Sharp Announces Sales of DC Powered Air Conditioner, Other Products To Follow

AmiMoJo writes: Sharp has announced that sales of DC powered air conditioners will begin by the end of the year. Most appliances use the standard AC electricity supply in homes, but as solar panels become more common switching to DC can save on conversion losses. Solar panels produce DC, which is then typically converted to AC before being fed into the house's wiring, and then converted back to DC again by appliances. Sharp has announced that it intends to produce a range of DC powered appliances for home use.

Comments

DC power?

[Joce640k]

Nikola Tesla is turning in his grave.

Re:DC power?

[K. S. Kyosuke]

Let's attach him to a DC generator, then. He can make himself useful once again. ;)

Re:DC power?

[MichaelSmith]

At 60Hz?

Single phase obviously.

Re:DC power?

[fuzzyfuzzyfungus]

Did he have anything against DC period, or just take the position that it wasn't so hot for transmission systems, especially since high efficiency DC-DC converters weren't exactly off the shelf items at the time?

Re:DC power?

[AmiMoJo]

That's it exactly. Back then things like frequency conversion and DC level switching had to be done mechanically. To change frequency you ran a motor that drove a gear that drove a generator at the new frequency, and did something similar for switching DC voltage levels.

Nowadays high voltage DC is used widely for transmission. Everything is solid state and highly efficient.

Re:DC power?

[cheesybagel]

Both.

Re:DC power?

[Big Hairy Ian]

Nikola Tesla is turning in his grave.

Spinning my dear boy spinning!

Re:DC power?

[Spazmania]

Basically, Sharp is eliminating the rectifier circuit from one of their existing products. Sharp currently sells it as an 'Inverter Air Conditioner".

Unlike most air conditioners, inverter air conditioners are always-on. The inverter varies the -frequency- of alternating current sine wave in order to change the cooling output of the air conditioner. It continuously outputs just enough cooling to maintain a steady temperature in the room.

To do this, the A/C converts the incoming wall power to DC and then back to variable frequency AC. Eliminating the initial AC to DC conversion here makes good sense.

Re:DC power?

[Mr D from 63]

High Voltage DC transmission makes sense in some applications. Its best as a single point to point solution over a fairly long distance, as line losses are minimized. High voltage DC switching and DC to AC conversion equipment is very expensive compared to AC, and typically has a shorter lifecycle, so you don't really want to have a lot of DC switchyards.

Re: DC power?

[Anonymous Coward]

His main interest was in using high frequency for wireless power transmission. His lack of E-M understanding caused him to waste a lot of his talent pursuing a failing approach.

Re:DC power?

[Anonymous Coward]

Came here to see this joke... Slashdot is still alive!

Re:DC power?

[rmdingler]

Not only Sharp, but nearly every major ductless split air conditioner runs the inside air handler with DC voltage from an inverter in the outdoor condenser.

This is a clever, but predictable evolution in design production.

Re:DC power?

[cayenne8]

Unlike most air conditioners, inverter air conditioners are always-on.

Hmm...apparently you're not familiar with "AC" Air Conditioners in the New Orleans area.

My Air Conditioner comes on basically late April and does not shut off till about early November.

:)

Frankly, I don't know how people lived down here in the old days without A/C....not to mention looking at the old pictures from the 1800's and so where men have full suits on and women with 14 layers of clothing on, and deodorant not really being prevalent either...UGH!!

But yes...my A/C really never shuts off all summer, which down here lasts for about 6mos of the year or so, give or take.

Re:DC power?

[Anonymous Coward]

1) They built houses that were passively cooled
2) Linen. Try it.

Re:DC power?

[Gryle]

Early Southern architecture had passive cooling systems, like double-hung windows, high ceilings, and constructing doorways and windows in such a way as to generate breezes as the air heated up during the day. The Spanish were particularly good at things like this.

Folks also had a different daily routine. They were up before the sun, worked until about lunch, had a big meal and a nap during the hottest part of the day, and then worked until late in the evening. The Spanish siesta is a good example of this, but the Italians have a similar concept. In modern days I see a lot of construction workers doing this too, particularly on road construction. The job site will be empty during the afternoon and work begins in the evening and lasts all night. (Some of this is to keep from interfering with day-to-day traffic patterns too.)

Re:DC power?

[AmiMoJo]

The other part of this is that they are running it from solar PV. The air con must have some intelligence to handle a variable supply. It's a great idea IMHO - as the sun warms the house, the air con has more clean energy to keep it cool.

Re:DC power?

[thinkwaitfast]

Can't be that clever; I had the idea 20 years ago when I first learned of bang-bang controllers

Re:DC power?

[macs4all]

Nikola Tesla is turning in his grave.

...at 1800 RPM

Re:DC power?

[macs4all]

High Voltage DC transmission makes sense in some applications. Its best as a single point to point solution over a fairly long distance, as line losses are minimized. High voltage DC switching and DC to AC conversion equipment is very expensive compared to AC, and typically has a shorter lifecycle, so you don't really want to have a lot of DC switchyards.

How are line losses minimized for DC over AC, given the same "RMS" voltage (yes, I know the term means nothing in DC) and Current?

It seems to me that the only application of HV DC would be to stack up a bunch of PV panels in Series, and then buss the result together before sending it to a HV-input Inverter.

But, unless the EPA revised the laws of physics, it seems like AC transmission will pretty much always "win" over DC, especially over distance.

Or is it because maybe HV DC has less coronal discharge loss over AC?

Re: DC power?

[macs4all]

His main interest was in using high frequency for wireless power transmission. His lack of E-M understanding caused him to waste a lot of his talent pursuing a failing approach.

Oh, I think Tesla understood Electromagnetic principles a bit better than most...

Re:DC power?

[macs4all]

To do this, the A/C converts the incoming wall power to DC and then back to variable frequency AC. Eliminating the initial AC to DC conversion here makes good sense.

Are you sure it's variable frequency AC, and not variable pulse-width AC? Or are they just using a Vector Drive to ultimately power a 3 phase AC motor?

So, in essence, you're building a big ol' SMPS (Switch Mode Power Supply), like in my Panasonic "Inverter" Microwave Oven. So Sharp is just eliminating the "DC Link" part of the SMPS/Motor Drive. Yawn. However, since that involves some fairly stout rectifier diodes and monster capacitors (that eventually wear out from the ripple currents), maybe it's not so dumb afterall.

Re:DC power?

[barc0001]

If your source of generation is DC like solar panels, then you're inducing conversion loss. AC is far more efficient at transmitting power long distances than DC is, but in this case you're only moving it a few meters so the benefits in efficiency are worth doing. Pretty sure Nikola would have been fine with this.

Re:DC power?

[Anonymous Coward]

Taking the following two formulas:
P = U * I (power = voltage * current)
U = I * R (voltage = current * resistance)

Follows:
P = I * I * R

The power lost in a transmission line is based on the resistance of the power line (which is pretty much constant) and the current flowing through it. Reducing the current will reduce power loss in a transmission line. To transmit the same power you will need to increase the voltage to compensate.

With an AC signal the ratio off transmission loss vs delivered power changed during each period. With DC you maintain the same high ratio.

Re:DC power?

[Mr D from 63]

How are line losses minimized for DC over AC, given the same "RMS" voltage (yes, I know the term means nothing in DC) and Current?

In the simplest terms, DC current losses are only result from the line resistance. AC losses are due to the line resistance plus reactive losses. Reactive losses are basically the capacitive and inductive losses that result from the constantly changing current and voltage. Various factors are at play, but you wind up with the voltage and current sine wave getting out of sync with each other. If that doesn't seem intuitive, imagine something at work trying to bring those two back into synchronization... that would be the reactive losses.

The higher the voltage, the less line resistance becomes a factor, but that does not apply equally to reactive losses.

Re:DC power?

[Anonymous Coward]

With AC power, your total line resistance is a combination of resistance, inductance, and capacitance. With DC you only have resistance. So you end up having a little less effective resistance when using DC. Because the converter stations on both ends of the line are expensive, this only pays off on long transmission lines. You also only need to run two conductors (positive and negative) for DC versus three for three phase AC power.

Re:DC power?

[zlives]

yes but windows 10 will render everything usless... o wait forgot to take my M$ shill hat off.

Re:DC power?

[david_thornley]

It isn't just line losses, it's also line capacity.

AC current isn't on steadily, and so the RMS voltage is about 70% of the maximum voltage. DC is steady, and so it can make full use of the voltage capacity of the line.

Another problem DC faced was switching. It has to be possible to turn connections on and off. It's harder when high voltage is going through, and AC has momentary zero-voltage points.

Most of what I know about electrical power transmission is from a job I had about fifteen years ago, so I don't know what switching technology is available.

Re:DC power?

[rtb61]

Yep that windows 10 marketing hype and B$ reviews are certainly getting more than a little over the top. That don't seem to be doing that well though and they are running into a real problem. As they push more the windows 10 hype is starting to becoming more annoying and putting people off, so they try harder to promote hype and instead of winning converts they are just becoming more annoying. I think their key market is baulking and looks to be waiting more than a quarter for the B$ to die down so that reality can start to leak through prior to wasting the effort and frustration on yet another unwanted upgrade.

DC makes sense with your own battery capacity and only using minimal AC from the mains as emergency charge up, if you solar capacity or battery capacity was too small. So likely houses will go all DC as it is much safer and reduces capital cost (all appliances without transformers), with AC only going to the battery outside. Mains power could of course become hugely expensive with so little current flowing to pay for that infrastructure. Interesting problem though of medium and high density dwellings with insufficient area for power generation, not to bad if they are all together but really bad when they are scattered amongst low density dwellings with sufficient area to generate their own electricity.

Re:DC power?

[Anonymous Coward]

With DC you don't have to factor in the inductive/capacitive effects of the line as much.

Line resistance is still a concern, naturally.

If you don't know what I am talking about, the capacitive loss is due to the fact that the wire and the ground make a (weak) capacitor, over long distances, this will leak the AC current. Also, radio antennae are basically just a long wire that you run AC into and it pumps power into the air in the form of radio waves; the difference between an antenna and a transmission line is the the power loss is desired in the case of a radio antenna.

Re:DC power?

[macs4all]

If that doesn't seem intuitive, imagine something at work trying to bring those two back into synchronization.

No, I get it. I just wasn't thinking. Thanks!

Re:DC power?

[DarkOx]

my A/C really never shuts off all summer,

That sounds to me like you have a badly undersized unit. Its also possible that something isn't working right maybe the A-Coil is caked with crud and inst transferring heat into the loop efficiently?

Either way that sounds terribly inefficient and is probably costing your a fortune to run. Its probably worth talking to an HVAC guy.

Re:DC power?

[maestroX]

At 60Hz? Single phase obviously.

aerosun-panel-powered generator?

Re:DC power?

[cayenne8]

1) They built houses that were passively cooled

2) Linen. Try it.

Well, I've been in a LOT of #1 type houses here (they still have 100's of year old homes here, and it still ain't cool enough.

And even wearing linen...single layers are still too hot, much less the many layers they seemed to have worn then.

I"m just guessing they were a LOT tougher than we are now.

:)

Re:DC power?

[cayenne8]

That sounds to me like you have a badly undersized unit. Its also possible that something isn't working right

No, I just had it checked, is a good system, etc.

It is just THAT fucking hot down here, and humid. A lot of the AC time on is to dry the air out in the house.

