Stories
Slash Boxes
Comments

SoylentNews is people

posted by janrinok on Wednesday July 23 2014, @08:51PM   Printer-friendly
from the more-bang-for-your-buck dept.

Google and IEEE (The Institute of Electrical and Electronics Engineers) are not satisfied with the power density of current inverters and are willing to pay for a major breakthrough.

The winning inverter will be the one that achieves the highest power density and meets a list of other specifications, as determined by a panel of judges, while undergoing testing for 100 hours.

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/cu in
  • 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 cu in
  • Will be taking in 450 V DC power in series with a 10 ohm 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
  • Cannot use any external source of cooling (e.g. water) other than air
  • Does not require galvanic isolation

Our testing philosophy is to not look inside the box. You provide us with a box that has 5 wires coming out of it: two DC inputs, two AC outputs and grounding connection and we only monitor what goes into and comes out of those wires, along with the temperature of the outside of your box, over the course of 100 hours of testing. The inverter will be operating in an islanded mode---that is, not tied or synced to an external grid. The loads will be dynamically changing throughout the course of the testing, similar to what you may expect to see in a residential setting.

A more detailed description of the technical specifications for the inverter, the testing procedure and the requirements for technical approach and testing application can be found in this document (pdf).
[...]
Does Google own the intellectual property created during the competition?

No. Google is not requiring any IP or licenses be granted except a non-exclusive license to be used only for the purpose of testing the inverter and publicizing the prize. We want entrants to benefit themselves through the advancements they make in order to help grow an advanced power electronics ecosystem.

However, in the spirit of advancing this power electronics community, Google may choose to make public some or all of the teams' high-level technical approach documents. These documents outline the key innovations used to overcome the problems which currently limit high power density in inverters. They do not need to reveal any IP. We only want to show the world what techniques are possible in creating a new generation of power electronics. More details on the requirements for this document can be found in the full terms and conditions and detailed inverter specifications for the prize.

This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 2, Informative) by tftp on Wednesday July 23 2014, @09:20PM

    by tftp (806) on Wednesday July 23 2014, @09:20PM (#72985) Homepage

    The market of DC-AC inverters is already not that huge; the market for low power inverters (up to 2 kW) is even smaller; the market for inverters with such a high input voltage (420V) is ... just Google. Normally you would get that voltage from a solar panel; however a non-synchronizable, 2 kVA inverter is of no use there.

    The R&D that goes into such a project would be considerable; higher for a company that has a chance to achieve results; lower for one-man teams who have no chance to outdo the existing designs.

    Besides, hardware companies are not setup to participate in such competitions. They work on their own products; however there is little added value in pushing the power density up by 0.1% if the price of parts increases by 500%. Companies do not make products in hope that Google will like them. Companies operate on more realistic plans, those that have a specific customer in mind, or a specific market that is ready to be sold into. For example, a 12V DC to 120V AC market is well defined, and you can always sell into it.

    • (Score: 3, Informative) by pe1rxq on Wednesday July 23 2014, @09:50PM

      by pe1rxq (844) on Wednesday July 23 2014, @09:50PM (#72991) Homepage

      This thing sounds exactly like an inverter for solar.
      As soon as you consider non-US households its power output fits perfectly.

      I currently own an inverter in the same 2kVA range. (It provides about 2/3 of my total electricity use)
      It does not operate in island mode, but if they really do get the wanted density it probably won't be hard to use the same technology for grid connected devices.

      • (Score: 0) by Anonymous Coward on Wednesday July 23 2014, @10:39PM

        by Anonymous Coward on Wednesday July 23 2014, @10:39PM (#73015)

        > This thing sounds exactly like an inverter for solar.

        Yes. That's exactly what it says on the website.

      • (Score: 2, Informative) by tftp on Wednesday July 23 2014, @10:48PM

        by tftp (806) on Wednesday July 23 2014, @10:48PM (#73020) Homepage

        A household inverter does not require high power density. A few more cubic inches won't make a difference if your inverter is in the garage. What you are looking for, though, is reliability and low cost - and both of these are better served by low power density. (It buys you larger safety margins, cheaper and physically larger parts, larger heatsinks.)

