« Power Water Networks | Main | How to Get Your Apartment Off the Grid »


Feed You can follow this conversation by subscribing to the comment feed for this post.



Another idea on how to reduce cable loss: distribute batteries to the high-power peak-load appliances and only distribute a low(er)-current charging current to those.

You probably use the 600W microwave a good dozen times a day for few minutes. Let's guess one hour of daily operation in total.
In peak you need 600W = 50A for a 12V system. Thus one hour @600W = 600Wh = 50Ah for a 12V system. Both is met by a typical small car battery.

To tricle-load this over a day a current of (a tad more than) 2A is enough - and easy to handle over common cable diameters.

Distribution electr(on)ics for such a decentralized can be quite simplistic, too: a germanium diode towards the decentral battery, a silicium diode for current going back to the "12"V charging grid grid. This way you'll have a 0.3V hysteresis preference for the local battery before all decentralized ones kick in and redistribute power to the depleted one. But a proper battery management electronics for each local battery probably is an even better solution...

Thom Westergren


That should be "Fewer solar panels . . ." Not "less" solar panels.

You'll look a lot smarter if you get your grammar right. Especially in your call outs.

Rufus Reno


It is continually refreshing to read thoroughly thoughtful articles. Refreshing to see the comprehensive scope. Nothing is particularly straightforward...

What about using a hydraulic accumulator as a "battery"?

Linda Foss


I'm so glad you wrote this & included the home design. I live in hurricane country. Hurricane Ike left me without power for 11 days. & here, it's usually very hot afterward. Luckily we got a cool front after a couple days.

Grid tied systems with micro-inverters have to detect power on the line or they shut down to keep from putting power on the lines & hurting workers. So you have no benefit from your solar in an outage. I know there are ways around this, but not approved by our electricity providers/municipalities.

So, since observing how my new flat screen TV is DC & all the audio equipment I want is USB, I've had the idea of having a grid-tied array for some AC & another for DC with storage to power some essentials/desirable like some lights, electronics, fans. (The fan is the most important thing!!!) & DC powered kitchen appliances depending on how reasonably they can be obtained. And i figured most of the DC power/wiring would be in some central location ie. the open kitchen which would share a wall with at least one bath.

And speaking of DC appliances & alternatives, I've been collecting examples. https://www.pinterest.com/betterways/better-appliances-alternative-ways/ There are solar thermal boosted mini split ACs for example. And DC brushless? motored fans are more efficient than AC which use as much power as the incandescent lightbulbs I no longer use. Cool cupboards can be used to store many foods so you have a smaller more efficient fridge. And the fridge can be kept closed at night. There's also solar thermal heat & hot water. Waste heat from the fridge or perhaps a dehumidifier.

I'd like to know who "we" is when you say in the next article "we" will.... I always thought this was a one-man operation.

Robin Wilson


My wife and I installed a grid tied solar PV system five years ago and our neighbors have a seven year old off grid PV system (see www.spectrumz.com ). Our neighbors with only 4 panel off grid do as your article suggests using DC power for most of their system. They make little boxes for making sure the DV voltage fits with the device which obviously requires seperate AC DC wiring. We have a micro grid so we can share power during low sun winter days or for power outages.

I disagree with your idea that off grid is cheaper. You didn't mention the price of the charge controller and batteries. In our grid tied system since our house is happily located in the woods so we had to run a line between our house and our PV panels 230 feet away in a clearing. As you point out high voltage loses less electricity so with the Enphase Microinverters, with a 25 year warentee on each solar panel we could send 220 V AC as opposed to 48 V DC to the house.

Off grid systems have no where to put the extra solar power in the summer so your panels and not fully utilized. Another advantage is grid tied systems qualify for SREC's which in West Virginia is only $20/MWh (coal lobby!) but with our total production to date is 23 MWh it helps. Enphase also gives online updates of your PV power production which I enjoy.

Off grid systems definitely encourage people to be more frugal in their electric consumption. Our neighbors use 2 to 3 KWh/day and we use 3 to 4 KWh/day. We share our laundry washer. We both like solar dried cloths hung on the line.



We've been using 48V DC in the Telecommunications and Data Center world for decades and we've pretty much addressed these issues. That ranges from the milliamps of 48V that power-over-ethernet uses to run your office VoIP phone to supplying high amperage to things like motors. You should check out our world before you let some newcomer vendor lock you into a proprietary voltage technology. There are tens of thousands of vendors in the 48v ecosystem who all basically inter-operate.



Excellent article. I can see 'possible' benefits if able to design from scratch, but if retrofitting not economic to replace an AC device with a DC one just for this. MY question is, how could you convert/bypass AC input to an existing device and connect directly to the DC side (safely)?

Nick Laing


This was a great article and I donated because it gave me an idea.

Thanks 'Low tech Magazine'.



Going from 12V to 48V would decrease loss by a factor of 16 (e.g. 4 squared), not by a factor of 9.

Sherwood Botsford


While there is some merit with this, the solution is not that clear cut:

My computer uses 5v and 12v internally. I have a digital picture frame that uses 9v. My iphone actually uses 3.3v but can be charged from a 5v USB port by throwing away 1/3 of the energy.