I like to keep the temp about 74F during the day time I"m here, and about 72F at night to sleep. That's comfy to me.

But it gets HOT here at the end of April usually, and we don't even start to get breaks in the heat and humidity till about mid Oct I think....

Re:DC power?

[swb]

All the HVAC techs I've ever talked to have told me that it's better to have the compressor run continuously than it is to short cycle it.

I had new AC put in about 10 years ago and if cleaning the A coil is something that's supposed to get done, they sure don't do the sheet metal like its something that's supposed to be done.

My A coil failed (I think they used a non-R134a coil) after two years and the guy had to do a lot of sheet metal surgery to get the new coil in.

Re: DC power?

[Anonymous Coward]

He "determined" that E-M waves in the air with earth as a ground plane travelled faster than light. He was very smart (eg, he had an awesomely intuitive understanding of resonance), but also a bit deluded.

Re:DC power?

[Anonymous Coward]

Its zero hz. DC means that it does not alternate. Phase does not apply.

Re:DC power?

[rthille]

I think the problem with your A/C unit is the location. It appears to be located in an unbearable hell-hole :-)

Re:DC power?

[Agripa]

RMS works fine for DC; it is just that for DC, the RMS and average are the same.

Coronal losses are some nonlinear function of voltage which becomes significant during the peaks of the AC voltage cycle at high voltages; lowering the peak to RMS ratio minimizes this for a given voltage.

Re: DC power?

[macs4all]

He "determined" that E-M waves in the air with earth as a ground plane travelled faster than light. He was very smart (eg, he had an awesomely intuitive understanding of resonance), but also a bit deluded.

Well, since he probably "determined" that before Einstein's General (or is it Special?) Theory of Relativity, he might be excused that faux pas.

And I would characterize him as "Brilliant", not just "Very Smart".

And yes, he had a few bats in his belfry; but that seems to often go with "Genius", ya know? It seems that, when a person has the ability to "see beyond" what people think of "reality", they don't just stop at one particular topic. But that peccadillo is certainly not unique to ol' St. Nikola.

Re:DC power?

[geckoFeet]

> deodorant not really being prevalent

Uh, New Orleans was settled by the French. What's the mystery?

Re: DC power?

[smaddox]

The waves probably did travel faster than light. This is a well known property of waveguides below cutoff. Even the group velocity can travel faster than light if the attenuation is large enough. However, energy transfer is limited to the speed of light.

The whole "nothing can travel faster than light" claim is an over generalization. A lot of things travel faster than light, just not matter, energy, or information.

Re: DC power?

[Anonymous Coward]

Whoosh!

Re:DC power?

[cayenne8]

I think the problem with your A/C unit is the location. It appears to be located in an unbearable hell-hole :-)

Yeah, but on the other hand.

Come Oct/Nov....I'll be wearing shorts outside for Thanksgiving and enjoying the weather while much of the US is indoors freezing their asses off already...so, like anything, I guess it is a trade off.

Re:DC power?

[stoatwblr]

Here are two reasons why AC is generally better than DC at consumer level:

1: AC arcs are self-extinguishing every half cycle.

Anecedotally: Back in the days when Sydney Australia had 250VDC distributed in industrial areas as well as AC, a DC-powered lamp in a factory blew (filament failure) and started arcing internally. That arc eroded away the glass stem of the bulb, its base, the ceramic bulbholder and most of the flex leading back to the ceiling rose before someone managed to kill the power.

2: Human safety:

DC is more deadly than AC, for pragmatic reasons: If you get attached to a DC power line for any reason, continual muscle contraction will ensure you cannot free yourself. With AC, the pulsating nature of the beast has a greater chance of making you kick yourself free.

I've known people who've been bitten by high voltage DC, AC and RF feeds. The DC guys came off worse than AC, but RF can cook you.

#1 means that every single piece of switching equipment in a DC circuit needs to be derated over its AC counterpart. Take a look at any switch you can lay your hands on and you'll notice that there are AC and DC ratings. This is down to arc resistance, not current carrying capability.

Re:DC power?

[cwsumner]

Here are two reasons why AC is generally better than DC at consumer level:

1: AC arcs are self-extinguishing every half cycle. ...

2: Human safety: ...

The first is true, but the second is not so much. 60Hz current can scramble the coded signals used by your nervous system, in effect "crashing your network". And 60Hz can cause muscle contraction, it just shakes a bit. I think it is harder to let go when hit by 60Hz. It does depend on frequency, though, higher frequencies do not cause this. I think 400Hz was considered less dangerous, in that way.

It's beside the point,though. The application in question does not require long distance or high voltage. For one house, 24 volts might be fine.

I think the real news is, that they are using the newer high efficiency DC motor technology allowed by electronic motor controls.
Brushes and mechanical commutators were why this was not done long ago.

Re:DC power?

[cwsumner]

Modern power control electronics can produce a smooth sign wave, at a pretty good efficiency. The first harmonic as at like the 8th overtone, which is easily filtered. However, there are motor controllers that don't use sign waves, but form the current to match the motor characteristics. Even better.

Re:DC power?

[rthille]

Or, you could live somewhere nice like Santa Barbara and have both :-)

Re:DC power?

[toddestan]

One of the reasons you want to run the A/C as close to continuously as possible is that you want it also to act as a dehumidifier. If it short cycles it's not going to be able to pull the humidity out, leaving you with a cool, damp house. With that said though, I'd be a bit concerned if it never shut off as I would want to know I have at least some spare capacity.

Why not both?

[aaaaaaargh!]

Wouldn't it be possible to have both in the same appliance?

Re:Why not both?

[K. S. Kyosuke]

Possibly, but maybe it's simply cheaper for the manufacturer to simply sell 95% of the same product plus different power interfaces. I've long thought that whole local power grids would switch to DC eventually anyway. With broader deployment of PV technology (and potentially power-to-gas) and improvements of power electronics, it only makes sense.

Re:Why not both?

[LinuxIsGarbage]

It makes sense, except where industrial users use a ton of 3-phase AC induction motors, which are a simple design. And AC transformers are still a more simple method of switching voltages for distribution.

Re:Why not both?

[K. S. Kyosuke]

Maybe you could always have a mid-level AC grid or something similar (below long-range HVDC and above municipal DC). I assume that fewer large inverters would be better for the overall economy than every small PV installation having their own stuff.

Re:Why not both?

[PhunkySchtuff]

AC has far lower transmission losses over long distances, however with the advent of rooftop solar now people can generate DC closer to where they're using it.
Converting AC to DC and DC to AC will always incur losses, and as AC is far more efficient for transmission, we've historically used AC everywhere (well, except inside just about everything you plug into an electrical outlet, where DC is generally used)

Re:Why not both?

[K. S. Kyosuke]

AC has far lower transmission losses over long distances

Actually, it doesn't.

Re:Why not both?

[fuzzyfuzzyfungus]

Certainly possible; but probably only efficient or sufficiently cheap for some items.

My (admittedly limited) understanding is that motors are one of the parts where you can get a relatively refined and mature design for almost any remotely normal flavor of electrical input; but the design of the motor is going to reflect your choice, and work either inefficiently or not at all from some other input. Since AC units are basically all motor(some fans, a compressor, maybe a pump), that is a problem.

Some devices just don't care very much(resistive heating works for basically anything except voltages too much lower than the system was designed for); and others(pretty much all modern electronics) always use DC internally, so you can re-use almost the entire design; but need to choose the correct PSU. Some hardware makes swapping one after the fact really obnoxious; but at very least it's an easy thing to choose at assembly time.

Re:Why not both?

[heypete]

AC has far lower transmission losses over long distances

Does it? I was always under the impression that AC was used for long-distance transmission because it could be easily stepped up to very high voltages with transformers while efficient DC-to-DC conversion was not possible until relatively recently. For the same power transmitted, resistive losses are lower at higher voltages as power lost to heat goes as I^2*R and lower currents could be used.

However, modern solid-state DC-to-DC converters are extremely efficient, can step DC voltages up to very high voltages and thus benefit from lower resistive losses in transmission. HVDC also benefits from not having to deal with inductive or capacitive losses in the cable.

In short, as far I know the key to minimizing losses in transmission lines is to use high voltages, not because of any inherent advantage of AC.

Re:Why not both?

[AmiMoJo]

It would, and some specialist devices do support that. It's not clear if that is what Sharp intends to offer, or if they are going to be DC only. It might make more sense to do the latter use an inverter that can run in "reverse", i.e. using the mains AC to provide central DC when solar energy is not available.

AC made a lot of sense when solid state switching regulators didn't exist, but nowadays having a central DC supply in a home that also generates its own DC isn't a bad idea.

Re:Why not both?

[Anonymous Coward]

No, AC's just cheaper to transmit.

Re:Why not both?

[Crashmarik]

Sure instead of using synchronous AC motors which are simple and efficient you can switch to much more complex DC digital motors.

Sounds brilliant.

Re:Why not both?

[amalcolm]

The advantage of AC for transmission is that it can be transformed up into the 100kV range, at relatively low current, thus minimizing losses in the resistance of the transmission cables. To do the same with DC would require up/down converters which would require chop->transform->rectify anyway (the transformers might be smaller and more efficient, though)

Re:Why not both?

[amalcolm]

I don't think you'll find solid state DC-to-DC converters that can upconvert to 330kV @ 100A

Re:Why not both?

[Anonymous Coward]

AC has far lower transmission losses over long distances

Actually, it doesn't.

Because not only does it have resistance losses like DC, but also induction losses.

But you can jack up the voltage easily and efficiently with a transformer and high voltages travel distances better.

Re:Why not both?

[K. S. Kyosuke]

And yet, in practice, HVDC is still more efficient than current AC lines in the end, even if still somewhat more expensive at the moment.

Re:Why not both?

[monkeyxpress]

The growing trend for appliances is to move towards using brushless DC motors instead of traditional induction motors. These are more compact, more efficient, and variable speed in nature - which opens up new opportunities in some applications. The main problem is they cost more as they require a controller, unlike induction motors which will run directly off the AC supply.

The neat thing about this product is that it recognises that the DC to AC inverter in a solar power system is basically just a motor controller (a box of power electronics). So by moving that box of electronics into an area where it has an additional benefit, they have offset the inverter cost of a solar installation. This has potentially huge implications for the solar industry as inverter cost is becoming one of the dominate components of a solar power system as panel costs continue to reduce.

It would be interesting to see how the added efficiency of the brushless motors and extra abilities - such as being able to vary compressor output to match solar input - impact the overall economics. There are also new technologies such as linear motor compressors that could continue to tip the economics in favour of solar systems. Either way an interesting development.

Re:Why not both?

[aix tom]

Well, here you have 350 kV / 700 MW thyristor converter. It's easy to find, because it's pretty big. ;-)

http://new.abb.com/systems/hvd...
https://en.wikipedia.org/wiki/...

Re:Why not both?

[amalcolm]

Wow, I stand corrected. I wonder how reliable they are vs. a transformer

Re:Why not both?

[Spazmania]

High AC voltages have induction losses. They don't travel as well as low voltages.

BUT

The goal is to send lots of watts, not lots of amps or volts. Low amperages travel well. High amperages don't travel at all -- they lose most of their energy to heat. Simple transformers (which are basically just coils of wire) can swap amps for volts so that lots of watts can travel a long distance at low amperages.