        I have a 6 kW solar inverter (made by SunPower.) It is not particularly power-dense. I cannot care less.

        With regard to ease of adding the grid tie, I'd be wary about such a claim. I must admit that I never looked into how this is done; but I would presume that you need a pretty accurate control over the voltage and the phase of the signal that you are feeding into the grid, with current transformers and a proper feedback. I wouldn't do it with anything less than a Blackfin DSP, for example; but perhaps an aficionado of analog purity would come up with a sea of OAs instead :-) Regardless of how you do it, the grid tie is bound to be a major headache - at least because no utility will ever allow you to connect a device into their grid without some serious testing - and that device may not ever fail, otherwise it will burn the house down in no time. My inverter has some impressive relays, as you'd expect them to be for switching 6 kW (at 240V RMS that would be 25A.) They alone will ruin your power density. Make no mistake, this inverter that Google is soliciting bids on, is a very, very custom part that you probably won't be able to sell (profitably) to anyone else. Perhaps even the design would be not very useful for other applications. Most products on the market are not of revolutionary design; but they are cheap and reliable - and that's what sells.

        • (Score: 2) by WillAdams on Thursday July 24 2014, @11:25AM

          by WillAdams (1424) on Thursday July 24 2014, @11:25AM (#73206)

          Density has a huge impact on shipping costs, esp. if the device is coming from China.

          • (Score: 1) by tftp on Friday July 25 2014, @12:44AM

            by tftp (806) on Friday July 25 2014, @12:44AM (#73543) Homepage

            Perhaps. But the shipping costs themselves are insignificant, compared to the cost of the device. Those invertors cost above $1K. You can estimate the shipping costs if you consider that shipping of dirt cheap but physically large plastic tubs and lawn furniture from China is still profitable.

      • (Score: 2) by frojack on Wednesday July 23 2014, @11:11PM

        by frojack (1554) on Wednesday July 23 2014, @11:11PM (#73027) Journal

        As soon as you consider non-US households its power output fits perfectly.

        If you look at their required wiring diagram, page 11 of the pdf, you can get to US power requirements simply by
        putting a center tap a the isolating transformer.

        I have no clue what voltage Google runs their data centers at.

        --
        No, you are mistaken. I've always had this sig.
    • (Score: 2) by frojack on Wednesday July 23 2014, @09:57PM

      by frojack (1554) on Wednesday July 23 2014, @09:57PM (#72995) Journal

      Yeah, 420V threw me for a loop too.
      Its not unusual to find 440V (AC) on ships, but its an odd number to want in DC unless they are coming from solar.

      I suspect they will have takers. Even companies working on their own projects have EE hours to spare, and if it works and they can keep the IP they can build a market around it. If it is a good fit for solar input there will be a home market.

      --
      No, you are mistaken. I've always had this sig.
      • (Score: 2) by frojack on Wednesday July 23 2014, @10:02PM

        by frojack (1554) on Wednesday July 23 2014, @10:02PM (#72997) Journal

        Typo, meant to day 450VDC.

        --
        No, you are mistaken. I've always had this sig.
  • (Score: 2) by Alfred on Wednesday July 23 2014, @09:40PM

    by Alfred (4006) on Wednesday July 23 2014, @09:40PM (#72990) Journal

    Obviously these are for use in server farms. Google wants to cut some % off of their current electric bills and it is worth more than a million to them.

    But I don't see where in the electric infrastructure 450 VDC comes in to play. Big power delivery is always AC, typically 480V 3 Phase with some places like industrial plants getting higher voltages even up to 60kV. For this to fit in the current scheme the AC delivery would have to be converted to DC which is just a waste. Unless they are going to use batteries but 450 V is a lot of batteries. Maybe they are wanting to do some peak shaving? I don't think solar is good enough to bother powering a server with it. There are some HVDC delivery lines but that is tens of kV and wouldn't apply here.

    Any ideas on where this fits because I'm not seeing it.

    • (Score: 3, Informative) by BsAtHome on Wednesday July 23 2014, @10:08PM

      by BsAtHome (889) on Wednesday July 23 2014, @10:08PM (#72999)

      450V DC is (relatively) easy to convert to 240V AC (i.e. 340V peak) because you can do H-bridge step-down conversion. That saves you a very expensive transformer and the associated loss. The numbers, including the series resistor (representing feed-line resistance) are apparently chosen so that you still have head-room in a step-down design.

