Design a better inverter- win a million bucks!

[qno]

Maybe less switching.

Check out Don Lancaster's Magic Sinewave pages.

http://www.tinaja.com/magsn01.shtml

[xygor]

Maybe less switching.

Check out Don Lancaster's Magic Sinewave pages.

Danm! You've revealed my secret.

That technique requires choosing from canned waveforms to do voltage regulation, so the output voltage moves in steps.

[qno]

Maybe less switching.

Check out Don Lancaster's Magic Sinewave pages.

Danm! You've revealed my secret.

That technique requires choosing from canned waveforms to do voltage regulation, so the output voltage moves in steps.

Did you ever implement succesfully one of Dons algorithms?
I have been looking at it for years but still was not able to formulate the thing for a PIC.

[jucole]

Maybe less switching.

Check out Don Lancaster's Magic Sinewave pages.

http://www.tinaja.com/magsn01.shtml

In the rules it said something about having to have an output frequency of between 59.7 - 60.3 hz;  so would that magic sine thing be no good in this case?

[zapta]

Do the rules say how long it needs to operate?

[jucole]

This document gives a lot of info.

https://www.littleboxchallenge.com/pdf/LBC-InverterRequirements.pdf

[electronic_eel]

The problem I see with household DC power distribution is switching: If you use more than about 24V, you'll get long arcs when switching off. All kinds of switches and relays must therefore contain arc breakers. With AC this is not neccessary as the arc goes out at the next zero cross. Look at the prices and sizes for DC relays and switches which are designed for these voltages and some amps. This also was one of the main reasons why the car industry stopped their plan to switch from 12V to 48V some years ago.

If you stick to 24V and less, you need thick wires and have considerable cable losses.

So there are technical reasons for AC beyond the installed basis.

[rob77]

This document gives a lot of info.

https://www.littleboxchallenge.com/pdf/LBC-InverterRequirements.pdf

the smart way of outsourcing the engineering job it will cost them 1meg , but they will get a lot's of ideas and probably even the product in return

[diyaudio]

I will eat my shoes if I see a dspPIC or some sort of "dspish" controller in the final winning project, would be a giant step for innovation.

[David Hess]

From the contest site:

One promising set of new technologies which may allow for the achievement of higher power densities are wide bandgap (WBG) semiconductors, such as Gallium Nitride (GaN) and Silicon Carbide (SiC).

Maybe they think we might be able use these to get higher switching rates. I can't see how else these semi conductors would help power densities.

If you could achieve a switching rate of 10x does anyone have a rough calc on what power densities you could then achieve?

Switching, conduction, and drive losses will be lower at a given frequency so power total power dissipation will be lower.  The operation frequency may be selected to balance power dissipation between various elements.

[nctnico]

And the inverter would still need to turn DC into AC. IMHO a lot depends on the magnetics. I wouldn't be surprised if someone comes up with a very smart setup for the inductors which has been patented 30 years ago.

[David Hess]

And the inverter would still need to turn DC into AC. IMHO a lot depends on the magnetics. I wouldn't be surprised if someone comes up with a very smart setup for the inductors which has been patented 30 years ago.

If it was patented 30 years ago then it will have expired.

The two common tricks I have seen for high density and low loss magnetics are using copper tape instead of wire and enclosing traces printed onto the circuit board with a planar E core.

[xygor]

Did you ever implement succesfully one of Dons algorithms?
I have been looking at it for years but still was not able to formulate the thing for a PIC.

No, I have not tried it yet.  I have kept it in mind for years too.

In the rules it said something about having to have an output frequency of between 59.7 - 60.3 hz;  so would that magic sine thing be no good in this case?

There is nothing that would preclude using it to produce that frequency.

And the inverter would still need to turn DC into AC. IMHO a lot depends on the magnetics. I wouldn't be surprised if someone comes up with a very smart setup for the inductors which has been patented 30 years ago.

I don't think you need any magnetics for the power path conversion.  Unless you mean for EMI filtering.

Do the rules say how long it needs to operate?

Yes. 100 hours.

[Marco]

I don't think you need any magnetics for the power path conversion.

Can't just throw a PWM modified sine wave at the load since it can be partly capacitive which is going to generate losses ... if you want to do it with only say a full bridge worth of MOSFETs you'll need magnetics.

I guess you could do it without magnetics with switched capacitors, but you'd need a much larger amount of switches (and a lot of capacitors as well).

PS. the ripple spec would be hard to meet with switched capacitors though ... not impossible, but is going to increase the amount of switches/capacitors even more.

PPS. it doesn't make much sense to have the ripple requirement in my opinion, coping with ripple belongs with the preceding galvanically isolated DC DC converter which will generally be necessary (I wouldn't want solar panels with mains voltage on their ground connections for instance).