My irrigation system uses 18v. The backup sump pump battery charger uses 14v. My camera battery charger uses 4.5v

14 ga wiring carries 15A safely. The voltage doesn't matter. But a 12v line at 15A is only 180 watts.

what DC voltage to use? 48v has merit as having enough power to run stoves and air conditioners and microwaves -- even electric water heaters.

Lots of electronics could run on 3, 5, 12.

One of the issues with electronics is that they keep changing. Today's 5v device is tomorrow's antique.

Power supplies are getting smarter and more efficient.

To me it seems to make more sense to go after the big loads for DC. If you could power the fridge, air conditioner and water heater by DC, and use high efficiency converters for the leftovers it would make more sense.

kris de decker


@ Craig & Thom

Thanks, corrected.

@ Don

Laptops can be operated on DC if you buy a DC power cord. These are available for most models. For lighting, cut the plug and replace it by a 12V-plug, then replace the light bulb by a 12V bulb. Almost any lamp can be converted this way. More about 12V (24/48V) appliances in the next post.

@ Rufus

Thanks. Compared to electrical accumulators, hydraulic accumulators are very energy efficient. They have less than 2% charge and discharge losses, no self-discharge, and they can last for many decades. But you need some space. The typical 100-ton weight-loaded hydraulic accumulator from the early 20th century could store 2 kWh of power. Converted to electricity, at least 1 kWh of electricity would remain.



I would like to see an experimental ecovillage where all electricity is from a 12V DC microgrid. Cooking and baking is done with biomass (gasification/pyrolysis for clean burn, rather than combustion). Freezing is replaced with canning, fermentation and also food drying using superheated steam (see page I wrote at: http://opensourceecology.org/wiki/Food_Drying_with_Superheated_Steam ). A washing machine could also be biomass-powered for the heating part of it, with the mechanical part from another source.



The cable losses are limited by regulations.It is usually a single digit percentage of the voltage. And this is the main problem. Lets say 5% are allowed in a 12V DC system that only 0.6V in a 48V system its 2.4V. So if the current doubles the cross section has to be squared which at the end lots of copper has to be used. Copper is expensive whereas insulation material is cheap. and has to be used anyway.

Extra low voltage is less than 60V Dc which means a nominal system of 48 V would be a good trade off. If a one conductor is +60V and another is -60V there are 120V between but voltage against ground would be only 60V. However DC is more likely to generate arks as with AC contacts the spark will be extinguished every half period as the voltage is zero.

Also keep in mind that electrical power is usually generated in a three phase system. This is very efficient as in a typical 400V network the 400V is between to phases and 230V is between phase and neutral. The next higher voltage is 690V/400V. A 400V motor can be wired as wye at 690V or delta in 400V. Also three phase AC means that motors can be very simple without any brushes and they are maintenance free for decades.

A Swiss T25 Plug has the size of the Schuko which is used in most EU countries. But it can handle 3 Phase 400V 16A. It is much smaller and nicer than a CEE plug.
Modern motors are increasingly controlled by a variable frequency device which converts AC to DC and then back to AC. The losses for the whole conversions are less than 5 percent.

Dont forget that trams an underground trains are always powered by DC. Trains are much lighter as the don't have to carry an transformator. Control was also easier with DC. Todays AC locos use an AC/DC/AC conversion.

Also long distance power transmission is done by high voltage DC as there are some parasitic losses like skin effect and capacitance in AC networks which don't occur in DC.

There are also no issues with power factor are synchronize in DC which is beneficial for decentral power generation. DC use is definitely increasing. If higher power is needed using 3 phase AC is the most economical way.

At the end it is very likely that a power conversion must be made. The losses can be kept minimal by using a good power supply.

Herman Vanmunster


@Volker (#1)

Your idea of decentralized batteries is very original and would work from the technical point of view, but the solution you propose does not solve the problem. It only shifts the losses from the cable to the diodes and batteries.

A charge/discharge roundtrip has 70%..90% efficiency depending on the type of battery and of the way the battery is charged/discharged. Your Ge diode would also dissipate an additional 2A x 0.3V = 0.6 W all day long (if you charge 2A during 24h as you propose), so a loss of 0.6x24 = 14Wh for a total of 600Wh energy: an additional 2.4% loss (even more, because you would probably need to send 850Wh to the battery in order to be able to draw 600Wh from it).

Furthermore the Si diode in anti-parallel (going back to the home-grid) is a very bad idea. The system you propose will cause uncontrolled current-flows between the batteries in both directions (especially when an electric fault occurs or when overloading the circuit or when one batterie has an internal fault or...). You need to protect all the batteries and all the cables from overcurrent otherwise your batteries will explode sooner or later and/or your cables will get overheated, causing fire. As you suggested yourself: a proper energy management system would not only 'be better'. It is a must. And it is needed on every battery.