      The efficiency of >95% would probably not be achievable with a transformer at 60Hz and 2KVA of power. Such a transformer alone will set you back 2..5% on a good day. Line-synchronization is also a pain because the actual power-transfer is not distributed evenly throughout the period, but concentrated on the peaks. That makes the design much harder and stability also harder to control.

      • (Score: 2) by DECbot on Wednesday July 23 2014, @10:23PM

        by DECbot (832) on Wednesday July 23 2014, @10:23PM (#73010) Journal

        Today's inverter based welders will rectify the incoming AC power and then use IGBTs or the like to create a higher frequency AC square wave to send though a smaller (fist-sized) transformer. Once the voltage is stepped down, it's then rectified back to DC for welding. The same can be done here to achieve better efficiencies and cheaper components.

        Higher frequencies == smaller transformers

        --
        cats~$ sudo chown -R us /home/base
        • (Score: 3, Informative) by BsAtHome on Wednesday July 23 2014, @10:45PM

          by BsAtHome (889) on Wednesday July 23 2014, @10:45PM (#73017)

          True, but higher frequencies result in higher core-losses, although ferrite is better than iron. You still have (generally) about 2..5% loss in copper+core. Another problem, especially with very high currents, is that the coil-to-core coupling can suffer, resulting in wasted magnetic energy. Especially a problem with high frequencies combined with high currents because the wire will expel the current in the wire-core and therefore reduce the effective coupling.

          It is always a balance and a lot of compromise. Pulling too much at one end will often cause a fail at another end.

          It is quite rare to see a PSU (any kind) that has an effective conversion that is greater than 90% if you start to count all the losses (including wiring). Most numbers you see are measured under very specific and not always realistic conditions (just like the mileage numbers for our car).

      • (Score: 2) by evilviper on Wednesday July 23 2014, @11:05PM

        by evilviper (1760) on Wednesday July 23 2014, @11:05PM (#73026) Homepage Journal

        I was thinking much the same thing. Square wave AC output would be easy, low-loss and cheap to do, and would give a 1.0 power-factor for connected SPSes. However, Google did specify a very strict THD figure in the requirements, right along-side their minimum power factor, so I'm left scratching my head at the seeming contradictory requirements, and don't really understand what kind of (realistic) device they really want.

        --
        Hydrogen cyanide is a delicious and necessary part of the human diet.
        • (Score: 2) by BsAtHome on Wednesday July 23 2014, @11:58PM

          by BsAtHome (889) on Wednesday July 23 2014, @11:58PM (#73045)

          You can simply generate the sine in a class-D style with added LC filter. There are advantages to use a pure sine-wave as it has no harmonics and therefore few EMI problems. The power factor is important because many loads are (slightly) capacitive or inductive. You need to be able to handle that properly.

          • (Score: 1) by anubi on Thursday July 24 2014, @01:23AM

            by anubi (2828) on Thursday July 24 2014, @01:23AM (#73070) Journal

            I was looking at that sine wave part too. I have been tinkering with inverters for solar power for some time now... using Don Lancaster's "magic sinewaves".

            A problem I have run up against is on single phase, I have high current draws on the peaks, but none during the valleys. Then I use 60 Hz resonators. They are big and heavy, but do act as an "energy flywheel", taking a steady source of energy then outputting it as AC. I am not happy with the design. Its big, heavy, and ( like the old SOLA regulators ), not all that efficient.

            I am currently experimenting around with three-phase inverters, just so I have two phases to dump the energy to while each of the phases goes through null. It is very hard for me to design around the pulsing nature of single phase, but three phase gives me a much smoother power draw from the DC rail if I can keep the AC loads balanced.

            I am doing roughly what they want, but I *have* to be 3 phase ( at 60 Hz delivered, but its synthesized at around 100 KHz ).