Re:Why not both?

[Spazmania]

It's called an "inverter" air conditioner. It produces a variable-frequency AC sine wave from the DC voltage. The variable-frequency to the compressor changes the cooling output, so instead of turning the air conditioner on and off as the temperature wanders back and forth across the set point, it varies the frequency to keep the temperature steady.

http://www.acson-international...
https://en.wikipedia.org/wiki/...

It is brilliant.

Sharp already sells these air conditioners. They're just removing the DC rectifier circuit and running directly on DC instead of starting with 50hz or 60hz AC.

Re:Why not both?

[K. S. Kyosuke]

To my knowledge, these converters are still fed from matching AC transformers. Granted, that's mostly because AC is what the whole station is connected to. But since technology keeps marching on, we'll probably see a day when direct LVDC to HVDC converters (for example, for utility-scale PV) will be a commonplace affair.

Re:Why not both?

[thegarbz]

They aren't as reliable but also do not require the same amount of maintenance. Transformers can be pricks of things if they start leaking oil, absorbing moisture etc.

Re:Why not both?

[thegarbz]

DC and AC motors are fundamentally driven in different ways. You either need a really esoteric motor design which will come with it's own set of drawbacks, or you will need to provide conversion in the gear itself at which point why would you bother going to DC at all.

Re:Why not both?

[OzPeter]

Wouldn't it be possible to have both in the same appliance?

Yeah, that AC/DC appliance would be really rocking, and a company that succeeds in doing it really would be a rising power. But I'd be worried about the low end manufactures doing dirty environmental deeds, dirt cheap as well as the build quality of the system overall. After all you don't want to be shaken all night long by your air conditioner - that would really lead you down a highway to hell.

Re:Why not both?

[fuzzyfuzzyfungus]

There are good reasons to be skeptical of complexity(such as basically all software); but at least for motors small enough that solid state switching isn't heroically expensive and/or practically disposable, 'complex' and 'digital' also mean 'has convenient ways to monitor and control the motor's activity in some detail'. Simple is good; but when the complexity is in solid-state parts you get a serious discount and sometimes some cool features.

Re:Why not both?

[fuzzyfuzzyfungus]

Especially if you are dealing with solar-derived DC, which is going to be (reasonably) steady. For some purposes, it doesn't matter that much whether or not the voltage actually crosses zero; but the ripple is essential if you want the magnetics to work correctly. If you have to go to the trouble of switching the DC to get that working, odds are that it's time for a pure DC design.

Re:Why not both?

[Mr D from 63]

HVDC equipment will always be much more expensive due to the high cost of switching and conversion equipment, which must have a much greater interruption/arc reduction capacity.

Re:Why not both?

[Anonymous Coward]

Stop being a dick, he made an assumption that was incorrect, when proven otherwise he adjusted his stance.
If everyone who made assumptions acted like him most of the world problems would be solved.

Re:Why not both?

[Mr D from 63]

Your points are valid, I'll just add that we are not really even talking about a high level of complexity either way. These power/control approaches are nothing new, they are used and proven and quite simple and reliable, they are just being adapted to this application. In some ways having DC input really makes motor control easier and you don't have to deal with current/voltage frequency to begin with, and digital controllers are so easy to implement.

Re:Why not both?

[aix tom]

Another drawback is also that those DC transmission is still pretty expensive compared to AC. So it's mainly used when other factors make it cheaper than AC. Like in that case the long transmission distance, including undersea cables where AC has a lot higher losses.

But the cost might actually be brought down a lot with production volume and new technologies. There were not much real technological breakthroughs in transformer technology in the last ~100 years, but they still come up with new semiconductor technology.

Re:Why not both?

[Anonymous Coward]

i think you are referring to "hybrid inverters". they have a A/C motor driving the compressor/pump. not sure.
however there are "pure" inverter, that have a (pulsed frequency?) DC motor that drives the compressor/pump.
also i think the DC motor uses permanent magnets (brushless).
i suppose these are not so common and expensive because building a DC motor is more complicated/expensive?

Re:Why not both?

[Actually, I do RTFA]

It is. Fridges desiigned for RVs and such are (often) dual input.

Re:Why not both?

[Anonymous Coward]

I don't think local grids are going to switch to DC anytime soon, but I'd wager that house/building level DC is inevitable. Virtually all devices in the home these days except for large appliances (Fridges, Washers, etc) run on DC anyway, they just need a power block somewhere between them and the socket. Future homes will probably still have a few AC sockets, but will also have DC power sockets fed by solar, wind, batteries and when none of those are available a whole house AC->DC converter.

Re:Why not both?

[trawg]

Thanks for the explanation. When we talk about "more efficient", how much are we talking about here? The article link mentions "5% or more of the power loss occurs" but as it's a Google Translate, not sure if it's talking about the same thing.

Does it basically mean these DC A/C units are 5% more efficient?

Re:Why not both?

[Smidge204]

More and more commercial AC equipment is moving to variable frequency drive compressors for efficiency, so if you can provide sufficient DC current at the necessary voltage to can shave a few percent off of the conversion losses.

=Smidge=

Re:Why not both?

[Anonymous Coward]

I don't think so. AC has a benefit of allowing to turn load on and off at 0V,.

Re:Why not both?

[Anonymous Coward]

This has potentially huge implications for the solar industry as inverter cost is becoming one of the dominate components of a solar power system as panel costs continue to reduce.

Not 5 minutes prior to seeing this article, I was pricing out solar power inverters.

A good one appears to cost somewhere between $2000 and $4000, depending on capacity. That's really not a terribly huge price for something you're going to install once and let run for 20+ years. (Unless, of course, they don't last that long, in which case that's the problem that needs solving. I'm not calling in a $Licensed$ $Electrican$ every 5 years to replace a permanently installed component that happens to burn out at a ridiculous rate. And this sort of thing is why I'm doing some research before I install anything in my house.)

Re:Why not both?

[Smidge204]

And yet, in practice, HVDC is still more efficient than current AC lines in the end, even if still somewhat more expensive at the moment.

Yes and no.

AC power is far more efficient at higher voltage and short to medium distances, and you save a lot of material (and thus money) on conductor sizes. The voltage can be changed easily and it is safer and easier to switch on and off since there's 50 or 60 times per second where the voltage/current is zero - allowing for the circuit to be opened without arcing or inductive voltage spikes. AC arcs also tend to be self-extinguishing for this reason.

But AC systems also have inductance and capacitance to deal with. For very high power, very long distance runs, the capacitive losses start to add up. More current is required to charge/discharge this inherent capacitance, which means more power losses. This is where HVDC really shines.
=Smidge=

Re:Why not both?

[Anonymous Coward]

The problem is one of standardisation. Can we come up with a good voltage to standardise on? For that matter what will be the allowed ripple of our standard DC voltage? Still, there is no great issue in providing a house wide DC supply in the wiring area, but I suspect for reasons of efficiency one voltage would not cut it, although 2 or 3 might.

Perhaps HVDC in to a module that generates whatever voltages are needed? We certainly want to get rid of all these AC/DC LED drivers. Even if those are efficient, they must be making bulbs more expensive.

Re:Why not both?

[Anonymous Coward]

AC power is far more efficient at higher voltage and short to medium distances, and you save a lot of material (and thus money) on conductor sizes.

Line voltage is limited by peak voltage (both from spacing and things like corona discharge), while carrying capacity comes from the RMS values. At higher voltages, DC wins because you can drive a higher RMS voltage, allowing a lower current. AC wins at the short to medium distances precisely because those are not the higher voltage links, and it comes down to switching equipment cost instead of transmission line costs. In other words, shrinking the conductor doesn't save much on short runs compared to other equipment, so it is cheaper to use larger conductors but cheaper switching equipment. But when you need long lines where you would save more by using smaller conductors, you go DC now. At high power loads, you can also run fewer lines too because the skin depth at 60 Hz starts to approach the size of conductors on major lines.

Re:Why not both?

[Anonymous Coward]

There is a unit on the market that takes DC from solar panels and AC from the house. When the sun is shining, the DC supplies ~ 85% and the AC the other 15%. When the sun is down the AC does the entire load.

Re:Why not both?

[Anonymous Coward]

Actually, you're full of shit.

Re:Why not both?

[jimbolauski]

There are all ready refrigerators that can run on either AC or DC that are used for boats and RVs.

Re:Why not both?

[tyen]

Inverters are typically warranted for 10-15 years, with the majority of the market's volume in the 10-12 year range. Enphase has a very controversial 25 year limited warranty on some of their inverters; user reports are all over the map on these units and the warranty claims service, so I wouldn't take them as any indication of what to expect on inverter life expectancies. Get rid of or mitigate the load handled by inverters in your solar power system design, and your inverter component cost goes down, but that is offset by the lack of selection in DC-based devices and for larger setups, the complexity of a safer DC power distribution design; there is no free lunch here going to straight DC, unfortunately. Definitely do not make your inverter installation permanent or even semi-permanent; it must be easily serviceable and removable.

Re:Why not both?

[thinkwaitfast]

High voltage only saves you on resistance losses. There are still inductive and capacitance losses with AC that you can't avoid.

Re:Why not both?

[SIGBUS]

One potential problem with the typical brushless DC motor is that it has permanent magnets using rare earth elements. The rotor in an induction motor, on the other hand, is just copper and iron. Induction motors can operate on variable frequency as well (the Tesla cars are a very good example of this, as are locomotives with AC traction motors).

Induction vs. brushless is a tradeoff of compactness vs. expensive rare earth elements.

Re:Why not both?

[macs4all]

Sure instead of using synchronous AC motors which are simple and efficient you can switch to much more complex DC digital motors.

Sounds brilliant.

Um, despite the Dyson commercials to the contrary, there is no such thing as a "Digital Motor".

Re:Why not both?

[monkeyxpress]

From the rough information I have about the appliance industry (I used to work for a company that had an appliances division) you go from around 80% efficiency with a cheap induction motor to around 90% for a brushless motor. However the controllability of the motors provides the really big benefits. For example, in a direct drive washing machine the motor can more efficiently change direction during agitation strokes, adjust speeds for things like load size, and implement more effective cleaning cycles (I have been assured there is real science behind wobbling the clothes back and forth different ways). In air conditioners and fridges the area of linear compressors is getting a lot of attention as you can remove substantial mechanical losses in the system. I don't know exactly how much these sorts of things add up too, but I'd say another 10% isn't unrealistic.

The 5% efficiency they are talking about refers to the losses in a typical cheap inverter, so you gain that as well.

I think the offsetting of the inverter costs is the real advantage though. Every bit helps towards making the economics stack up.

Re:Why not both?

[thinkwaitfast]

Inverters have a lifespan of ~10 years. There are no moving parts so killer is heat. If you run 1000w through a 93% efficient inverter, you generate 70w of heat. Running 100w generates 7w of heat. If your inverter is rated at 1000w it is built with 70w heat dissipation in mind. If you have a 5kw inverter, it would be built with 350w of cooling so running 1000w (70w of heat, 350w of cooling) will remain much cooler providing longer life.

Re:Why not both?

[Smidge204]

if you're going to have an internal module that "generates whatever voltages are needed" then you're just going through the conversions again, and you're saving nothing on efficiency (and losing a lot on costs).

The only way it'll work is if the DC input, eg solar panels, is matched to the unit's requirements. Anything other than that and you're just running in circles.
=Smidge=

Re:Why not both?