Sadly it is an all too common mistake to think that low voltage is safe. It is called 'safety voltage' because people can not get electrocuted. But as soon as medium power is involved (say >> 100 W), a proper protection against faults is needed just a badly as for today's AC system. Without proper protection, you system will cause fire sooner or later. I strongly advise anyone to get help from a qualified person before installing any power-system as soon as the power is >> 100W or as soon as you intend to connect energy-storage systems (peak-power) to the system.



As far as i can see most of the high load devices are essentially water heaters. For that problem you can use a cavitational water heater powered by a low load dc motor, which is according to youtube videos even faster than a traditional resistance heater. You can use one smaller for kitchen needs, and a larger one that feeds hot water to your shower and washing machine(which should be stripped off the heater, and at this point it is no more than a programmed motor hooked to a tub). If you want, you can automate the larger heater-reservoire with a few sensors and a dirt cheap single board computer, and then you can constantly have 100 liters of 80C. Even if you wouldn't ever heat the water to boiling point, i would still install a steam-release pressure valve just in case...



Speaking of static load storage/ water accumulators, this interesting idea, which is getting a pilot in Nevada, showed up on hackernews yesterday:


Essentially, they are investigating using a Roll-Under-Roll-Off cart system to store/retrieve energy by moving dense weights up and down a hill, with a pilot project in Nevada. When power is generating, the electrified carts pick up dense weights and move them up hill. When power needs to be retrieved, carts pick up weights and move them downhill. They are reporting 80% efficiency, so not quite as efficient as hydro=accumulators, but seems like it has a lot of potential for arid areas near elevation grades,

I also wonder if it would be feasible to combine with the rope transport concept which was discussed earlier in this blog, although perhaps nothing similar to the Roll-Under-Roll-On concept exists for rope transport due to the lowered interest since the 40s.

Herman Vanmunster


Hi Viktor,

Can you please explain how a cavitation water heater could reduce the electrical load? I can not imagine that anything could be more efficiënt than a resistor to produce heat? There are no rotating parts in a classical electric water heater. The resistor is inside the heater, so the efficiëncy is nearly 100% (not considering the way how electricity is produced). If you drive a (dc) motor, then electricity needs to be converted first to mechanical energy (by the motor) before it is further converted into heat by the cavitational propellor . The rotation of the motor causes losses (friction)outside the heater, so a certain amount of enegy is clearly lost. I really don't see how this could be more efficiënt than a resistor. Or am I missing something?

You need a certain fixed amount of energy (Joule) to raise the temperature of a certain amaount of water by 1 degree Celsius, no matter which source of energy you use. For water this is 4186 Joules for 1 liter of water for each degree Celsius. In the best case your dc-motor will still draw more current from the dc grid than a ristor for heating the same amount of water in the same time and for the same raise in temperature.

Paul Holden


I enjoyed the article, may I suggest that:
A "3-wire" 48-0-48 centre tapped dc supply of 48V to earth but 96V +ve to -ve would allow more flexibility for low and high power loads. Most old d.c. power distribution was always 3-wire for example 230-0-230 was common to provide 460V for industrial use and 230V for domestic use.
You can't escape conversion electronics, the output of solar panels have to be regulated to protect the storage batteries, motors need drives, even LED lamps have to be regulated even if only by an internal resistor, so there will always be conversion electronics and losses - the art is to design a scheme which minimises losses vs. acceptable cost.
Working in the railway business (as I do) "low" voltage dc is something of a curse, in a typical 750V traction system 30-35% of the power is lost to cable and rail resistance. If the voltage is raised to 1500V the losses drop to about 8%, and even lower at 3000V, there are applications where higher voltage is better. You might like to look up "Rene Thury" - modern IGBT Multi Module Converters ("MMC")are making Thury's pioneering work on d.c.power distribution a modern reality.



Hi, Herman,

You might be right. To be honest i've only seen youtube videos about these water heaters, never an actual one in real life, but the vids seemed quite impressive. Not an expert on these kind of things either, and i'm planning to cover all my heat-related energy needs from biomass that is readyly available from the land my family owns, and the house would still be on grid, as it makes more sense in my area to sell back solar energy to the company with a simple two-way meter, than spending loads of money on (partially) rewireing the house and spending more on batteries, charge controllers and other whatnots.



"Thus, no dishwasher, but doing the dishes by hand. No washing machine, but doing the laundry in a laundromat or with a manually operated machine. No tumble dryer, but a clothes line. No convenient and time-saving kitchen appliances like electric kettles, microwaves and coffee machines, but a traditional cooking stove operated by (bio)gas, a solar cooker, or a rocket stove. No vacuum cleaner, but a broom and a carpet-beater. No freezer, but fresh ingredients. No electric warm water boiler, but a solar boiler and a small wash at the sink if the sun doesn't shine. No electric car, but a bicycle."

You might as well admit solar power is a dismal failure and will never compete with other forms of power. Oh and I'm not living the pre-Industrial Age lifestyle to appease you environmentalist wackjobs.