            On a practical side, I note I can have various panels delivering various energies. Tying them in series is not a good idea. The strong panels can reverse a weak or shaded panel. A shaded or weak panel will also limit the system. Also I do not like several hundred volts at tens of amps DC floating around the roof where not only a short circuit can easily start a fire, its also a single point of failure. I much prefer distributed inversion, where I have a microinverter for each panel doing what it can to synchronize to and feed the main high voltage DC bus that feeds the three phase inverter. These inverters are typical 10V to 60 VDC input and use maximal power point tracking to maximize power transfer. Current-Fed resonant Royer topology. These are kinda heavy as they have a large input inductor for energy storage. I prefer inductors for their reliability. All upconvert to the positive DC rail of about 330 volts DC. This goes to a "Magic Sinewave" three-phase switch. The output is 120 volt 3-phase ( 208V line-to-line ) RMS. Note the 330 volts represents the peak-to-peak voltage. Batteries can be charged from the 330 volt bus, or discharged back through the inverters to regenerate the 330 volts which becomes switched into 120V three-phase.

            What this boils down to is any heavy user of power I have ( refrigerator, air conditioner, washing machine, battery charger, 12 volt DC secondary bus driver ) has to be powered three phase so I can keep the load balanced. I want to put a 12-volt "lighting and utility" bus up so that all the electrical outlets will put out 12-volt power for lighting and miscellaneous stuff like laptops. I figure I would not use more than 500 watts at 12 volts for all lighting and miscellaneous combined. I would just use the standard automotive type devices in lieu of the 120 volt devices for most things. Fusing is a lot simpler too. PTC thermistors.

            This has been a work in progress for me for several years now. I am not done yet. I am now developing the Arduino logging systems which will control it all. My tactic is not to use one big switch, rather I will be using hundreds of tiny switches ( dirt cheap on AliExpress ) which I replicate on PCB plug boards. Failures only reduce capacity of the system. It should work with only one card functioning, albeit maybe all I can do is charge a flashlight with what I can put on my local grid.

            Anyway, for me its a toy I have been working on that when I can build my dream house, that will go in it.

            I have had a lot of time to ponder on this kind of thing. There sure are a lot of details to make something like this work, but it looks like it will be very replicatable and scalable if done right.

            --
            "Prove all things; hold fast that which is good." [KJV: I Thessalonians 5:21]
    • (Score: 1) by Urlax on Wednesday July 23 2014, @10:10PM

      by Urlax (3027) on Wednesday July 23 2014, @10:10PM (#73001)

      so this is a 450VDC->240VAC inverter.

      My best guess is that they want to use this to transport power from the battery backup to the rack itself.

      If you have a battery backup that is kept in shape with solar, it's quite easy to stay at 48VDC and go from there inside the building to the rack itself. at large loads, this requires enormous amounts of copper/aluminium, and going back to AC for the (relatively) short distance is quite wastefull.

      so they want a lot of small inverters to get to AC, using COTS PSU's in their servers to go back to DC.

    • (Score: 2) by DECbot on Wednesday July 23 2014, @10:14PM

      by DECbot (832) on Wednesday July 23 2014, @10:14PM (#73004) Journal

      Europe is 230VAC single phase for home and 460VAC 3-phase for industrial. Not too sure where the 450VDC requirement is coming from, but I bet Google has a particular power source in mind. Perhaps they are tying a large solar array together and the total volts was calculated to 450V, or they have a DC generator, or other DC power source and are looking for a better inverter.

      To my knowledge, the server farms are all DC input power and DC to DC power supplies. There's no sense in inverting DC power from the UPS to only rectify it back to DC in the server.

      --
      cats~$ sudo chown -R us /home/base
      • (Score: 3, Interesting) by kaszz on Thursday July 24 2014, @12:11AM

        by kaszz (4211) on Thursday July 24 2014, @12:11AM (#73053) Journal

        Don't you mean 400 V AC 3-phase? very common in Europe.

        The reason to feed the servers 230 V AC is to be able to equip them with off-the-shelf power supply parts. Otoh any serious power reduction scheme would eliminate this DC/AC--AC/DC mess and go DC/DC directly. Perhaps even pump that DC distribution up to 1000 V or so as this will reduce ohmic power losses with square to voltage.