[fuzzyfuzzyfungus]

I didn't make it suitably clear; but the 'complexity' is really more of a historical issue. The fact that you can get power transistors, digital logic, and similar solid-state goodness for peanuts, possibly even less than the carbon brushes or other electromechanical alternatives, is a comparatively recent thing in historical terms.

Now that you can, doing so is pretty compelling for any but the highest-power tasks; but it has not always been the case that you can throw semiconductors at a problem for astonishingly tiny amounts of money. Today it is; but a lot of very clever electromechanical, inductive, and similar tricks were developed during the time that it was not.

Re:Why not both?

[jbengt]

True, but the trend for small, single phase motors is to use Electronically Commutated DC motors, converting the AC to DC and then using fast switching power transistors to drive the DC motor directly at whatever speed you want. The issue with using solar power would be that the solar cell DC voltage is lower than needed and varies. There would need to be a standard DC voltage established, and, in addition to putting the solar cells is series to achieve that, you may have to use DC to DC conversions to maintain a constant voltage (though, I guess, it wouldn't have to be exaclty constant, just within parameters req'd by the motor - still you need to standardize on those tolerances before you could see commercially useful adoption of DC in the home and office)

Re:Why not both?

[jbengt]

For smaller HPs, Electronically Commutated DC motors are becoming common where you want to vary speed and torque. But for larger motors (more than 1 HP), Variable Frequency Drives (AC to AC conversion) driving 3-phase induction motors are more practical and very common.

Re:Why not both?

[jbengt]

Just a question, wouldn't builld-up of static electricity be a problem for high voltage DC?

Re:Why not both?

[trawg]

Awesome, thanks for the info! Those sound like improvements worth holding off for. Be interesting to see what prices they'll start at & how they compare to existing A/C units.

Re:Why not both?

[jbengt]

It is brilliant.

Variable frequency AC inverters are nothing new (I've been specifying those for large HVAC systems since the late '80s/early '90s), they're just starting to get cheap enough to be used in low-capacity products like these. The only thing new is the possibility of having DC power available as the source, instead of 50 Hz or 60 Hz AC.

Re:Why not both?

[jbengt]

Generally, you are right. But an inverter 93% efficient at full power is likely much less efficient during low power operation. I've seen UPS literature (UPSs are basically AC to DC to AC converters with batteries included) where the total heat dissipation at 20%+/- load was greater than the heat dissipation at 100% load, let alone the efficiency.

Re:Why not both?

[Anonymous Coward]

> The voltage can be changed easily and it is safer and easier to switch on and off since there's 50 or 60 times per second where the voltage/current is zero

No, this happens 100 or 120 times a second.

Re:Why not both?

[ArcadeMan]

Can we come up with a good voltage to standardise on?

Sure we can. There's 48V, 36V, 24V, 12V, 5V and 3.3V

That's the nice thing with standards. You can have many!

Re:Why not both?

[compro01]

Or do combined-heat-and-power and use absorption chillers.

Re:Why not both?

[LinuxIsGarbage]

So called Brushless DC motors are actually permanent magnet rotor, synchronous AC motors?

I haven't seen a large change in small 1 phase motors. Most are still:
-Shaded pole induction motor for small fans
-Capacitor start induction motor for air conditioners, fridges, etc.
-Series wound "Universal motors" (that are basically a brushed DC motor) for intermittent high torque loads (vacuums, blenders, hair dryers).

Re:Why not both?

[LinuxIsGarbage]

What do we standardize on? 3 Phase 120V is 208V. Most American industrial is 480V 3-phase. Canada industrial is 600V 3-phase.

The DC bus for these drives run at a nominal 295, 679, 849 VDC respectively.

Plus you still have a lot of control equipment (relays, contactors, etc) running at 120VAC single phase.

There is already so much established equipment completely replacing the grid would be almost impossible.

Re:Why not both?

[Anonymous Coward]

There are multiple 1,000kV HVDC lines in North America.

http://www.npr.org/2009/04/24/110997398/visualizing-the-u-s-electric-grid

Re:Why not both?

[Anonymous Coward]

Inverters aren't new of course. I used one when I was in college to operate a 3 phase motor. Well it was technically called a variable frequency drive, but then that is likely what is really in the AC units. The biggest thing is that you had to watch certain frequencies since the motor would thrash, but then that motor wasn't even rated for VFD use so you can't complain too much.. We really only used it for soft starts/stops and limiting the max rate.

I actually seriously considered buying a 3 phase motor controller for my mother's house to increase the efficiency of the 4 ton heat pump here. Combine it with the right controller circuitry and you could likely bump up the efficiency a fair amount. There was a guy on youtube that actually did that. Of course to make it work I'd have to first replace a perfectly good single phase compressor with a three phase compressor and then I'd have to leave essentially an experimental system up when I was not around, so, admittedly I didn't consider it too seriously. A more realistic plan would be to figure out what frequency to run it for every month of the year. That wouldn't be quite as efficient, but it would be simple. I more or less sized the unit for the heating load, save it doesn't quite handle that on very cold days either. Still it has saved energy in winter. In summer it is sadly oversized. I half wish I would have put two units in at the time, one for upstairs and one for downstairs, but that is what I could afford then.

I'm actually visiting now and have put an unused small window unit in an out of the way spot in the living room. That is set to run most of the time, which means the big unit runs half the time and the cold air is more where it is needed. I suspect it will save money that way. Long term I still plan to get one of the more efficient 1 ton mini split heat pump, possibly from highseer dot com, and put in the main area, while moving the main thermostat to the main bedroom. I think that will cut the energy bill quite a lot, since it isn't heating and cooling so much unused space. Sadly it would still need to main unit to equalize the temps in the main bedroom, but it is likely a reasonable start with a fairly short payback time. It would also need the main unit when it gets really hot or cold of course, but most of the hvac work would be done by the higher efficiency unit... (Note I don't actually recommend anything from highseer dot com, but as for as value for the dollar they look reasonable, at least spec wise. I do not own any.)

Re:Why not both?

[Smidge204]

...why would we replace the grid?

=Smidge=

Re:Why not both?

[Smidge204]

So called Brushless DC motors are actually permanent magnet rotor, synchronous AC motors?

They are virtually identical, yes. There might be some nuanced differences in their physical construction or drive (sinusoidal vs trapezoidal waveform, for example) but the operational principle is the same.

For fractional horsepower motors there's no cost-benefit to doing sophisticated controls in most cases. You just need it to turn on and off at one, sometimes two or three speeds and the load is more or less constant. I wouldn't expect that to change any time soon.
=Smidge=

Re:Why not both?

[LinuxIsGarbage]

For fractional horsepower motors there's no cost-benefit to doing sophisticated controls in most cases. You just need it to turn on and off at one, sometimes two or three speeds and the load is more or less constant. I wouldn't expect that to change any time soon.

For which the shaded pole induction, capacitor start induction, series wound "Universal motors" do a great job. Especially equipment that runs for long periods, induction motors will run years without maintenance (like motors in fridges)

In the industrial setting, aside from a process that needs to vary speed during cycle / between qualities, efficiency gains from VFDs (on fans and pumps), usually means that the system wasn't properly speced out (when they're used for example, to run the same machine at 50% speed continuously). Across-the-line starter for an induction motor is always more efficient than a VFD running the same machine at the same speed. Change the gear ratio, change the motor rating, and you will get more efficiency out than installing a VFD.

VFDs are something like 97% efficent. It's incredible the waste heat that comes off a 2000HP VFD vs an across-the-line starter for the same application. Plus the VFD is higher maintenance than the starter.

Re:Why not both?

[K. S. Kyosuke]

Relevant?

Re:Why not both?

[K. S. Kyosuke]

I wonder, then, why don't I have a "simple and efficient" SSB transceiver in my cell phone...

Re:Why not both?

[Crashmarik]

Your phone is a software defined radio so you do.

Re:Why not both?

[Smidge204]

Offer void where constant-speed is not the most efficient. Pumps and fans that can match the actual demand by varying speed will be more efficient than running a full out and bypassing or artificially increasing head pressure to get the desired flow.

You're also not going to put an across-the-line starter on a motor larger than about 50HP unless you like replacing equipment. You'll always have a soft starter to get things going - this is doubly important if you're starting the motor under load.
=Smidge=

Re:Why not both?

[K. S. Kyosuke]

Ah, but that is also "much more complex" than a dozen transistors on a PCB, isn't it? ;-)

Re:Why not both?

[Crashmarik]

is it now ?

I can make the argument both ways. Can you ?

What Voltage?

[storkus]

Noticeably missing from both linked TFAs. As discussed here and elsewhere previously, 48V would probably have too much ohmic loss unless this A/C is right next to the supply. Higher voltage would work better, but call into question safety issues you don't have with AC due to it passing through zero volts 100-120 times a second.

Re:What Voltage?

[Great Big Bird]

> Higher voltage would work better, but call into question safety issues you don't have with AC due to it passing through zero volts 100-120 times a second.

AC running from hand to hand will likely kill you, causing your heart to try beating to it.

Re:What Voltage?

[Anonymous Coward]

Voltage doesn't kill; current kills and power burns. Higher voltage means lower current, and the same power. Your heart will fibrillate at 50mA AC or DC, unless the AC is 100kHz+ where the current won't penetrate to your heart. Your safety issues are therefore somewhat spurious.

At 48V, with 1mm^2 copper cable, the loss is ~8mW per meter, so your high ohmic loss is also somewhat spurious on a device that uses O(kW) of power.

 

Re:What Voltage?

[Anonymous Coward]

50V is generally considered 'safe'. Large parts of the world actually use 220V AC and it's not considered a serious hazard, as long it is installed and maintained by knowledgeable people.POTS (landline telephone) also uses about 50V, because it's considered safe - yet the shock can still be painful.

The difference (as i was told) is that if you touch a high voltage DC power line, the muscles in your hand may get cramped, so they you cannot 'let go' of the wire instantly. AC apparently doesn't have this problem, and will allow your reflexes to pull your hand back. If this plays on 100V or only above 1kV i wouldn't know for sure.

As far as losses go, see what other AC writes - within a household it's totally neglectable using proper wiring.

Re:What Voltage?

[Mr D from 63]

Noticeably missing from both linked TFAs. As discussed here and elsewhere previously, 48V would probably have too much ohmic loss unless this A/C is right next to the supply. Higher voltage would work better, but call into question safety issues you don't have with AC due to it passing through zero volts 100-120 times a second.

I'm not sure the resistance losses would be that significant as to be an issue. The safety question is a good one. 48V home DC systems are common enough and I believe they are well covered by code, so installations done right should be safe. DC can be problematic in that you can't always detect certain faults as there is not ground fault current, so there is inherently some greater chance of something like a bad connection overheating and causing damage, but that should not really be a concern if stuff is quality and installed correctly.

Re:What Voltage?

[Mr D from 63]

> Higher voltage would work better, but call into question safety issues you don't have with AC due to it passing through zero volts 100-120 times a second.

AC running from hand to hand will likely kill you, causing your heart to try beating to it.

So will DC. And at a given voltage, DC is worse as the current is constant and doesn't like to be interrupted.

Re:What Voltage?

[jenningsthecat]

Voltage doesn't kill; current kills and power burns. Higher voltage means lower current, and the same power.