Herman Vanmunster



I've visited a couple of sites. Some of them claim efficiëncies of over 100%, which is total nonsense [http://www.panacea-bocaf.org/cavitationheaters.htm]. They explain the >100% efficiëncy by unknown processes including nuclear enery. I would not want to use that! But there are other sites that seem to be more reliable. There I read that this type of heaters can be driven with any form of rotational energy. So apart from driving it with an alectrical motor, you could also drive it from other sources if no electricity is available. It appears to be safer than gas heaters and has a number of other advantages. But it is noisy! Anyway I had never heard of this type of heater before, so thanks a lot for sharing your idea!

I found this link among the reliable ones:
It's a download location to a pdf. It's a pity they did not calculate or measure the efficiëncy.


Herman Vanmunster


Hi Tim,

I totally disagree with your opinion. It is high time that we stop burning fossil-based fuels. We need to find valid alternatives quickly. Of course most people will not be willing to give up the luxury they know today, but we need to find solutions that either consume less energy, or that can be powered by renewable energy (or both: more efficiënt and driven by renewable power!).

Solar power is only part of the solution, it is not 'the' solution. But I would certainly not call it a failure. It's part the future. Solar panels will most probably play a leading role in dc home-grids in the near future. In fact, technology is ready for it. We only have to get rid of the ac-grid step by step. The conversion from ac to dc will not 'save the world'. It is only a small step towards more efficiëny and thus less losses/less consumption.



@ Victor and Hermann

Heat pumps are the most efficient way to use electricity for heating. Of course some low level heat source is need such as water or air. Usually its the best to generate heat either by solar thermal or by burning some kind of fuel.


As long nobody forces you to use solar power you just can ignore it. I have seen people which used an expensive AC inverter for their cabin or caravan. At the end they used the AC for LED light, laptops and TV. A DC system would have done the job and saved them a few hundred bucks. And in a truck store you can find many appliances that run from 24V DC.

kris de decker


@ Tim

You might as well admit that our way of life is a dismal failure that cannot be maintained without fossil fuels.

I'm typing this on a laptop that's powered by a DC off-grid system. It doesn't really feel like living in a pre-industrial age.

Tim Martin



Noble idea, but sounds like a nightmare to execute. Carve up the walls to bring in new wires? Some electronic devices operate on several dc voltages. I have a video projector that requires 5v, 15v, 24v, 85v all plus and minus, and plus 390vdc. One would have to be a very skilled technician to enable his existing ac powered devices to input one or several dc voltages. Warranties would be voided. Repair shops would find modified devices to be difficult or impossible to service.

Good luck

kris de decker


@ Tim Martin

I tape the wires against the wall :-)

If multiple systems are used and devices are clustered, cables can be very short.

The devices that I use the most (laptop and lights) are both running on 12V DC and are very easy to convert. An appliance that can't be converted, such as your video projecter, can always be operated with an inverter. This inverter can be much smaller and cheaper if it doesn't have to handle all power use in the building.

Boris Doderer


@ Herman and Viktor:

Concerning the comparison of efficiency of heating water by motion or "just" by adding heat, in a more general consideration:

You are right, that the amount of energy being necessary to heat a given amount of water from a given temperature to another given temperature, is always the same. But this concerns the energy REACHING the water.

To break down things to a simple explanation: Heat is always the result of motion, or molecular motion, and the correlated friction. The question is, where to create this motion.

The resistor in your water heater creates molecular motion, which must pass through several material layers (necessary for technical safety purposes) into the water. Every layer causes energy losses due to its thermal time behaviour (delay in heating up). Finally the water itself loses energy through the exterior wall of the heater.

If you produce the molecular motion (= heat) right where or at least closer to the place that you need it, there will be less energy losses.

A good example may be induction cooking (high frequency magnetic field increases molecular friction by idle currents directly inside the cooking pot material), or microwave cooking (the food itself is heated by molecular motion created by the electromagnetic radiation).

A cavitation heater also produces heat by internal motion and friction of the medium to be heated. Since the moving forces also stir the water around, the heat is not only created by bypassing several transition losses, but it is also distributed more evenly inside the water.

@ Kris:

Again an interesting article.

A former colleague (now retired) told me once, that there is acually no difference between DC and AC in transporting it by wire over a certain distance (in fact, accelerating electrons forth and back all the time in AC might even be more disadvantageous, due to feeding the parasitic capacitance over the length of the wires - even without load).

He mentioned the only advantage of AC in the vintage times was its transformability, so the voltage loss over long distances could be compensated by the transformer at the end of the line.

With todays progress in high energy/high voltage Solid State technology a compensation of DC losses by PWM shouldn't be a big problem, since the user's voltage is much lower than the one used for HV transportation.

Another question: You wrote that a hydraulic accumulator with a 100 ton weight would bring 1 kWh of stored energy after conversion to electrical use. Are you sure about that number? It's the equivalent of a small bathroom fan heater running far less than 1 hr ...

@ Matthias:

A modern washing machine (European Standard A+++) with heat pump is nice, but it trades its advantage in energy saving against a duration of up to 5 hrs for washing programmes that used to last 1.5 hrs with an old machine. Try to get three times washing on a Saturday done with that. I tend to consider things like that as being a nice mathematical trick. The German expression is "schönrechnen", which means "cosmetically recalculate". The user pays for it with his private time. I don't consider this being a real progress.