[xygor]

I don't think you need any magnetics for the power path conversion.

Can't just throw a PWM modified sine wave at the load since it can be partly capacitive which is going to generate losses ... if you want to do it with only say a full bridge worth of MOSFETs you'll need magnetics.

I guess you could do it without magnetics with switched capacitors, but you'd need a much larger amount of switches (and a lot of capacitors as well).

Yes, it will need an LC output filter, but that's to meet the EMI and THD requirements.

the smart way of outsourcing the engineering job it will cost them 1meg , but they will get a lot's of ideas and probably even the product in return

They don't get the design for the $1M.

[houdini]

If your submitting it for evaluation they get the design.

[xygor]

If your submitting it for evaluation they get the design.

They don't look in the box.

[xygor]

It does say they can reproduce it, but I don't see any requirement to submit any manufacturing data.  On the face of it it looks like Magic Sine concept is out since that is owned by Don and the entrant would not be able to grant a license to reproduce it.


12. INTELLECTUAL PROPERTY RIGHTS: As between Google and the Entrant, the Entrant retains 
ownership of all intellectual and industrial property rights (including moral rights) in and to the Device. As a 
condition of entry, Entrant grants Google, its subsidiaries, agents and partner companies, a perpetual, 
irrevocable, worldwide, royalty­free, and non­exclusive license to use, reproduce, adapt, modify, publish, 
distribute, publicly perform, create a derivative work from, and publicly display the Device (1) for the 
purposes of allowing Google and the Judges to evaluate and test the Device for purposes of the Contest, and 
(2) in connection with advertising and promotion via communication to the public or other groups, including, 
but not limited to, the right to make screenshots, animations and device clips available for promotional 
purposes. As stated elsewhere in these Rules, the Sponsor reserves the right to make public any technical 
approach document submitted to the competition.   

[NiHaoMike]

Most of the space in an inverter is for voltage step up. Take that part out and the inverter can be a lot smaller. I have seen a 1kW inverter just a little bigger than a deck of cards used in a hybrid bicycle.

[Legit-Design]

Most of the space in an inverter is for voltage step up. Take that part out and the inverter can be a lot smaller. I have seen a 1kW inverter just a little bigger than a deck of cards used in a hybrid bicycle.

It was a brushless DC motor controller? 3 phases and operates the motor at battery voltage. So if we have high voltage DC in houses now all old appliances that require AC needs one of those bricks between it and the plug. Might be a bit hassle to explain that to some people.... "ah but this has worked before, why doesn't it work anymore?"

A variable frequency drive for a motor should be able to operate with DC voltage. Some airconditioners already include that, to change the speed instead of crude on-off control.

[tszaboo]

In the rules it said something about having to have an output frequency of between 59.7 - 60.3 hz;  so would that magic sine thing be no good in this case?

240 +/12 V AC
60 +/0.3Hz
So they managed to make a specification, which isn't used anywhere in the world.

[Artlav]

220V at 60Hz?
South Korea, Philippines, Peru and a bunch of island states.
Not exactly "nowhere in the world", but a very peculiar choice never the less.

I was under impression that the big problem was the 12-48VDC to 320VDC step-up part, not the 320VDC to 220VAC part.
Why would they focus on the last part?

Or is it about making higher efficiency pure sine wave inverter?

[Fsck]

They are asking for power density, which is about 10 times the currently available inverters. I'm sure that the challange is possible, maybe it would be worth to make an asic for the control. But.
It wouldnt be cheap, it wouldnt be ready in time, and most likely it would be loud as hell. You still need to dissipate a lot of power and that needs cooling.

basically what I was thinking. the output quality specs are brutal for that density and size, but it looks like 450 VDC to 240VAC which is not really what I was initially thinking would be a 1M$ campaign.
then again, I'm not up to date on power electronics. maybe if you gave me like 5 years to throw everything at the wall to see what sticks.

[Bored@Work]

Could it be that Google simply needs better power inverters for their data centers? That this is not about bringing small inverters to the homeowner masses?

[ejeffrey]

220V at 60Hz?
South Korea, Philippines, Peru and a bunch of island states.
Not exactly "nowhere in the world", but a very peculiar choice never the less.

Also the US.  US residential supplies are 240 VAC.  It is a split phase configuration and normal outlets only use one phase to get 120 VAC, but the  backup generators and PV inverters normally produce 240 V.  High power outlets used for things like electric ovens, clothes dryers, and electric cars use one of the various 240 VAC outlets.

Industrial settings don't have split phase transformers they normally have 3 phase feeds.  In that case, you normally get the 208 V phase-phase voltage on the same outlets used for residential 240V.