Higher voltage only means lower MAXIMUM current, and then only if you assume constant power. However, it doesn't take much current to kill a person, and most real-world power sources can deliver enough current to kill under the right circumstances. (A mostly-dead flashlight battery can stop your heart if you bury electrodes deep enough in the right part of the body Also, think of Tasers - basically, low battery voltage raised to the point where it can stun or kill). Higher voltage usually makes death more likely, given the (approximately) constant resistance of a given current path through a body; I=E/R, so if E, (voltage) goes up, so does current. (Unless you're talking about static electricity from your clothing, or some other source which has high internal resistance/impedance and/or a small quantity of charge). And at still-higher prolonged voltages, the body's resistance can drop dramatically as parts of it start to boil and carbonize).

Your heart will fibrillate at 50mA AC or DC...

No. AC at a low enough frequency, (and at a surprisingly small current) will make the heart fibrillate; DC simply locks the heart muscle into a prolonged contraction. That's why defibrillators use DC - they temporarily 'freeze' the motion of the heart and give it chance to stop fibrillating and start beating normally.

Re:What Voltage?

[jenningsthecat]

DC can be problematic in that you can't always detect certain faults as there is not ground fault current, so there is inherently some greater chance of something like a bad connection overheating and causing damage, but that should not really be a concern if stuff is quality and installed correctly.

There's no connection between ground faults and bad connections that might cause overheating. But to the extent that DC systems might have lower voltage and therefore higher current, bad connections ARE more likely to overheat and cause fires. Also, there's no reason a properly installed DC system can't have Ground Fault Interrupters, although the ones currently used for AC won't work on DC. The ones designed for DC would be somewhat more complex, and probably bigger as well.

Another note: interrupting Direct Current without arcing can be difficult. AC has a zero crossing that extinguishes an arc across switch contacts, whereas the equivalent DC circuit may continue to arc across switch or relay contacts. Such switches and relays typically have heavier contacts and the contacts, when open, tend to have more space between them. The may also have permanent magnets nearby to act as 'blowouts' to extinguish any arc that develops.

Re:What Voltage?

[Mr D from 63]

There's no connection between ground faults and bad connections that might cause overheating.

I understand that and did not mean to imply there were. I suppose my wording could have been better. Just meant to point out some specific inherent differences that relate to safety.

Re:What Voltage?

[Anonymous Coward]

I must admit, I'm a bit surprised by all the talk of how dangerous DC is in all of the replies to your post. Clearly I'm quite ignorant on this topic. I've always thought that DC was safer. I remember reading that in the battle between Edison and Tesla over DC vs AC, Edison even electrocuted an elephant to prove how dangerous AC was. I've always assumed that implied that DC was safer, And over all these years, I've never seen anything commenting how ironic it was that Edison was demonstrating the dangers AC while promoting the more dangerous DC.

Can someone clarify this for me? Was it just that edison was backing a low voltage DC, and that low voltage DC is safer than AC (though less efficient, which is why it lost out in the market)?

Re:What Voltage?

[stabiesoft]

Yes I am curious too. My panels will generate somewhere between 250 and 330VDC if I remember right. The inverter refers to this as MPP maximum power something to extract the most power from the panels. I am wondering if the sharp unit allows for a range of voltages. If it does not, then it still may be more efficient to convert to AC then back to DC so the panel efficiency is maximized. If not that, then a DC to DC converter may be needed in front of the sharp unit if it does not like variable DC voltages. I assume the DC/AC inverter to convert leftover power from the panels will still be responsible for MPP, but what happens if a cloud comes and the panel power drops, I've seen swings as much as 70% from thick clouds. Does the sharp unit shut off, or does it pull from the grid thru an AC/DC converter. Article is very light on details of real world conditions, such as what happens at night?

Re:What Voltage?

[Anonymous Coward]

This is not true:
http://www.brighthubengineerin...

Re:What Voltage?

[Anonymous Coward]

unless the AC is 100kHz+ where the current won't penetrate to your heart.

The skin depth for 100 kHz in a human body is still 30-100 cm, plenty deep enough to reach your heart. You need to go up another order of magnitude in frequency at least.

Re:What Voltage?

[Anonymous Coward]

I'm fairly sure that they dedicate 2-3 panels to the AC, so maybe 36-45v (depending on panels)

Re:What Voltage?

[thinkwaitfast]

Maximum Power Point Tracker. The more current you draw, the lower the voltage. Since P(V,I) = ~V x I(V), you want to operate at the voltage that maximizes the power (at the first stage off of the solar panel). Once you have the energy stored (in a capacitor or inductor), you can convert it to whatever level you need.

Re:What Voltage?

[Polo]

As to loss, thicker wires would eliminate the issue.

I think it would be great for off-grid applications.

Re:What Voltage?

[Anonymous Coward]

From personal experience dealing with lab power supplies, DC hurts FAR less than AC for the same current and the same voltage. I'd go as far as to say it hurts at least half as much and I'm talking above the threshold where DC can actually hurt you, which is obviously much higher than AC. The reasons for that, I honestly have no idea. If I had to guess, I'd say it's either some weird resistance effect in the human body or a capacitance effect similar to "charging up" your body on a carpet. But then again, I'm no EE, I'm just some hobbyist who have gotten shocked far too many times.

Disappointed

[Anonymous Coward]

I'm still disappointed that there is no universal power converter chip yet. You know, a chip with a configuration interface (I2C or somesuch), two wires going in and two wires going out, taking anything from 12V DC to 240V AC, and providing anything from 1V DC to 48V DC, current limited.

Re:Disappointed

[amalcolm]

The will arrive just after my goose lays the first golden egg.

Re:Disappointed

[Anonymous Coward]

OK, that's the gut reaction I expect, but what's so outrageous about the requirements? If I don't need the output to be isolated from the input, I can just rectify the input and treat all input as (pulsed) DC, so that's one requirement down. Then it's a DC-DC up-down converter with a wide input and output range. Doesn't sound too outrageous to me. With high switching frequencies, on-chip capacitance should be sufficient; maybe off-chip, in-package capacitors are needed. Perhaps two more wires would be needed to attach a big capacitor so that the converter could store power to avoid pulsed DC input resulting in pulsed DC output.

IMHO a chip like that would be like going from discrete logic to microcontrollers, but for power instead of logic.

Re:Disappointed

[Great Big Bird]

The size of such a chip and the heat dissipation requirements would mean you would only want to use it for the higher end applications.

Re:Disappointed

[Anonymous Coward]

Like you would only want to use a microcontroller for the higher end applications? "You're using millions of transistors for WHAT?" "Calm down, it's cheaper."

Re:Disappointed

[ledow]

Compare size of fixed-voltage adaptors, to those able to cope with 110v/240v output.

Compare size of fixed-output adaptors to "multi-output" adaptors. Some laptops ones are tiny. All the "generic" adapators are huge.

And that's with a handful of options, maybe 2 options on the input and 3-4 on the output. Now combine sizes. Now add in an input capable of the DC (yes, you can Wheatstone bridge the AC, but that's got to be using diodes big enough for anything you put on). Now add intermediate paths capable of the MAXIMUM current that goes through it, as DC, all the way, down to the minimum voltage (i.e. if it can output 1v, that's going to give 12 TIMES the current that outputting at 12V would give, needing seriously large cables and intermediate circuitry for any practical purpose (even 30W, at 1v, would give 30A ratings... 30A DC @ 1V needs 120mm thick cable for a 1m run).

Even adjusting for "reasonable" voltages up/down, you're also expecting up and down conversion (i.e. mains to 12V and also 12V to 48V), which is rare to see in the same device.

Basically, the thing would be as huge and heavy as a car battery, get fucking hot, be prone to failure, and require internal circuitry and insulation from the power paths that's just not practical. And it would probably be quite inefficient across whole ranges of voltages.

You can't have some tiny 5v chip controlling a variable output of that kind of size without some huge, specialist "break open a UPS and see the kind of size we're talking about" components in it, with thick paths between them.

Honestly, the first thought that comes to mind is "fire", quickly followed by "fucking expensive". There's a reason that you set a standard and follow it and try not to change that standard (hence why half the world refuses to change to the other half's mains voltage). The conversion equipment is then static, and makes it practical. And there's a reason that 100+ Volts emerges as the winner every time, because you can get decent power down a decent sized cable.

Hell, even 12V DC / 240V AC input devices are "rare" in modern life compared to everything else - motorhomes and marine use, pretty much, and expensive and sometimes several years behind the technology. And internally all they do is convert to one or the other, so it's really just a TV and a convertor in a box wired to the same 12V DC input on the actual TV (which almost certainly then boosts the voltage for it's LCD panel) because that's easier to make than some generic device that can take in anything and shove it where it wants in whatever format it wants.

Fuck, even inverters or voltage convertors (110/240V) as standalone devices for generic use are bricks that weigh a ton, cost a lot, and have very limited power output.

Re:Disappointed

[Overzeetop]

So you're saying it's impossible to do an arbitrary DC/AC to DC/AC without a massive efficiency penalty?

Re:Disappointed

[Anonymous Coward]

You can't compare information processing to power. Jesus Christ you programmers are tiresome with that fallacy.

Re:Disappointed

[Anonymous Coward]

I understand that this isn't conventionally possible. I'm not trying to wish that away. I'm pointing out that there is a similarity to discrete logic vs. microcontrollers: Instead of looking for few things doing a lot, I want designers to think in terms of many things doing a little each, but achieving a lot in sum. Power supply engineering appears to be all about cleverness and tuning parameters just right. Instead I want an approach that embraces complexity, because logic is dirt cheap. Make something that depends on advanced control, lots of real time calculations and thousands of components working together. As long as it can be etched into silicon, it's still going to be cheap. Nobody would dare suggest building something as complex as a microcontroller for the things that many microcontrollers are used for today, but nevertheless they are cheaper than even very simple discrete digital logic.

Re:Disappointed

[fuzzyfuzzyfungus]

The competition for good DC-DC conversion is reasonably fierce(given the existence of DC telco and datacenter operations, and the fact that even 'AC' shops are really just doing the conversion in each chassis(and unlike the old AT PSU days, an ever larger chunk of the output power is 12v going directly to a DC-DC converter on the motherboard to feed the CPU and RAM, with fewer and fewer components, aside from HDD motors, being sufficiently high voltage to feed directly from the PSU); so even modest improvements in DC-DC efficiency would make you quite wealthy indeed.

I haven't kept a close eye; but I think that the present standard for DC-DC modules still uses a number of off-chip components(whether because the needed capacitance and such simply can't be done in silicon, or are cheaper as discretes, I don't know); but you can get some very, very, dense little modules.

Re:Disappointed

[ledow]

God, really? What don't you get about this?

THE SHEER PHYSICS INVOLVED MEANS HUGE POWER CONNECTIONS. What can fit in a Cat5e-thick cable at 240V needs a fucking iron bar to carry the current at 1V. DC requires even thicker iron bars than AC because it arcs like fuck.

You can't shove an iron-bar's worth of current through a microcontroller with teeny, tiny traces without things catching fire. Even if the logic is abstracted (e.g. UPS), still something somewhere has to shift enormous iron-bars carrying current around. Look inside a UPS at the size of the capacitors, the width of the traces, the amount of stuff just put through physical relays to isolate real-current-carrying stuff from the controller circuitry, and the sheer thickness of copper in the battery leads.

