likes to keep what he gets and we like to get our stuff back. When it gets back to the surface it calls the ship and tells us where it is. It knows by tracking GPS satellites and then radios its position to the ship.

The vehicle is designed to be long and narrow like a canoe pointing up. We’ve gone down in the Puerto Rico trench, a 17 kilometer round trip, and we were off by under ½ kilometer on the surface just by the hydrodynamics. These vehicles go straight down and straight back up.

But it’s a real big ocean. If there’s a storm brewing or conditions out of your control you want to have multiple ways to find it. It will have a flag, a strobe light, a radio beacon, but having this other tool is another chance to get this thing on board as soon as possible. The biology doesn’t suffer from decompression, but it does suffer from warmer water temperatures. We need to get it on board quickly and put it in the chill box. And also a ship can cost $30,000 a day, so if you’re spending two hours looking for the the lander that can cost money.

We’ve been out in the middle of the night during a tropical storm and you couldn’t see ten feet. It’s a big ocean and it’s always in motion and always changing. There’s a great quote by Nietzsche that I think of when when I’m standing on the edge looking down at 11,000 meters of water: “When you stare into the abyss the abyss stares into you.” If you think you’re ready come on down.

So, what you’re saying is, your job is awesome.

I find everything interesting. It’s fun trying to figure things out. I’ll be 60 this year and when I learn something new I still get goosebumps on the back of my neck. That’s why I like to stick my neck out in places where I’m uncomfortable. Even if I don’t answer a question, maybe I’ll push it far enough along so somebody else will pick it up.

The great thing about being an engineer is I get to go to Norway, Antarctica, Australia... I once landed on an uncharted island that nobody’s been to before except the turtles. It’s lovely to be alive.

Not all science is done in a lab by guys in white coats staring into microscopes. Lots of discoveries require brave men and women to put their boots on the ground and get down and dirty in dangerous environments. Every month we’ll profile one of these field scientists, tell you how they do their job, and explain the science behind what they do. If there’s a scientist or field of science you’re dying to hear more about shoot us an email or a tweet: erin at erinbiba dot com, @erinbiba