Not to mention the necessity of having lots of electronics and a heat pump system with refrigerant inside, which must be recycled after its lifetime somehow, somewhere on this planet, in a correct manner.

Don't get me wrong, I like heat pumps, too. But being an engineer myself, I see technologies nowadays being used in places, where there would have been more appropriate ways to do things. It just looks so conveniently "modern" to put them there. And it sells better.

kris de decker


@ Boris

Concerning the storage capacity of a hydraulic accumulator: I didn't measure it myself, the information comes from Ian McNeil's "Hydraulic Power" or B. Pugh's "The Hydraulic Age", see the sources below the article on power water networks. (I don't have the books at hand so I can't say for sure which of those is the source).



@ Boris Doderer

As you do not need to attend the laundry while in the machine the longer time do not need to be a problem. Run the machine every two days while you are working. For off grid systems low power and low energy consumption are the way to go. As there is less stress for the batteries.

The laundry machine could run during the day and consume the power generated by solar panels without storing in a battery first. In a system with a gas or diesel generator it might be better to use an older and faster washing machine and reduce the working hours of the generator as its efficiency is low in partial load.

With a heat pump system for heating a house, a kilowatt of electricity can get you many kilowatts of heat. And heat pumps can be used for cooling in summer too. There are systems which use the heat of the ground in winter to heat the house when there is not enough solar thermal energy. In summer excess heat from the solar thermal and from cooling the building is stored in the ground. In fact the underground is a seasonal heat storage.

Paul Holden


Kris, Boris

You wrote that a hydraulic accumulator with a 100 ton weight would bring 1 kWh of stored energy after conversion to electrical use.

E=mgh 100t x 1000kg x 9.81 x 6m = 5.9 MJ
1 kWh = 3.6 MJ therefore a 100 tonne accumulator with a 6m stroke contains 1.6 kWh. Allowing conversion efficiencies then 1kWh would seem about right.

Effectively the electricity industry does store electricity like this, using pumped storage.

Herman Vanmunster


Hi Boris,

Maybe we are a bit drifting off topic here, but we are both talking about different things. Your post applies to 'pass-through' heaters where cold water is heated instantly and immediatly used. In that case it is important to heat as fast as possible and as close as possible to where the heat is needed. You are also right about induction cooking. The faster the water is heated, the less losses there are because the cooker is not thermically isolated. So there is less time to lose heat.
But I am talking about a boiler where water is stored for several hours before being used. The resistor and its surrounding layers are placed in the center of the boiler. They are completely surrounded by water. So any 'loss' would eventually still leak to the water, where we want it. The heat has no other way to dissipate to than to the water.

If you drive a cavitation pump with electricity, then the losses of the motor are defintily lost because the motor is placed outside the boiler.

The situation is still different is you drive the pump directly from a locally available form of mechanical (kinetic) energy. In that case the cavitation pump will be by far a much more ecological choise than electricity. That is because electricity is still for the biggest part produced by converting heat into motion (in fossil fuel based power plants and in nuclear power plants). In this convertion 50% of the available energy is lost at the moment of production.

Boris Doderer


@ Paul

Thanks for recalculating. What I meant, was not just the way of energy being stored, but the thought of using the special technical application of a 100 ton weight for energy storage. I think we both agree that even a power water system covering a hole city means nothing compared to a pump-storage power plant. And as Kris already pointed out in the related article, power water networks are better for intermittent high power supply than for long-term usage.

@ Matthias

I usually don't attend my washing machine. If that was necessary, I'd need none. I also don't feed offgrid. I belong to the billions of people who still hang at the electrical grid.

What I meant was this:

Considering energy consumption of a washing machine, there are two modes where a machine needs exceptionally higher amounts of energy:

- For heating up the water at the beginning of the programme

- For the spinning cycle(s)

The energy consumption while spinning can be kept down by using higher efficient motors/transmission. It's done today by directly driving the washing drum with slow high torque motors instead of using high speed motors with belt drive.

The heating of the water is majorly done at the beginning of the programme. That's perhaps five minutes in the whole washing cycle. A heat pump as being used in a washing machine consists of several pipes, a compressor usually providing up to 20 bar of pressure for liquefaction of the refrigerant, the refrigerant being necessary for the operation, two heat exchangers, a waste water tank and a pump for flooding the water around.

The basic function of the heat pump in a washing machine is: The detergent solution of a previous wash is caught in the afore-mentioned wastewater tank. The warmth of this water will be extracted by cooling it down as far as freezing the water in the tank. When drained, the water from the current wash will then thaw this ice block and a part of it will go down the sewer.

Interesting question: How does the thawed part of that mixture seperate from the fresh, warm waste water, when going down the drain? And how warm is the waste water of the current wash AFTER having thawed the ice block it got its heat from? Greetings from the lessons of calorimetry.

What most people don't know about refrigeration/heat pump systems: They must remove their own heat (from the compression). This is a problem when being used for cooling, and a gain in case of usage for heating purposes. So, considering the not very high temperature of the waste solution from the previous wash, there's quite some amount of heat being added by the heat pump itself. The heat pump actually PRODUCES some of the heat by motion and friction (You read my remark to Viktor and Herman?). The only thing is: It doesn' provide so much of it.

