"For our test, we took power resistors and placed them at various orientations to understand different boiling affects," Weiss said. "The resistors used for testing are rated for 260W in a typical air-cooled heat sink-mounted arrangement, where in the dielectric fluid, we obtained power levels close to four times the rated power of the devices without destroying the device."
The SiC Module circuit card and the module were submerged in the Novec, but since the fluid has a dielectric constant that is nine times that of air, surface currents hindered the system from running for long periods of time. The design team developed an inductor for the overall 150kW design and tested that in the fluid. But they were unable to obtain boiling at 200A of power, which was the limit of the supply.
Overall test unit design. (Source: Moog Inc.)
To further the investigation into the reduction of the inductor, Moog performed the same test on a copper strip 4.5 inches long, 0.5 inches wide, and 0.0014 inches thick. This piece of copper, which would have vaporized in air well below the 200A level, was boiling in the Novec and remained at a stable temperature of 55C. These results led to a reduction in the size of the inductors and the overall filter, he said.
With the information from testing, compared to our 10kW design, we developed a controller capable of handling 15 times the amount of power and is smaller in volume by approximately 500 cubic inches. Although we found a tremendous benefit in submerging inductors in fluid, one other area that we saw a benefit in our testing is the dielectric fluid helped maintain extreme explosive arcs. The fuse industry could also benefit from the utilization of the Novec fluid.
Al, this is an interesting development in the power electronics field that parallels the ones I have seen in the computing world. Even on high preformance desktop computers liquid cooling is being used. For servers in a data center, the densities have forced that. I read a while back that Verizon and at&t were both specing liquid cooling for their data and switching center.
I could also see a scaled-down version of this for use in electric cars. Space and weight constraints are even more severe in electric cars than they are aboard ships. Automakers already have experience with liquid cooling and the coolant could also supply cabin heat, instead of the resistance heating or heat pumps now used with pure electric cars (hybrids still use engine coolant from the internal-combustion engine for cabin heat, of course).
Liquid cooling is more efficient, but also more troublesome to maintain. Note that Moog's Naval Systems is developing this. I can easily see the reduced size being very attractive.
The heat has to go somewhere though, so this liquid system would probably interface with some sort of seawater heat exchanger having its own maintenance difficulites.
I can see this unit being quite self-contained - literally sealed. Uncouple the heat exchanger connections, the line in and load out connections, and the control connection and yank the whole thing if it did fail.
Current technology devices that are air-cooled have replacable modules - only a part of them usually fails (well, hopefully only a part fails).
Don't get me wrong - this is really cool stuff (no pun intended). But it doesn't come without a price.
It won't be too much longer and hardware design, as we used to know it, will be remembered alongside the slide rule and the Karnaugh map. You will need to move beyond those familiar bits and bytes into the new world of software centric design.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.