Re: Build a better DC-AC inverter contest
Mads Barnkob, Wed Jul 23 2014, 07:07AM

In brief, the other specifications are :

Must be able to handle up to 2 kVA loads
Must achieve a power density of equal to or greater than 50 W/in3
Must be able to handle loads with power factors from 0.7–1, leading and lagging in an islanded mode
Must be in a rectangular metal enclosure of no more than 40 in3
Will be taking in 450 V DC power in series with a 10 Ω resistor
Must output 240 V, 60 Hz AC single phase power
Must have a total harmonic distortion + noise on both voltage and current of < 5%
Must have an input ripple current of < 20%
Must have an input ripple voltage of < 3%
Must have a DC-AC efficiency of greater than 95%
Must maintain a temperature of no more than 60°C during operation everywhere on the outside of the device that can be touched.
Must conform to Electromagnetic Compliance standards as set out in FCC Part 15 B
Can not use any external source of cooling (e.g. water) other than air
Does not require galvanic isolation

Complete set of competition rules in PDF:

I think the space limitations will throw off most people, you will properly see students in the fields of semiconductors/magnetics win this since you will need some expensive GaN switches to make a transformerless design or some moonmetalwitchcraft hybridtransformer.

Re: Build a better DC-AC inverter contest
Artlav, Thu Jul 24 2014, 11:33AM

240V at 60Hz?
South Korea, Philippines, Peru and a bunch of island states.
Peculiar choice, to say the least.

Also, i was under impression that the DC-DC part is the most lossy and voluminous?
Unless they want to find new ways of making sine wave inverters.

Re: Build a better DC-AC inverter contest
Patrick, Thu Jul 24 2014, 05:45PM

i dont see why this has to be true.

Re: Build a better DC-AC inverter contest
Steve Conner, Thu Jul 24 2014, 05:48PM

I believe they're thinking of the PV market where they typically connect the panels in 450V series strings (to avoid the need for said bulky DC-DC) and feed the output back into the grid with an inverter. The contest is all about improving these grid intertied inverters as there is expected to be a huge market for them in future.

The entries aren't required to operate in an intertied mode, but that is just a minor detail in the control circuitry, it doesn't really affect the design of the power electronics.

Not that I'm in the slightest interested in the contest, move along, nothing to see here

Re: Build a better DC-AC inverter contest
..., Thu Jul 24 2014, 07:29PM

Or you know, most of the US/canada/mexico

It is an interesting design challenge, the requirement on THD really does limit you to running a nice GaN device and some nicely optimized magnetics, although I think that modern server power supplies already are at the power density with
normal' fets and whatnot (running backwards of course, 240v ac in -> 12v/48v out)

Re: Build a better DC-AC inverter contest
Adrenaline, Thu Jul 24 2014, 07:57PM

The DC side ripple requirement is harder to meet than the AC THD. The AC THD is 'easy' to fix with high switching frequency to miniaturize the filter components, however the DC ripple is going to need some clever control algorithm.

Re: Build a better DC-AC inverter contest
Artlav, Thu Jul 24 2014, 09:12PM

Huh?
Aren't USA & north america a 120V area?

Re: Build a better DC-AC inverter contest
Houdini0118, Thu Jul 24 2014, 11:37PM

We are but the feeds that come into the house are 240 60hz with a grounded center tap to give 2 120v buses.

Re: Build a better DC-AC inverter contest
Steve Conner, Fri Jul 25 2014, 07:14AM

Yes, and larger appliances like electric stoves and air conditioners often run off the 240V. In other words, even though Americans "use" 120V they still admit that it's a wimpy voltage and useless for serious electrical power.

I thought GaN was best for lower voltage applications. This looks like a job for silicon carbide.

Re: Build a better DC-AC inverter contest
Myke, Fri Jul 25 2014, 06:58PM

I'm pretty sure GaN's good for high voltage applications since the bandgap is 3.4eV. Unfortunately, it doesn't look like high voltage devices are commercially available yet (EPC says they have 600V and greater devices in development).