So guess why the washing programmes in modern machines need so much longer?

It is the difference between efficiency and effectivity. The machines may be more efficient (energy), but they are less effective (time).

These are weaknesses of the system which must be compensated for. So, the machines are getting equipped with time control, programming features to make them convenient to use, etc. Things which must be produced and recycled later, which again costs ressources..

"Schönrechnen" - you remember?

I think you agree that a heating rod is much easier to built, to use and especially to dispose of/to recycle after end of product life. Considereing todays life cycles of 10 years or less ...

The washing temperatures today are much lower than in earlier times, since the detergents are constantly improving.

So, why not use a combination of the following:

- Use the hot water being available in every household (especially with solarthermics) to feed the machine. If you mix it up with cold water from the tap, you'll get the right temperature in most cases. It's efficient, since the hot water is there, anyway, and it's effective because the wash needs less time, (which again saves energy - double gain). You can still use a heating rod, but it will be there for correction purposes, only.

- Insulation of the machine chamber will help keep up the temperature (I may guess this is already done today).

- Especially the heat pump machines flood the water around. So why not skip the heat pump and use the heat of this water pump motor itself to keep the temperature up?

So the machine will be fed pre-heated water, need less time for the wash and keep the temperature up by its design.

Just a few ideas, but you see, there are many ways to do things. They just don't look so cool.

But I think, this is slowly but steadily drifting off topic, for it's got not so much to do with DC any more...

Sorry, Kris

Boris Doderer


@ Herman:

I agree. Nothing to add.

@ Rasmus:

This might still be off-topic, but in slight alteration of your wish:

Have you ever seen the videos about the solar test village in Tamera (combined use of solar radiation for solarthermics and a greenhouse, heat transport and storage by vegetable oil, mechanic and electrical energy produced by low temperature difference stirling engines)? The way they cook there is also very interesting.
Look for "solar village Tamera" and "sunpulse" at Youtube

X Darcan


I've been thinking that if neighbors would share their PV power and battery capacities the need to design systems for peak usage would be greatly reduced. The major problem that comes to mind is the cabling.

Even only 250 W PV panel and perhaps a 100 Ah battery per person would be enough, imagine each household had one of these packages for each individual and a thick DC cable or perhaps even a solid piece of metal would go from house to house connecting all the neighbors batteries and pv panels. If cables were not so wasteful, each person in a neighborhood would have access to all the capacities of panels and batteries in the neighborhood.

In a neighborhood of 100 individuals each would have access to 100x250=25 kw PV panels and 10 000 Ah battery capacity. If during the day the peak usage would be greatly supplied by the panels, and since the use of high power devices are mostly random during the day and ON for a short while, the design needs would be lowered. Disregarding devices that are continuously during the day of course.

Another idea I have is to make the PV panel makers to include a small simple voltage limiter that would prevent over charging of the batteries, perhaps adjustable or if the market could decide on a voltage that suits most type of batteries.

This would remove the need for a charge controller I think based on what I know. As said, DC 12/24 systems would also remove the need for an inverter

The charge controllers with high amps and inverters above 3kw are ridiculusly expensive and the prices increase in huge steps after 3kw. The PV panels are almost the cheapest components in the systems it seems and they have very long lifespans. On second glance it is more likely the other components that prevent people from buying them and not PV panels.

You may disregard the 100 Ah example for batteries per person, consider a higher number, yet if shared with neighbors the need to design individual systems for individual peak usage is eliminated.

Bruce MacLean


This is on the market and seems an interesting direction. Not cheap
DC high efficiency refrigerator/freezers are also available from these folks:

Paul Holden


To those that want to get rid of the a.c. grid, may I make a plea that it was us railway engineers that invented the grid in the first place as a means of transmitting electricty over long distances to supply electric trains. 132kV ac transmission was invented in the US around 1914 to enable hydro-electric power to be distributed over 100's of miles for electric trains. If you want to keep electric trains, the power has to get to them somehow.
"King of the Rails" from 1915 https://www.youtube.com/watch?v=o_LmA8_7oW0 shows one of the early schemes.

Using the 132 kV technology for a grid to connect fossil fuel fired power stations came later,and was done to reduce the amount of generating plant that ran at low loads for most of the day. In the UK prior to the 1930's 132 kV grid 7 GW of local power stations supplied typically 4GW of load.

One correspondent mentioned that long ac underground cables carry a lot of current just to serve the capacitance of the cable insulation, this is true and sets a practical limit to the length of underground cables. So some of the offshore windfarms are connected to shore by dc systems.

kris de decker


Lots of comments also at Resilience and Treehugger.





I would like to add some informations to this article.

1) It was possible to use HVDC back then, but efficiency was about 60-70% of conversion. in 1882 there was 2kV HVDC in Germany, In 1903 was local railway Tábor-Bechyně at Austria-Hungary electrified with voltage 1400-1500V DC. Between 1890-1910 many systems emerged with voltage between 6-60kV, usually under 4.5MW. Beside steam there was no powerful source of electricity back then.

