Will AC power distribution become obsolete?

[Andy Chee]

I likely won't live to see it, but I foresee a world of ubiquitous, safe, DC power in the future.

Case study: Japan

In case you weren't aware, Japan operates two AC grids, Western Japan runs at 60Hz and Eastern Japan runs at 50Hz (as a result of war rebuilding with electrical generating equipment sourced from USA & Europe, respectively).

The two grids have DC interconnectors, but they were inadequate to carry the load when the Fukushima nuclear plants went permanently offline, hence Western Japan has surplus energy, Eastern Japan had to ration power.

But think about why they didn't just agree on a single grid.  The logistics of swapping out existing infrastructure makes it unpalatable.

The same argument applies to why AC power grids will never become obsolete. 

[madires]

I can see it going obsolete for lighting, it just makes sense to run a lot of LEDs from a single DC supply, most likely 48V. No flicker and dimming would be excellent.

Absolutely, and it's actually an old idea from the era of halogen bulbs. Halogen bulbs are a bit more efficient than standard incandescent ones. So a few nerds went for a 24V DC distribution just for lighting. Fast forward to today and it makes even morse sense with LED lights, i.e. no need for a cheap PSU in each bulb. A simple resistor would be sufficient, also increasing the life span considerably.

[Xena E]

I can see it going obsolete for lighting, it just makes sense to run a lot of LEDs from a single DC supply, most likely 48V. No flicker and dimming would be excellent.

...no need for a cheap PSU in each bulb. A simple resistor would be sufficient, also increasing the life span considerably.

Agree, a simple system, but series dropper resistors are inefficient. The proportion of lost efficiency in a line voltage lamp with a constant current supply is much less, and it is possible to regulate the brightness better.

If on the other hand you self generate then I'd say its more efficient to use the power at whatever voltage it is produced or stored at, my system is 48Volts but the lamps themselves use switching constant current generators and multiple underrun LED's

(Except work lights where I prefer tungsten).

[David Hess]

My experience with electronic LED and CFL ballasts and electronically commutated motors have been all bad.

For instance the EPA now requires refrigerators to use electronically commutated motors in place of the shaded pole motors for the evaporator in refrigerators.  These motors cost 10 times more, and fail because of their sensitivity to power quality.  The shaded pole motors they replaced lasted decades.

So what was more wasteful?  I have to replace the evaporator motor in my refrigerator about every 2 years, at a cost of $30 plus my time.  Do you think a shaded pole motor, which will never fail, draws $30 worth of power over 2 years?  Of course not.  It is not even close.

LED and CFL bulbs with electronics ballasts have the same problem.  The operating life specifications of LED bulbs are bullshit based only on the operating life of the LEDs, when the electronic ballast is what will fail, either due to poor power quality or high temperatures, often in months.

DC power is not going to make that any better.  You might claim that ubiquitous DC to DC conversion will result in better power quality, but now it will also be the ubiquitous and expensive DC to DC converters which fail.  As a bonus, they will spew out RFI.

[jzx]

I wonder why they dont make the leds and the drivers separate units. (well, I can imagine a reason).

With fluorescent lamps, you have the ballast and the tube, if one of then failed, you changed only this one. You could buy good tubes and good ballasts because you have not to throw away both. Today the drivers are crap, because you have to dispose all  if the leds fail, and the leds are crap because you have to trhow all if the driver fails.

[Connecteur]

If I were to guess, it will probably come in stages.  The same high tension lines that carry AC could probably carry HVDC.  The transformers will have to be replaced with solid state converters.  Once in the home, it can be converted to a variety of useful voltages, probably not as hazardous as our current outlets.  I don't know which protocols will be established, but some will be for electronics and others will be for power appliances, I imagine.
Of course this is all a ways off, and probably beyond most of our lifetimes, but I definitely foresee the end of AC power.

The only place where 'grid scale' DC conversion does make sense is for international bidirectional import/export interconnects, typically sub-sea cables. These remove the need for absolute frequency and phase locking between the two countries. [EDIT: Oh, and offshore windfarms.]

About a mile from here there is a DC transmission line, running at 500,000 volts. The higher the voltage, the lower the amperage, which reduces power loss through heat...

In that case, I suspect that you live within a few miles of the coast. 500kV AC transmission lines are common BTW, nothing special about the voltage.

Perhaps such converters could incorporate some sort of inductive coupling to achieve isolation. It wouldn't need to create AC to do it, a pulsing current may be all that's needed.

I may not be an electrical engineer, but I don't need to be insulted for being an amateur.

[Zucca]

Since Jan 2024 I have an ESS Victron System installed in my home.
I designed it myself with Victron devices.

I am on grid about 10% of the time...

No, AC power distribution will not become obsolete, it will become a very well maintained, cheap ($12.5/month here) and 0dB silent emergency AC generator*.

*:I hope, I really hope...

[T3sl4co1l]

Perhaps such converters could incorporate some sort of inductive coupling to achieve isolation. It wouldn't need to create AC to do it, a pulsing current may be all that's needed.