Sheer physics says that what you want, something that can take account of all situations, means you have to design to the highest spec in order to not start fires. And that means a huge stonking great, very hot, dangerous device.

Honestly, I can take out a room by shorting a decent-sized car or UPS battery. Fuck knows what happens if you get an arc inside your multi-convertor device if it develops a fault.

There's a reason it doesn't exist, for the same reason that 10GHz chips don't exist. Simple laws of physics.

What the bollocks was about nobody daring to suggest building microcontrollers for the things they are used for today, I have no clue. That's EXACTLY what microcontrollers were specifically built to do and have been doing since their invention.

P.S. Discrete logic runs generally slower than IC because of one very simple thing - the speed of light. Look up any of the wirewrap/homebrew "computers" made from discrete components of ordinary size. They are severely limited because the length of their traces (on the order of centimetres rather than micrometres/nanometres) means that signals get out of sync before you get past 8MHz or so. This is the same reason that you don't have 10GHz chips yet, but can have lots of multicore 3GHz chips in the same package.

Re:Disappointed

[Anonymous Coward]

You don't seem to understand that it's not a logic issue, it's a physics issue with size constraints being put up by coil sizes, switching FET sizes, etc. to do what you suggest. As someone else said: it'll be the size of a car battery, get freaking hot, and won't be reliable - plus it would be very, very expensive. You can't compare this to a microcontroller...

Re:Disappointed

[Anonymous Coward]

Who said anything about amps? Obviously I don't expect to run a hairdryer on 1V 1000A. What I want though is to plug an LED, a tiny controller or a 30W laptop into a 240V outlet without needing a wall wart. Or power these same things from a 12V battery. That should be doable. What I don't want is a box full of power supplies for every imaginable combination of input voltage, output voltage and AC or DC. If you look at the designs in use today, they're still mostly analog, even if they're integrated circuits. And even though they are designed to be used with few external components, one of them always is a comparatively big inductor, because these chips are designed to be used in buck/boost converters. Those are design choices, not matters of physics.

Re:Disappointed

[thinkwaitfast]

They have them, but the converion process requires some amount of energy storage which can't be done at wafer scale without dielectric breakdown.

Re:Disappointed

[Anonymous Coward]

Just to counter that ridiculous size argument: Look at these things: 90W laptop power supply, 1.23V-30V buck converter (max 3A), 150W DC-DC Boost converter. Between these three items, you almost have everything I asked for: PSU for AC input (100V-240V), buck converter for going down to almost 1V, boost converter for going up from 12V. That stuff is neither as big as a car battery, nor does it get freaking hot or cost an arm and a leg. Yes, I am asking for a solution that's one to two orders of magnitude smaller, and I want it all in one and automatic, but that's not as outrageous as you make it appear.

Re:Disappointed

[Agripa]

So you're saying it's impossible to do an arbitrary DC/AC to DC/AC without a massive efficiency penalty?

Efficiency is not so much as a problem as cost. It is easier to design converters with specific input and output voltages.

Converters commonly have a fixed output and 2:1 input voltage range which covers universal input power supplies with a 120 through 240 AC input . Converters with a 4:1 input voltage range are not as common but are available for a price premium. 10:1 input ranges are rare.

The problem with large input and output voltage ranges is that for the same power, the electronics have to handle both high voltage at low current and high current at low voltage making the design less optimal, less efficient, and more expensive.

Re:Disappointed

[Anonymous Coward]

That's with current buck/boost designs. There are other ways.

DC power

[wonkey_monkey]

At last, they're harnessing all that hot air coming out of Washington for something useful.

On a more serious note, what are the benefits/costs of using AC over DC in the home? Would things be more efficient if we continued to receive AC down the wires, converted it to DC in the basement, and fed everything with DC? Seems like plenty of devices really want DC, and come with those lovely hand-warming bricks.

Do modern TVs run on AC, or are they just converting it to DC internally as well?

Despite getting good grades in physics, I've never quite grasped electricity as a concept...

Re:DC power

[LinuxIsGarbage]

Even when they take DC, a lot of them are different voltages. Phone / USB chargers are 5V. Things like Laptops will be ~20V. So you will still have conversions. And the power going around, 120V or 240V, allows you to run moderate sized appliances without ridiculously thick cables. High voltage, low current, rather than low voltage, high current.

Re:DC power

[peragrin]

This is it exactly. 5V for usb, 1.5 V for the processor, 48 volt for the back light, etc. and what about things like your stove, refrigerator etc? things that use AC power for their motors.

you get 120/240V at every outlet, you can then convert it to any voltage you need fairly easily. if you are running 48V DC it is a lot harder to convert that down to 5 V.

Re:DC power

[Anonymous Coward]

if you are running 48V DC it is a lot harder to convert that down to 5 V.

A stamp sized board which will do that with high efficiency costs about $0.70 including eBay fees and shipping from China.

Re:DC power

[Anonymous Coward]

> if you are running 48V DC it is a lot harder to convert that down to 5 V.

No, it's not. DC-DC-conversion is very efficient and cheaper than AC/DC conversion. Having 12-48V locally is pretty efficient. Just look inside your computer (not the power supply), there's half a kilowatt in tiny DC-DC-converters converting 5V or 12V down to what's actually needed.

The problem with high power DC is
1) high voltage DC is more dangerous than AC and requires different cabling and switches.
2) low voltage DC would require higher amperage and therefore a lot thicker cables.

Re:DC power

[AHuxley]

Re "On a more serious note, what are the benefits/costs of using AC over DC in the home?"
AC gives you the national grid, hydro, power stations and epic scale.
DC gives a solar setup one less DC to AC to DC loss conversion to get the same result in the home setting (lots of roof panels, sun, short DC wire length to correctly sized air con unit).
Re: Do modern TVs run on AC, or are they just converting it to DC internally as well?
A boat, RV or truck shop can help with a list of DV 12v and 24v devices. Wire thickness, length, amps, devices used, storage then gets to be interesting design cost in the home setting.
With AC within reason any electrician can give you a great deal of "power" from the grid, 24/7 per room. With DC the length, width, usage, voltage math starts to get more interesting per device added.

Re:DC power

[Megane]

The main advantage of AC is that you can use higher voltages safely, and higher voltages mean higher wattage with the same wires. And bigger wires are more expensive.

AC versus DC load breaking comparison with a knife switch

That was 220 volts, but 110 volts isn't much better on the DC side. There's a reason why DC-powered telecoms equipment uses 48 volts; much more than that and switches start arcing.

Ohmic loss is an issue when DC power is transmitted over power lines, but not so much when the DC is generated in the same building (solar panels, etc.).

Re:DC power

[K. S. Kyosuke]

DC gives you efficient long-range transmission. I.e., that "epic scale" you spoke of.

Re:DC power

[Megane]

I'm pretty sure most of those DC-DC converters (the modern efficient ones) work by converting to high-frequency AC internally, running it through a transformer (higher frequency = smaller transformer), then back to DC.

Re:DC power

[Anonymous Coward]

No, they're usually buck converters.

Re:DC power

[AHuxley]

Yes looking up from a home setting AC? Or down the grid (HVDC). https://en.wikipedia.org/wiki/... and https://en.wikipedia.org/wiki/...
The DC air conditioner might still be interesting with a savings % on site for solar been more direct and less AC to DC conversion loss.

Re:DC power

[Eunuchswear]

Thanks for linking that -- great demo.

Re:DC power

[squiggleslash]

Well, I think we are getting better at converting DC voltages, which is why HVDC is being used for transmission lines for example.

I suspect the reason is in part portable electronics. We're trying to eke out as much power as possible for multivoltage devices (one voltage for the processor, another for the screen, another for the HDD (portable electronics includes laptops too...) another for the USB bus, etc) from a single (DC it goes without saying) battery. The amount of R&D into the voltage conversion field over the last thirty years must have been extraordinary, yet not sexy enough to warrant much media coverage.

Re:DC power

[Anonymous Coward]

Sort of. They switch power into a coil at high frequency, then switch it out at a different rate which lets you choose the voltage.

Re:DC power

[Anonymous Coward]

Yes, very good. Bravo for that trolly FUD.

He's using wire-wound resistors that have very high inductance, and then he puts a huge negative d(current)/dt onto them, so he gets 10kVs of back EMF, which then arcs, completing the circuit while the capacitor discharges at leisure through the load. This circuit would actually make a pretty good arc welder - it generates the 10kVs of strike voltage and then provides the 100Vs of maintenance voltage. So this is a great demonstration of parasitic inductance, and a great demonstration of bad DC design when faced with an inductive load.

What does good DC design look like for this case?

If you want fast switching, and diodes have been invented, you add a diode to prevent back EMF. This is why motor controllers have diodes on them - the back EMF can easily destroy the FET driver otherwise.

If you don't mind sacrificing switching speed, and diodes haven't been invented yet, a capacitor can also be used to drive dI/dt down however low you like, at the expense of somewhat "soft" switching relative to power consumption.

None of this is news. 48V alone doesn't guarantee that this won't happen. dI/dt determines the back EMF, not V.

Re:DC power

[thinkwaitfast]

But buck converters do operate by converting to high-frequency AC internally, running it through a transformer (higher frequency = smaller transformer), then back to DC.

Re:DC power

[Charliemopps]

You're forgetting that AC is significantly more efficient at driving a motor than DC is. Go look at any good power tool, washer, drier, they all require dual phase AC for a reason.

This is a step backward.

Re:DC power

[Anonymous Coward]

No, that's not a transformer, that's an inductor. There is no secondary winding. Modern AC-DC converters (and the much less common isolated DC-DC converters) use small transformers and high switching frequencies (flyback converter).

Re:DC power

[Anonymous Coward]

No, they usually switch the coil at a fixed frequency. The coil is either charging from the source or discharging into the load. The output voltage is controlled by varying the duty cycle (how long it charges vs how long it discharges each cycle).

Solar Powered Aircon

[Anonymous Coward]

Go one better and make a solar powered airconditioner that uses the heat of the sun directly on a phase change material, without the need for the electricity in the middle.

aka. The Solar Powered Fridge

https://en.wikipedia.org/wiki/Solar-powered_refrigerator

Since aircon is a big energy user, and it uses that energy when it is sunny, it follows you could cool a big thermal mass with a solar powered cooler and that would keep the house cool during summer months.

Re:Solar Powered Aircon

[K. S. Kyosuke]

Too complicated energy pathways. You can't power everything by heat. If your cooling needs are saturated, you can't power anything else with it. So you need a parallel supply of electricity again. And once you have it, why the redundant heat engine in the first place?

Re:Solar Powered Aircon

[Smidge204]

Of course you can power everything with heat. Indeed, nearly everything *is* powered with heat; Most conventional power plants use thermal processes, converting heat energy into mechanical energy then into electrical energy.

A solar powered refrigerator (or any refrigeration cycle driven directly by heat) allows the use of fairly low quality heat sources to do useful work without losses converting it to electricity first. Very useful in some circumstances.
=Smidge=

DC AC

[sglow]

I love it, a DC AC

Re: DC AC

[Anonymous Coward]

What are you taking about? The abbreviation for air-conditioner is A/C. Idiot.