2) Efficiency of power production in DC plant of 430kW was about 95-98%

3) Microwave oven require few kV of AC that is rectified and doubled. This power is around 1kW! Nothing suitable for low voltage DC.

5) Almost everything today with motor is "DC Ready" as universal motor that is in almost all 1 phase devices could be operated on AC or DC.

6) you have wrongly calculated looses in one part, it's not 10% + 15%, but 1 - (1 - 0.15)*(1 - 0.1), giving us 23.5% (Not much of difference, but more precise, in case of larger looses it could make huge difference.

--------MY OPINION--------
Any low voltage system is nonsense in case of buildings, as well each electronic device require different voltage, sometimes even few voltage levels, symmetric voltage... you will not eliminate all inverters. Sometimes this voltage change is done by dissipating power in regulator (78xx and 79xx integrated circuits), sometimes it require inverter in device.

Despite fact that up to 24V you don't have to have insulation etc., what you will save on plastics you will spend on copper. And there you go with main problem, choosing voltage, mobile phones tablets etc. - 5.5V. routers, normal things - 9-12V (due low power it could often be reduced by (7909, 7809)), tools - 12-24V. On other hand you have looses that require higher voltage to be eliminated.

With higher voltage you will have problems as DC burns everything it touches with sparks. Reserving DC only to low power devices will not help much as power generated by PVE and not stored or used would be lost, grid connected PVE, under certain conditions, could sell power to grid and reduce production of other power plants.

Fire could be started by any grid if conductors become too hot and are located near flammable things. As heat is generated by current, low voltage with higher current could start fire more easily, but for sparks it could be true. In case of let's say 24V system you would need 120A or more for peak load.

Your proposal is interesting, but I don't think it will be useful as you either would have to go for higher DC voltage of 48-100V in one "leg," or do double wiring (limits are 120W max. for 12V, 240W max. for 24V, 480W for 48V (multiply by 2 for two legs)), in houses of common people, will not happen. And even after that you will have to adjust voltage for many device.

Yes, going for some standard that is here is good idea, 48V seems as good compromise, if used as - 0 + we could get 960W with 10A. If it is compatible with PoE, it's even better.

TO: Sherwood Botsford
I agree. Those high power (and where is border? 5W 10W 24W?) devices would cause higher looses and often run independently on owner (fridge, air condition, heating) small devices might not be as important. But it's question whether use 48V in one leg or 2x24V thus 48V over both. If it would be 48V in one leg then we could go directly to 100V DC. But in light of what was stated by KJ it seems like anything not compatible with PoE

Paul Holden


Australian ABC program on battery powered homes




So you support home battery storage. And now Tesla and Mercedes, two automakers, are making them. But you oppose car ownership. The logic of automakers building storage systems is that eventually they will have to replace batteries in their cars because they will be down to 70% original capacity (although no Tesla batteries have reached that point yet), yet those can still be used for many other things. Like home battery storage systems. Your worst-case 2010 attack on EVs was based on purely coal-generated charging, which is rapidly becoming obsolete in parts of the world where EV demand is highest (which you failed to consider), and the embodied energy in producing batteries that you expected would be disposed of. So since you're already requiring a government powerful enough to outlaw automobiles, why not instead use that government to make people buy home solar and battery storage systems, and let them buy the EVs they want and thereby create the national supply of salvaged batteries that they can charge using their own home systems? It's not like the cost of new lithium ion batteries was falling over 10% per year. Oh wait...

kris de decker


@ super390

You should read my articles more carefully. My 2010 article was not "an attack on EVs" but an attack on a certain type of EV -- a vehicle that is as fast, heavy, and "smart" as today's gasoline powered cars.


See also the follow-up article on electric velomobiles:

I'm making the same argument for cars as I do for home electricity use or the internet: we can keep most of the technology if we adjust it to limited energy sources.



You'd be safe with up to 300 V DC wiring without increasing safety measures, I'd bet. As this engineer demonstrates [1], it's much harder to shock someone with DC than it is with AC. That's because of the capacitance of the skin tends to block DC (but not AC).

Great analysis by the way. Really thorough.

[1] https://www.youtube.com/watch?v=snk3C4m44SY



To make DC competitive a std would have to be set for the voltage of all the intended appliances envisioned. Heck just in my garage alone I have 12v, 18v drills, a 48v weed wacker, and a 24v sander. It gets worse from there.

In the telecom industry 48v was a designated standard for all central office machinery as the back up tech at the time was massive battery banks down in the basement. That makes implementation quite easy.

So that is the first hurdle to consider.



Success in a DC centric power system will not focus on production, but consumption. As one commenter has already observed the Telco industry has solved many of the issues of using 48v as a standard. Therein lies the crux, the mfrs of consuming devices (phones, appliances, computers, etc) do not conform to a standard power input paradigm. Its why wall warts abound even in AC systems, consuming devices requiring different voltage outputs.

Till standards are developed on the consuming side, a DC powered system would still need converters to meet the needs of the myraid of voltage demands of mfrs.

Good article otherwise.