Yeah, I'm thinking that this is a more academic research design project but it's definitely possible for a hobbyist to have a good shot at such a power density. They would have to have some specialized tools to work with RF power electronics though.

Re: Build a better DC-AC inverter contest
Carbon_Rod, Sat Jul 26 2014, 10:25AM


The rules do seem rather odd from a design perspective... as most modern engineering goals are driven by cost optimization problems. Accordingly, the winning entry may be impractical for mass production if it uses an unobtanium based component.

I don't think they define cryogenic temperatures for superconductors as cheating... yet...

Re: Build a better DC-AC inverter contest
klugesmith, Sat Jul 26 2014, 03:19PM

Candidate devices will be tested under load for 100 hours, with free air circulation but no external fan.
The rules specify a range of air temperature, but not (yet) the test site elevation or air pressure.

THe Little Box Challenge is a contest of smallness, with the qualifying bar set pretty high.
I think the real-world importance of smallness is debatable,
but the parameter is easier to measure and invites more creativity
than, say, efficiency.

Who remembers the million-dollar Netflix prize?
I just stumbled upon this follow-up critique:

Re: Build a better DC-AC inverter contest
hboy007, Sat Jul 26 2014, 08:12PM

Cui bono? If there's 1M in for the price, the real value ought to be much higher. So who gets the 99%?

Re: Build a better DC-AC inverter contest
Steve Conner, Fri Aug 01 2014, 06:34PM

I think the idea is that the PV industry ought to make at least 99 million in cost savings.

But you might as well ask when Google will see a return on their investment in Google Glass and self-driving car technology.

Re: Build a better DC-AC inverter contest
Hazmatt_(The Underdog), Sat Aug 02 2014, 03:32PM

Yea, if you do enter, and work very hard at your design, expect some rich tycoon to steal your work.
All this is, is a bait scheme.

CM did the same thing years back with his "Clean Renewable Energy Challenge" contest, of which I proposed current, easily practicable techniques for geothermal energy extraction in a closed loop system, that works 24 hours a day. I have not seen the $10,000 promised, and never will.

Re: Build a better DC-AC inverter contest
Patrick, Sat Aug 02 2014, 10:13PM

Who is "CM" ? i like to keep track of reputations, the good and the bad.

Re: Build a better DC-AC inverter contest
Carbon_Rod, Sun Aug 03 2014, 12:17AM

@Hazmatt
I know some of the world's leading engineers who have built numerous geothermal systems.
The story I know... is when a $40 million government subsidized agricultural project was sabotaged by a natural gas company. i.e. they directly approached the partner company and offered fixed fuel prices for 10 years (at 90% discount). Around 130 engineers lost 2 years of design and preparation work about 2 months prior to the main well being made. Accordingly, my Prof at the time had a personal hatred for that pointless scam they pulled.


This contest is a “Black Box” demonstration, and does not require full disclosure.

Re: Build a better DC-AC inverter contest
Ash Small, Sun Aug 03 2014, 01:11AM

This is all about optimizing ferrite losses, I think. Switching losses, etc will play a part too, and, obviously, minimizing copper losses, etc.

Not so much max power as min losses

Re: Build a better DC-AC inverter contest
Steve Conner, Sun Aug 03 2014, 09:42AM

I think I would trust Google to stump up a million bucks more than I'd have trusted CM. For a start, Google probably have the money.

Patrick, CM was a free energy entrepreneur who appeared to be living in a trailer somewhere. We eventually banned him from 4hv because he was so full of s***.

Re: Build a better DC-AC inverter contest
Mattski, Wed Aug 06 2014, 04:09AM

I think that GaN will eventually beat out SiC for the high-voltage market at least up to a kV or two, but it will still take a while to design and qualify especially as most (all?) companies in the GaN power market are growing the GaN on Silicon, which is not the easiest way to grow GaN.

I know that Transphorm has 600V GaN parts available, not widely available at the moment but they will be providing samples to qualified participants in this contest.