Good to see someone has managed to change physics

[Anonymous Coward]

In a electrical engineering class we were introduced to why we didnt have DC in our houses, and right amongst the reasons learned from the very first days of DC housing installs were that if you were unfortunate enough to get a shock from one, it was much more damaging to you as your natural response was not to spasm muscles thus releasing, but to remain gripping the origin of the shock. In short you get "stuck" to a dc shock.
How have they mitigated this major downside?

Re:Good to see someone has managed to change physi

[Anonymous Coward]

spasm muscles thus releasing

Here on Earth we have different anatomy.

Already been done in China for a while

[RobinH]

I've been saying for a few years that if you just had a few solar panels in your back yard, and didn't want to go through the expense of all the inverter stuff, you could just use it to charge a small battery and power a DC air conditioner. That's because you generally want air conditioning at the same time that you have the most solar power. At the time, the only DC air conditioners available were for marine use, and so they were expensive. However, in the last year and a half I noticed a lot of DC air conditioners on the marker on AliExpress (in China). Some of them even come as a kit including solar panels. The difference here is that presumably the Sharp ones are UL and/or CSA certified, so you could use them in North America.

Honestly, some of the stuff on AliExpress is impressive for how cheap it is. You can buy 500W grid-tie inverters for a solar array for the $200 range. Unfortunately they only have a CE rating, so they're not OK for North America yet. In comparison you can spend 3 to 4 times that much here.

Re:Already been done in China for a while

[Dog-Cow]

I don't know about building codes, but the UL is a private organization. There's no legal need, in general, for anything to be UL-certified in the US.

Re:Already been done in China for a while

[Anonymous Coward]

There's no legal need, in general, for anything to be UL-certified in the US.

Legal, as in criminal? Perhaps not. There may be a few jurisdictions in certain circumstances that require it on inspection, but even those would hardly notice.

Legal, as in your insurance company won't pay out? That's another matter.

Re: Already been done in China for a while

[Anonymous Coward]

No legal requirements, but good luck getting an insurance claim paid if your non-UL appliance causes a fire.

Re:Already been done in China for a while

[dj245]

I've been saying for a few years that if you just had a few solar panels in your back yard, and didn't want to go through the expense of all the inverter stuff, you could just use it to charge a small battery and power a DC air conditioner. That's because you generally want air conditioning at the same time that you have the most solar power. At the time, the only DC air conditioners available were for marine use, and so they were expensive. However, in the last year and a half I noticed a lot of DC air conditioners on the marker on AliExpress (in China). Some of them even come as a kit including solar panels. The difference here is that presumably the Sharp ones are UL and/or CSA certified, so you could use them in North America.

Honestly, some of the stuff on AliExpress is impressive for how cheap it is. You can buy 500W grid-tie inverters for a solar array for the $200 range. Unfortunately they only have a CE rating, so they're not OK for North America yet. In comparison you can spend 3 to 4 times that much here.

It's really eye opening how much middlemen mark up chinese goods. My wife has ordered high-end clothing and a fabric baby carrier from Aliexpress. Sometimes the quality is typical of cheaply made goods, but in many cases we can't tell if it is an excellent knockoff or a case of "Prada ordered 10,000 units, let's make 12,000 and sell the extras ourselves". The markup on mainstream high-end goods is extreme. Independent entrepreneurs have taken advantages of this in some sectors, but not others. Only the threat of legal action is holding back the tide in some cases.

That said, I used an NEC relay in my last project, as opposed to the equivalent chinese model. Dresses and purses might have the ability to catastrophically fail, but when they do, my apartment doesn't burn down.

Re:Already been done in China for a while

[RobinH]

Well, I'm in Canada and you absolutely cannot connect anything to the grid unless it's cUL or CSA certified. That's on the 220/110 side at least. If you go through a CSA certified power adapter and come out at 12VDC, they don't really care. I know from experience in the US that things are a lot different.

Re:Already been done in China for a while

[bluegutang]

Yeah, but do you trust a random product purchased from AliExpress to actually meet the certifications it claims to? AliExpress has the problems of online merchants and contemporary Chinese business ethics all put together.

Re:Already been done in China for a while

[dj245]

I don't know about building codes, but the UL is a private organization. There's no legal need, in general, for anything to be UL-certified in the US.

I have had fire marshals come down hard on my company for electrical equipment that didn't have the sticker. I've experienced it in 2 different states, at 2 different companies. And I have heard of other companies having similar problems. If it doesn't have the sticker than it falls outside of common exclusions for inspections. We had to have electricians come in and verify that the equipment was safe by checking every wire to code and generally accepted practices.

Maybe for a residential installation it would be fine. Or maybe it wouldn't be. You're basically betting that the home inspector wouldn't see a problem with it. Home inspectors are generally assholes so that's not a gamble I want to take.

Re:Already been done in China for a while

[Dan East]

That is not a DC air conditioner. Note that it comes with a 2kw inverter. It's just a regular 220V AC air conditioner. Also, that whole package is sketchy. The stated BTUs don't match throughout the page (title and description says one thing, specs say another). It comes with 4 solar panels, but there are absolutely no specs on them - not even the wattage. Anyway, that package is solar panels, batteries, huge inverter and a regular air conditioner. The efficiency would have to be very low.

Re:Already been done in China for a while

[Thelasko]

I've been saying for a few years that if you just had a few solar panels in your back yard, and didn't want to go through the expense of all the inverter stuff, you could just use it to charge a small battery and power a DC air conditioner.

The big advantage of the DC air conditioner is that you can run it in parallel to a traditional air conditioner and don't have to worry about the grid connection. This will take load off of the main air conditioner and save money. No special controllers needed, just set the target temperature on the DC unit below the target of the traditional system. That way the traditional system should only run when it's both hot and dark/cloudy.

Re:Already been done in China for a while

[jbengt]

I don't know about building codes . . .

That's obvious.
Most building, electrical, mechanical, etc. codes require certain products to be tested by a nationally recognized testing agency, of which UL is commonly named in the code. (BTW CSA and UL are largely interchangeable these days, as they cross-list many, if not all, of the products tested by one or the other.)

Re:Already been done in China for a while

[jbengt]

I know the feeling.
Many years back, I was involved in the design of a university chemistry lab where the lab equipment was pre-wired and UL labeled. However, the City of Chicago had different electrical requirements (they're somewhat less rigid and arbitrary these days) so it had to be re-wired. The kicker is, the City of Chicago required it to be UL labeled, so after re-wiring, the contractor had to pay around $5,000 (in 1980s dollars) to get UL to come to the job site, inspect & test the equipment, and slap a UL label on it.

DC is great until you touch it

[Anonymous Coward]

And the current flowing through you into the ground locks your body onto the conductor, juicing you until you're dead. AC at least has the decency to throw you across the room.

I'm surprised Slashdice hasn't linked this to those home battery units that Musk was pimping. We haven't had a minky musky story in days.

A much more efficient air conditioner, too?

[swb]

Is this not just a change in power input but a substantially more energy efficient air conditioner, too?

I've seen small A/C systems for cars and marine applications that can run off DC power, but they're usually pretty small which helps cut the overall power consumption. In marine applications they also have the advantage of being able to use sea water to move the heat versus a fan and coils in open air.

One of Sharp's smallest split system units has 8500 BTU of cooling with an EER of 13 which is roughly 650 watts. That's about 14 amps @ 48v, 27 @ 24, and a battery sucking 54 amps at 12v (run with welding cable).

8500 BTU might cool a room reasonably well, but its not going to provide whole-house cooling, either, and would require a pretty large battery array to run off battery. It might make sense for some kind of supplemental cooling setup where it ran direct off solar panels.

Re:A much more efficient air conditioner, too?

[ledow]

An A/C capable of cooling a car or boat isn't really comparable. First, they move through outside air when in use or are submerged, so cooling is aided quite a lot. Second, they barely have to cool the equivalent of one room in a house. Maybe more rapidly, but that's just a case of brute power.

However, the volume covered means they aren't actually cooling at max power for more than a few minutes anyway, and the sealed interior of a car means they can dial-down quite quickly. They also don't have to worry about killing the battery at all, they are effectively "plugged in", and I'm not sure how long they'd last on battery - certainly it would be much worse than leaving your lights on!

Applying such things to a household running on solar? I think there's going to be more than a few small problems.

Re:A much more efficient air conditioner, too?

[Overzeetop]

The good news is that, potentially, this would be tied to a large PV battery bank. He who shall not be named's battery packs are Li-ion at 400V which, if run directly, would be just over an amp and a half (about 25% of the basic battery pack's continuous delivery capacity). I agree it's not going to make a whole house system work, though. I don't do the hardcore side of ME, so I don't know what separates the typical residential 2.5-3.0 CoP of a whole house unit from the mini-splits that seem to get closer to 4.0 and can suck heat out of colder sinks. The mitsubishi in my home office is a beast, pumping out hot air even when it's 10-15F outside and the main system is hitting the resistance coil backup heat.

FWIW, AC systems for cars are actually pretty demanding - a typical system for a full sized sedan is ~20kBTU. Then again, you have mechanical compressor power to spare.

Re:A much more efficient air conditioner, too?

[DigiShaman]

8500 BTU might cool a room reasonably well, but its not going to provide whole-house cooling, either, and would require a pretty large battery array to run off battery. It might make sense for some kind of supplemental cooling setup where it ran direct off solar panels.

Translation: Turn your detached garage attic into a whole new room. Place solar panels on the roof and cool it with a Sharp Air Conditioner. With renting, you can now pay back your investment and then some. Winning!

Re:A much more efficient air conditioner, too?

[Anonymous Coward]

Today's most efficient mini-splits offer double that efficiency; though 27 Amps is still horrendous, it's manageable with enough batteries, and one needs to realize that a properly insulated room will only need to run the compressor perhaps 30% of the time (on average throughout the day and night). This brings us into the less than 10 Amps per hour average range.

Golf cart batteries can easily supply 100 Ah without experiencing any damage, thus I predict 6 batteries would cover cooling for 1 room if the product is designed well.

Hopefully they can start making DC powered compressor based fridges. I believe the last company doing that went out of business, and it's a pain in the ass powering a fridge from an inverter. Powering a decent sized A/C from an inverter is horrendous. Both devices have horrible startup current requirements that require ridiculously oversized inverters (typically 3 - 5x the plated power requirements).

Re:A much more efficient air conditioner, too?

[swb]

It'd be interesting to know what Sharp plans for the power input. I would suspect the market starts to shrink dramatically for input voltages over 48V because pretty much all battery arrays are 48V or lower and AFAIK (which isn't very far) only the newest solar installs run at high DC voltages.

I'd guess that this would be a 24-48V system (highest common DC voltage in battery arrays) and lets say you have 6 hours runtime after dark (pure battery load), you're burning 4000 watt hours of power or 80+ amps @ 48V and 160+ amps at 24v.

The daily use PowerWall is only 7kw and I'd guess a summer of that kind of use would put a serious dent in its lifespan.

The only thing I can think of is that the Sharp DC A/C is designed for sucking direct from larger solar panel installs during the sunny days and really isn't practical to use for night cooling without some kind of other prime energy source (generator, grid, etc).

Re:A much more efficient air conditioner, too?

[Charliemopps]

Bascially Air conditioners only use electricity in 2 ways. A motor in the compressor, and a motor to drive the fan. AC motors are MUCH more efficient than DC motors. Any industrial electric motor you find will be AC. This is a step backwards.