I may not be an electrical engineer, but I don't need to be insulted for being an amateur.

Any change, is by definition, AC.

Tim

[Zucca]

Any change, is by definition, AC.

Tim

Any change = AC

🥶

[Connecteur]

Perhaps such converters could incorporate some sort of inductive coupling to achieve isolation. It wouldn't need to create AC to do it, a pulsing current may be all that's needed.

I may not be an electrical engineer, but I don't need to be insulted for being an amateur.

Any change, is by definition, AC.

Tim

I'm no engineer, but I think I know the difference between alternating current, and pulsing DC.

[Gyro]

If I were to guess, it will probably come in stages.  The same high tension lines that carry AC could probably carry HVDC.  The transformers will have to be replaced with solid state converters.  Once in the home, it can be converted to a variety of useful voltages, probably not as hazardous as our current outlets.  I don't know which protocols will be established, but some will be for electronics and others will be for power appliances, I imagine.
Of course this is all a ways off, and probably beyond most of our lifetimes, but I definitely foresee the end of AC power.

The only place where 'grid scale' DC conversion does make sense is for international bidirectional import/export interconnects, typically sub-sea cables. These remove the need for absolute frequency and phase locking between the two countries. [EDIT: Oh, and offshore windfarms.]

About a mile from here there is a DC transmission line, running at 500,000 volts. The higher the voltage, the lower the amperage, which reduces power loss through heat...

In that case, I suspect that you live within a few miles of the coast. 500kV AC transmission lines are common BTW, nothing special about the voltage.

Perhaps such converters could incorporate some sort of inductive coupling to achieve isolation. It wouldn't need to create AC to do it, a pulsing current may be all that's needed.

I don't understand. International DC interconnects are to isolate Frequency and phase differences (not galvanic). The only 'inductive coupling' that works with decent power efficiency is a transformer - that's what we have at the moment. A transformer won't work with pulsing DC, it will saturate and (for a grid scale transformer) explode.

I may not be an electrical engineer, but I don't need to be insulted for being an amateur.

Huh? I don't understand. The only things I put in my post were simple [Edit: non insulting] facts. The only DC interconnects shown in France in by Open Infrastructure Map ( https://openinframap.org/#5.35/48.43/1.731 ) are the ones from Tourbe converter station in Normandie and the Mandarins HVDC converter station near Calais, both of which run undersea to the UK, hence my "coast" reference. The 500kV DC transmission line that you referenced doesn't appear on the map, could you clarify the location?

[tooki]

I wonder why they dont make the leds and the drivers separate units. (well, I can imagine a reason).

With fluorescent lamps, you have the ballast and the tube, if one of then failed, you changed only this one. You could buy good tubes and good ballasts because you have not to throw away both. Today the drivers are crap, because you have to dispose all  if the leds fail, and the leds are crap because you have to trhow all if the driver fails.

Ummm... only for LED replacement lamps (i.e. things designed to replace an incandescent, halogen, or fluorescent lamp) and very cheap LED lighting. High-quality natively-LED lighting uses separate drivers and LED modules.

So basically, you ask "why don't they"... well they do, actually.

[ArdWar]

But DC power can now be transformed to higher or lower voltages without transformers.

Who said HVDC don't need transformer? How do you think the transmission voltages are converted? With huge transformers!



Even with "solid state" switching it still need huge transformers simply because shunting around huge amount of current can't be done at a too high frequency.

[tooki]

I can see it going obsolete for lighting, it just makes sense to run a lot of LEDs from a single DC supply, most likely 48V. No flicker and dimming would be excellent.

Absolutely, and it's actually an old idea from the era of halogen bulbs. Halogen bulbs are a bit more efficient than standard incandescent ones. So a few nerds went for a 24V DC distribution just for lighting. Fast forward to today and it makes even morse sense with LED lights, i.e. no need for a cheap PSU in each bulb. A simple resistor would be sufficient, also increasing the life span considerably.

LED lighting has been creeping up and up in voltage, since voltage drop becomes an issue on long runs. Since LEDs are natively tiny, and we basically just use tons of them to get the needed brightness, strip-form LED lighting (where we basically distribute the lighting) is extremely common, and voltage drop is a real problem across these strings. Originally they were 12V, now 24V is common, and higher is becoming more popular. Non-strip LEDs often use higher voltages already, often through a constant-current driver with a high compliance voltage.

[tooki]

But DC power can now be transformed to higher or lower voltages without transformers.

Who said HVDC don't need transformer? How do you think the transmission voltages are converted? With huge transformers!

(Attachment Link)

Even with "solid state" switching it still need huge transformers simply because sunting around huge amount of current can't be done at a too high frequency.

Oh, let's not pollute the waters with pesky things like facts! 

[AVGresponding]

I wonder why they dont make the leds and the drivers separate units. (well, I can imagine a reason).