        • (Score: 2) by DECbot on Thursday July 24 2014, @03:51AM

          by DECbot (832) on Thursday July 24 2014, @03:51AM (#73108) Journal

          Do'h! Yes, I meant 400V. Our equipment is manufactured in Europe and is rated for 460V +/- 10% for the US market. Good thing too, for the past few months our '480V' has been more like 497V.

          Anyway, it seems that I got some of those numbers jumbled in my head when I posted last.

          --
          cats~$ sudo chown -R us /home/base
    • (Score: 0) by Anonymous Coward on Thursday July 24 2014, @03:03AM

      by Anonymous Coward on Thursday July 24 2014, @03:03AM (#73093)

      Big power delivery is always AC, typically 480V 3 Phase

      It constantly amazes me there are still people who see a building and don't immediately think "That would be a great place to put a bunch of solar panels" [thegreenage.co.uk] and who instead think "Y'know, if we just let enough sunlight hit the roof of our building, it will get so hot that we'll have to pump out that heat--and we could pay some 3rd party to provide us with the electricity to do that".

      .
      I don't think solar is good enough

      Welcome to the 21st Century, Mr. Van Winkle. How was your nap?

      -- gewg_

  • (Score: 1) by arslan on Wednesday July 23 2014, @10:18PM

    by arslan (3462) on Wednesday July 23 2014, @10:18PM (#73007)

    So if some does indeed come up with something like this, could they make more money patenting it and selling it to the rest of the world? If so why would they even bother turning it over to google for $1m... $1m doesn't get you far nowadays.

    • (Score: 1) by tftp on Wednesday July 23 2014, @10:33PM

      by tftp (806) on Wednesday July 23 2014, @10:33PM (#73013) Homepage

      Per the summary, Google does not want you to "turn it over" to them. They are happy just buying the finished inverters from you. However their intent is to request an R&D done for free - say, each OEM does R&D for $500K (which is peanuts today,) and 100 such OEMs invest 25 million combined. But only one OEM gets paid by Google, and the rest are eating the costs. Per my understanding, very few OEMs are set up to bid on contracts. The reason is that unlike Phase 1 SBIRs, this is not a mere paper research - this is some serious engineering, with prototypes built and brought up and perhaps respinned. Note also that Google requires compliance with a bunch of EMC standards, which requires testing - and that is not free, and that is not going to be a waltz-in-get-approved method; not with currents that flow within this inverter. High efficiency will require fast switches, and they will generate harmonics higher in the spectrum - so you have to suppress emissions, both conducted and radiated, in a very wide range of frequencies.

      • (Score: 0) by Anonymous Coward on Wednesday July 23 2014, @10:45PM

        by Anonymous Coward on Wednesday July 23 2014, @10:45PM (#73016)

        > Per the summary, Google does not want you to "turn it over" to them.
        > They are happy just buying the finished inverters from you.

        Neither the summary nor the web site itself say they plan to buy any inverters.
        They might, but they aren't even talking about that.
        What the website makes clear is that this is a project in the mold of the X-Prize.

        The guy who runs X-Prize does it for himself. Diamandas makes a buttload of incidental income from all his "non-profit" endeavors. Not that google needs that kind of chump-change. Just that, like Peter Diamandas, we shouldn't assume that Google has any intent of using the end product themselves.

    • (Score: 1) by epitaxial on Thursday July 24 2014, @01:15AM

      by epitaxial (3165) on Thursday July 24 2014, @01:15AM (#73067)

      Patenting this requires full disclosure of your design. Then China can start cranking out cheap clones with impunity.

      • (Score: 0) by Anonymous Coward on Thursday July 24 2014, @03:37AM

        by Anonymous Coward on Thursday July 24 2014, @03:37AM (#73105)

        I hate to even think about patenting anything I do. I would just as soon try to carry a load of beef past a pack of hungry wolves. Anyone who would want to copy what I have done is most likely far bigger than I am and their lawyers earn more in five minutes than I would earn all year. I do not think I could defend myself against that.