This is logical next step

[140Mandak262Jamuna]

People in USA and Europe with excellent grid connections are not aware of it. But in places like India with unreliable grid, people have been using backup electricity storage for quite some time. Typically truck lead-acid batteries are used to store enough energy to power a couple of ceiling fans, a few lamps and the TV, never forget the TV, for a few hours. They put up with power outages using these contraptions.

They use inverters to convert the DC to some square wave and approximate it to A/C using electronic gimmicks. Not a pure sine wave A/C, but close enough to run fans and the lamps. Energy conversion efficiency is not bad, the inverters do hot heat up too much. But they play havoc with the motors. So the Japanese A/C makers have been selling ruggadized air conditioners that can run on the inverter electricity.

The logical next step is to create A/C to run purely on DC. Probably it would use AC to DC converters to use grid electricity. Again this DC would be poor in quality compared to battery DC. So this Aircon also would need to be ruggadized.

All these calculations about when residential solar will become viable compared to coal or natural gas are completely different between G8 and rest of the world. Places like India will pay well over the current grid price for steady electricity supply. Not all of them. But the affluent population of India is about the size of Japan, some 120 million people. They have been making do with truck-battery-inverter contraptions, small gasoline generator sets etc. They would probably form the wave of early adopters who pay for the early fixed costs of solar panel factories.

Re:This is logical next step

[dj245]

People in USA and Europe with excellent grid connections are not aware of it. But in places like India with unreliable grid, people have been using backup electricity storage for quite some time. Typically truck lead-acid batteries are used to store enough energy to power a couple of ceiling fans, a few lamps and the TV, never forget the TV, for a few hours. They put up with power outages using these contraptions.

They use inverters to convert the DC to some square wave and approximate it to A/C using electronic gimmicks. Not a pure sine wave A/C, but close enough to run fans and the lamps. Energy conversion efficiency is not bad, the inverters do hot heat up too much. But they play havoc with the motors. So the Japanese A/C makers have been selling ruggadized air conditioners that can run on the inverter electricity.

The logical next step is to create A/C to run purely on DC. Probably it would use AC to DC converters to use grid electricity. Again this DC would be poor in quality compared to battery DC. So this Aircon also would need to be ruggadized.

All these calculations about when residential solar will become viable compared to coal or natural gas are completely different between G8 and rest of the world. Places like India will pay well over the current grid price for steady electricity supply. Not all of them. But the affluent population of India is about the size of Japan, some 120 million people. They have been making do with truck-battery-inverter contraptions, small gasoline generator sets etc. They would probably form the wave of early adopters who pay for the early fixed costs of solar panel factories.

When I visited North Korea, individual solar panels were everywhere, in cities and in the countryside, charging lead-acid batteries for lighting at night. Being on China's doorstep, it may be cheaper to do this than to build out the grid. Widespread availability of food refrigeration would help poor countries tremendously.

Re:This is logical next step

[Anonymous Coward]

We're in the States and have been planning a new home that includes Solar Panels. Suprisingly, the added cost of the PV Array would be easily offset by the cost of running Power lines where we plan on building so it's a net gain. Now that we're seeing them begin to bring out DC powered AC units, it makes it even more likely that we'll be building the new home within the next 24 months.

Part of our planning has included Solar water heating combined with hydronic baseboard heaters (same principals as the heater in your car) as we can easily get enough heat even on a cloudy day to keep things comfortable during the winter.

Re:This is logical next step

[Anonymous Coward]

I see this as a great thing for a cabin etc -- or .. my house. I'm in the USA and I'd love to cut down on the Electric bill in the summer as efficiently as possible -- and this is perfect for that exact application...

Re:This is logical next step

[Anonymous Coward]

People in USA and Europe with excellent grid connections are not aware of it.

WTF? How does grid connection affect what I know?

Re:This is logical next step

[Bing Tsher E]

But the affluent population of India is about the size of Japan, some 120 million people. They have been making do with truck-battery-inverter contraptions, small gasoline generator sets etc. They would probably form the wave of early adopters who pay for the early fixed costs of solar panel factories.

Anybody who understands the culture of India knows that the upper castes would see to it that somehow the spread of solar panels in India would be done in a way that insures the affluence would not spread far.

Installed base of AC

[sjbe]

I've long thought that whole local power grids would switch to DC eventually anyway.

Unlikely to happen any time soon. Too much installed base of AC power. Not like people are going to rip open their walls to switch from AC to DC and virtually everything you plug into the walls is designed with AC in mind. The only wide spread DC cabling standard is USB and that's mostly low power stuff.

I have no principled objection to DC power but I think any switch will take many decades if it happens at all.

Re:Installed base of AC

[Holi]

>The only wide spread DC cabling standard is USB and that's mostly low power stuff. more like RV and yachting equipment, it all runs off DC.

Re:Installed base of AC

[Anonymous Coward]

Don't forget PoE, the new new wiring plant

Re:Installed base of AC

[macs4all]

>The only wide spread DC cabling standard is USB and that's mostly low power stuff. more like RV and yachting equipment, it all runs off DC.

Yeah, where anything over a couple of Amps requires a cable as thick as your little-finger. And with the price of Copper, that sounds like a giant step in the wrong direction.

And I doubt you'll find any marine or RV applications for HV DC, for some pretty obvious (and deadly) reasons. And once you get past the consumer-level of marine and "RV", generation and distribution goes back to AC pretty damned fast.

Re:Installed base of AC

[cwsumner]

You don't need new wire, you just need new sockets so you don't plug into the wrong voltage. There are already DC socket standards.

But if you just wire the air conditioner to the solar panels, you still save a lot.

voltage

[Skapare]

what voltage do they expect? 12 volts? 48 volts? 100 volts? can i use my 2 AA cells?

Re:voltage

[freeze128]

It probably just runs off the power from a USB port.

What happens when they feed off the grid?

[acoustix]

So what happens when their home system craps out? Can the unit run when they pull AC power from the grid?

Re:What happens when they feed off the grid?

[Anonymous Coward]

All they'd have to do is include a AC-DC converter, which might be difficult for an HVAC system compared to your phone or otherwise, but it's not impossible.

As mentioned above, these units already are converting AC-DC, this is just having a DC-input and leaving that bit out.

Inverters on the panel

[ISoldat53]

The trend in solar is to put mini inverters on the panel itself giving them an AC output. This makes the setup and wiring of an array much easier and cheaper.

Don't get it..

[Anonymous Coward]

Isn't this post basically saying we are going to sell European AC's in the US?

inverter is in the AC instead of the house

[Anonymous Coward]

This is not really a no-inverter appliance. It's the difference between 1 and 2 inverters. It's possible to make an AC aircon with no inverter, using single-speed capacitor-start compressor, if you need normal rather than exceptional efficiency. It's also possible to make a DC aircon with no inverter, using a brushful motor, if you want low efficiency. But this particular aircon will contain a 3-phase motor and a dc-to-3phase inverter, which is more efficient than dc -> 1phase -> 3phase. In fact, dc->3phase inverter is cheaper than 1phase->3phase inverter, so if your aircon efficiency goals require using an inverter anyway then the DC aircon will be cheaper equipment. For solar, probably people will want the most efficient aircon because that's cheaper than more panel (the original discussion), so this aircon is both more efficient and cheaper to make than the likely alternative, but in both cases only slightly so, I think.

Good News Everyone!

[DarthVain]

I was literally thinking about this problem the other day.

I was day dreaming about the idyllic rustic life living in a cabin in the woods away from civilization etc... Then I got thinking, you know what? I would hate that, this week the temperature has been hovering around 32-38 degrees, which is hot. Sitting in a hot cabin all day doesn't sound so romantic, nor does hiding in a lake all day. AC is pretty much out of the question, as I'm pretty sure it would be too inefficient and drain whatever power you have in short order...

However DC AC (lol new Aussie band name?) might work better. Presumably when it is hottest out when you want AC, it will also be very sunny for your solar panels, which if more less feed directly into your AC might work out better.

Anyway it all sounds rather interesting... OK I'm off to start day dreaming about unrealistic rustic off-grid living again!

solar home appliances

[Anonymous Coward]

YES!
If they sold a washer/dryer WITH the solar panels just for them, it would be great!
If they sold a dual system waterheater WITH solar panels just for them, it would be great!
( dual= DC/Solar and 120/220 VAC for nights and cloudy/cold days )
repeat above statement for refrigerator.
DC/Solar for LED lights...and my electric toothbrush, beard trimmer, etc. - still might need DC storage...
sometimes a beard trimmer is needed after dark...

Nothing screams Green like

[kenh]

Solar-powered Air Conditioning!

Seriously, how large must the solar array be to provide a days worth of power for an air conditioner? Air conditioners are notorious power consumers, and I don't think their problem was that they ran on A/C not DC...

solar powered?

[serbanp]

There is no way to supply a high-power appliance (such as an AirCon) directly out of a PV array. None.

The panel is always operated at a point (MPPT) where it produces the most power, but that power is highly fluctuating whenever there's a slight obstruction in the incident sunlight (clouds, stray leafs, even passing birds).

You need a DC/DC power converter feeding a battery array. Then the DC appliance can be powered from that battery pack. Overall, an expensive solution mostly due to the need of local energy storage.

Re:solar powered?

[Agripa]

There is no way to supply a high-power appliance (such as an AirCon) directly out of a PV array. None.

The panel is always operated at a point (MPPT) where it produces the most power, but that power is highly fluctuating whenever there's a slight obstruction in the incident sunlight (clouds, stray leafs, even passing birds).

You need a DC/DC power converter feeding a battery array. Then the DC appliance can be powered from that battery pack. Overall, an expensive solution mostly due to the need of local energy storage.

This is true in the sense that nobody bothers to make appliances which can be run directly from solar arrays. Air conditioners which use a variable speed drive instead of on/off operation could adjust their load to match the maximum power point of the solar array.

(OT, Sorry)

[JazzLad]

I, for one, am disappointed to not see you modded :)

Edison has won...

[Anonymous Coward]

... after all.

LED/CFL and DC

[maestroX]

So, can anyone explain how this is going to affect LED lightning etc.?
I remember reading something about the changing load on AC due to the adoption of LED/CFL instead of bulbs (spike usage).
Will a central DC unit in my house run LED/CFL more efficiently?

off grid AC - sounds .. cool

[Anonymous Coward]

now , you need garden of solar panels and perhaps that nice big battery and you can magically convert heat to nice cold @ home.
Not worrying about electric bill

Wouldn't be all that useful to me...

[rthille]

I don't turn on the A/C until late in the day when i get home from work and the sun is already at a low angle. I suppose if I had a Tesla battery to go with my PV system it'd make a bit more sense...

sharp no longer producing solar panels

[Anonymous Coward]

this is great, except that sharp is pulling out of the solar panel market. at one time they made them domestically, then shifted production offshore. they still can't compete with asian manufacturers.

http://asia.nikkei.com/Business/Companies/Sharp-to-end-thin-solar-cell-production-focus-on-residential

Early Southern architecture

[jsrjsr]

More wealthy folks also used domes and towers with windows at the top for the thermal siphon effect. Some even had and gas flames near the windows to enhance the siphon. One of the This Old House episodes showed an example of the type with gas flames.

Re:Early Southern architecture

[Anonymous Coward]

The gas flames near the windows is a new one to me. I've only seen torch sconces in old missions and cathedrals and always assumed they were for lighting purposes. Thanks for the information.