With fluorescent lamps, you have the ballast and the tube, if one of then failed, you changed only this one. You could buy good tubes and good ballasts because you have not to throw away both. Today the drivers are crap, because you have to dispose all  if the leds fail, and the leds are crap because you have to trhow all if the driver fails.

Ummm... only for LED replacement lamps (i.e. things designed to replace an incandescent, halogen, or fluorescent lamp) and very cheap LED lighting. High-quality natively-LED lighting uses separate drivers and LED modules.

So basically, you ask "why don't they"... well they do, actually.

Indeed. They aren't any more reliable though, frustratingly (comparing good quality replacement lamps with good quality LED luminaires with separate drivers).

[David Hess]

I wonder why they dont make the leds and the drivers separate units. (well, I can imagine a reason).

It would not be cost effective, although halogen spot lighting worked that way.

With fluorescent lamps, you have the ballast and the tube, if one of then failed, you changed only this one. You could buy good tubes and good ballasts because you have not to throw away both. Today the drivers are crap, because you have to dispose all  if the leds fail, and the leds are crap because you have to trhow all if the driver fails.

Oddly enough I never had an electronic ballast for a linear fluorescent tube fail.  So the replaceable ballasts were the ones which never needed to be replaced.

[David Hess]

I don't understand. International DC interconnects are to isolate Frequency and phase differences (not galvanic). The only 'inductive coupling' that works with decent power efficiency is a transformer - that's what we have at the moment. A transformer won't work with pulsing DC, it will saturate and (for a grid scale transformer) explode.

Combination motor-generators were used for DC transformation and conversion at one time and can have reasonable efficiency.  They are still found in applications which require the highest isolation.  Mechanical losses from spinning in air can be reduced by sealing the unit and replacing the air with hydrogen, as in done with synchronous motors

Diesel electric locomotives and some large scale construction equipment uses this method even now to replace a mechanical or hydraulic transmission.

[T3sl4co1l]

Any change, is by definition, AC.

Tim

I'm no engineer, but I think I know the difference between alternating current, and pulsing DC.

Yeah, I'm sure you're right. Centuries of us engineers must be absolute dimwits. I expect your revolutionary theory of analysis will take the world by storm in the next couple of years and leave us old people, too stuck in our ways, behind in the proverbial dust.

Y'know, if only there was some way to check whether ones' opinion, on some objective topic, was correct...

Tim

[bdunham7]

I'm no engineer, but I think I know the difference between alternating current, and pulsing DC.

The distinction isn't simple and depends on what aspect you are looking at.  If you are concerned about galvanic corrosion or charging a battery, then there is a clear functional distinction between an AC current that actually reverses its polarity periodically and a varying DC that does not.  However, when you are talking of mathematical analysis or transformers, you break your "pulsed DC" into DC and AC components to analyze and describe it. 

[NiHaoMike]

So what was more wasteful?  I have to replace the evaporator motor in my refrigerator about every 2 years, at a cost of $30 plus my time.  Do you think a shaded pole motor, which will never fail, draws $30 worth of power over 2 years?  Of course not.  It is not even close.

Next time, look for a different brand of fan with similar specs. Or mod it to take a standard 120mm computer fan (and whatever voltage converter module if it's not a 12V supply), of which there are lots of cheap and reliable ones out there.

[soldar]

LED and CFL bulbs with electronics ballasts have the same problem.  The operating life specifications of LED bulbs are bullshit based only on the operating life of the LEDs, when the electronic ballast is what will fail, either due to poor power quality or high temperatures, often in months.

I have had many LED light fixtures fail because the LEDs fail because they are overdriven. At half the current they would last forever.

Same thing with a TV backlight I have here. Failed because they stress them.

[David Hess]

Next time, look for a different brand of fan with similar specs. Or mod it to take a standard 120mm computer fan (and whatever voltage converter module if it's not a 12V supply), of which there are lots of cheap and reliable ones out there.

The original fan motor was Swiss, and I replaced it with the same exact Swiss motor a couple times, and then I bought a Chinese one which has lasted the longest so far.  Since the newer EPA mandated motor is a direct replacement for the older shaded pole motor, I should be able to get the original shaded pole motor, but I do not know the part number.

But the whole EPA mandated fiasco really pisses me off.

[vk6zgo]

Perhaps such converters could incorporate some sort of inductive coupling to achieve isolation. It wouldn't need to create AC to do it, a pulsing current may be all that's needed.

I may not be an electrical engineer, but I don't need to be insulted for being an amateur.

Any change, is by definition, AC.

Tim

I'm no engineer, but I think I know the difference between alternating current, and pulsing DC.

Put it through a transformer & the different becomes blurred.
Have you asked yourself why SMPS have rectifiers on the transformer secondaries?

[schmitt trigger]

But the whole EPA mandated fiasco really pisses me off.

Do you have an estimate of how much power is “saved” between a shaded-pole motor vs the BLDC one?

I don’t understand the EPA mandate,  the shaded-pole motor, which is quite inefficient, could be replaced with a permanent split capacitor one. The efficiency gains may not be as large, but have proven ruggedness.