        The way I see it, patent law is not in place to protect the little guy, rather its to give the big guy the tools he needs to keep competition at bay. "Free Enterprise" sounds good, but its hard to practice in a Patented environment. Trying to be productive in the Land of Patents is like trying to sell drugs on someone else's "turf". That is just asking for a fight. If you are going to build something, have it built in China. Ship it worldwide. China is getting strong enough to protect itself from the patent trolls.

        Americans don't need jobs anyway. We have well funded entitlement programs to take care of that. Unlike the rest of the world, we can simply print money over here to pay all of our bills. Don't believe me? Watch Congress!

        I am going to post this troll anonymously, but I hope it made you think.

      • (Score: 2) by Fnord666 on Thursday July 24 2014, @04:11AM

        by Fnord666 (652) on Thursday July 24 2014, @04:11AM (#73118) Homepage

        Patenting this requires full disclosure of your design. Then China can start cranking out cheap clones with impunity.

        But that's why we have a patent system. It protects US inventions from having this sort of thing happen to them. Really.

      • (Score: 0) by Anonymous Coward on Thursday July 24 2014, @04:53AM

        by Anonymous Coward on Thursday July 24 2014, @04:53AM (#73133)

        Don Lancaster has already been mentioned in this thread.
        He's a real freak when it comes to "intellectual property".
        He advises everyone NOT to patent anything, saying it's just a way to piss away money on lawyers and gov't application processes.
        He's so paranoid that his website is done as PDF pages.
        A real freak.

        .
        Now, how are you going to keep the Chinese from using chemicals to dissolve the potting on your gizmo and reverse-engineering it.
        Hasn't anybody seen "A Piece of the Action"? (ST-TOS)

        -- gewg_

  • (Score: 1) by anubi on Thursday July 24 2014, @06:49AM

    by anubi (2828) on Thursday July 24 2014, @06:49AM (#73147) Journal

    about those three phase sinewaves... I am basing it on this... delta friendly magic sinewaves ( small .pdf ) [tinaja.com] for precisely the reasons Don speaks of. Don is definitely the guru here... I am just following the path he is blazing. That link is an overview. The implementation can get pretty sophisticated.

    I find storing the patterns in a 27C128 EPROM and simply marching them out with a counter to work pretty good. Because all the signals are pulsing rapidly, it simplifies driving the power MOSFETS, as I can get by with N-Channel MOSFETS/IGBT's for all switches, using capacitive coupling to drive the high-side switches. The EPROM is 8 bits wide, I use six, having the seventh and eighth bit available for putting in trigger codes to help mark a certain portion of the cycle for closer observation.

    I have three high side switches and three low side switches. Each phase gets one high and one low side switch.

    I am trying to find patterns where at all times I have one and only one high side and one low side switch on at any time - kinda acting like a three-legged differential amplifier. The idea is to keep the input current flow absolutely steady while smoothly switching the power into three phase AC.

    Hat tip to Nikola Tesla, who came up with this three phase AC thingie in the first place...

    --
    "Prove all things; hold fast that which is good." [KJV: I Thessalonians 5:21]
  • (Score: 0) by Anonymous Coward on Thursday July 24 2014, @09:37AM

    by Anonymous Coward on Thursday July 24 2014, @09:37AM (#73189)

    On first look, with the 450VDC input, the converter could be single stage with a full bridge and a low pass LC filter for the output. The sinewave can be generated with PWM modulation of the diagonally opposed top/bottom switches (each half period, one acts as a 'buck' switch and the other as an active rectifier, the next half period the other diagonal is used). Like a class-D amplifier really. I can't immediately see why this topology wouldn't work within the given constraints, and if it really can't, what else to use?

  • (Score: 2) by VLM on Thursday July 24 2014, @12:22PM

    by VLM (445) Subscriber Badge on Thursday July 24 2014, @12:22PM (#73228)

    I have an almost uncontrollable desire to enter a motor-generator as a gag entry.

    Its pretty close to what it sounds like, a plain old motor, a coupling, and an alternator or generator. On a chassis. Of the appropriate phases and voltages. Add circuit breakers on the inputs and outputs, and add a flywheel if you want.

    They're not all that bad of an engineering design, especially if you have to deal with momentary spikes and the like.

    Kind of a big "F you" to them. You want an EE to work for you, well, hire or contract one, not offer some idiotic contest.