Jim Pawley


First of all thank you for all the work. I really enjoyed this article on Slow Electricity. It is a topic I often include in my lectures on Responding to GW.

However, I was puzzled by this paragraph.

The energy loss in the cables equals the square of the current (in ampère), multiplied by the resistance (in ohm). The resistance is determined by the length, the diameter, and the conducting material of the cables. A copper wire with a cross section of 10 mm2, distributing 100 watts of power at 12 V (8.33 A) over a distance of 10 metres yields an acceptable energy loss of 3%. However, with a cable length of 50 metres, energy loss becomes 16%, and at a length of 100 metres, the energy loss adds up to 32% -- enough to negate the efficiency advantages of a DC grid even in the most optimistic scenario.

As I see it, US AWG #7 copper wire is close to 10mm2 in cross section and the resistance as listed at


and the resistance is .00163 ohms/m, (8.33)*2 x 10 x .00163 = about 1.1watt, or a bit more than 1% of than 1%, not the 3% you state. Of course, #7 wire is pretty thick and hard to work with (not to mention that it includes a LOT of copper that is very energy intensive to mine and purify), so perhaps you meant something close to AWG #12 this is 2.05 mm in diameter and 3.31 mm2 in area (and squared this would be about 10?) and the resistance is now .00521 ohms/m. This leads to about a 3.6watt loss or a 3.6% loss on a 100w load.

Personally, I see the the utility of a parallel battery buffered 12v or 24v DC network to run the low-power digital electronics and perhaps lighting and only available in a few rooms. Most of the PV power would still be inverted to AC for net "storage" and running high power equipment through wall sockets (Inverter efficiency continues to improve as does that of DC switching power supplies).

I don't' see ladies giving up their 1650w hairdryers, clothes dryers are often 6kw, and microwaves work at high voltages (though this may change as magnetrons are replaced by GHz power transistors.)

Keep up the good work.



Hackaday has featured a few things of this ilk recently too..


the comments aren't bad either.

and they link also to this project


Kolya Ivankov


Thanks for this really mind-provoking article. Let me add a personal comment.

I am an AI engineer. Since 2016 I've been working in rationalizing animal husbandry, and there I've learned pretty too well that high-tech is not always better. Just because one uses neural networks, it doesn't mean that the old technology won't perform better. Right now I am switching to the AI use in energy, primarily in wind turbines.

My overall opinion is that one has to switch to low-tech as soon as one can. That's precisely why I am reading and supporting your blog. My goal as a scientist is to figure out how AI may suggest low-tech algorithms and appliances there, where it could be done, not necessarily relying on a too sophisticated machinery. After all, I have a low-tech background as a mathematician, and in my math studies I've been doing "downshifting" theories - making theories with less assumptions.

That said, I am an immigrant, and, being quite underpaid in comparison to the people with my education and of my profession (people in agriculture were slow to understand the rules of software development, it was a reason for change), me and my wife had to live as close to off-grid as it gets.

We use public transportation only, despite of the fact that we need about two hours to go in one direction. A good part of the distance is covered by us on foot - if fact, I am walking about 12km every day. We almost never use the fridge, plugging it out altogether at the cooler times of the year and eating fresh. We don't use dishwasher, having just enough cups and plates for us two and washing them with hands. We don't have washing machine, instead I go to laundry every 3 weeks or so, after accumulating a large backpack of clothes to wash. In our flat, we've been using local warmth appliances to heat ourselves.We don't have a coffee machine whatsoever. We did use microwave, because...

Well because if you live a semi-nomadic life of an immigrant with residence permit bound to the job and the contracts never exceeding a year, sometimes lasting only several weeks, you can't allow yourself having too much things. I was living almost off the grid not because I wanted to, but because I've had to. The life has shown us that we can live like that. But do we wish to?

In the early to mid 90s Russia, in the time of my childhood, we were to poor to allow ourselves a vacuum cleaner or a wash-machine. I still remember too good the laundry stuck in the bath for days because the parents were too exhausted to do it. I've been using the same underpants for weeks sometimes. Same went for the dishes.

Our purchase of a wash-machine in 2000 was like starting a new life. We were finally able to put on fresh clothes, to sleep on fresh blankets, to actually buy new blankets. The machine is still operational in my parents' flat, by the way, just like the vacuum cleaner we've bought about the same time.

So, living as close as it gets to the standards you propagate in this article, I have to say one thing. I am out. As soon as I get a residence permit and hopefully a citizenship, that is, as long as I won't be forced to go to more and more orwellian Russia, I will buy a wash-machine. I will buy a normal fridge. And - finally - a dishwasher. And I'll be looking a place with floor heating, for I am tired of my and my wife's feet being constantly frozen - my current place obviously fell victim of "good energy passport". And I'll be seeking to buy a self-driving car for my wife, because she - like my mother, by the way - is simply not apt to drive a car herself, and yet I want her to be safe, healthy and in time. There are good chances I will take part in developing these cars.

Meanwhile, the community in rural Germany I've been working an draws 50% of its electricity of to wind turbines.

TS Gordon


I'm currently building my first 2Kw backwoods solar back-up system and addressing these efficiencies is of utmost importance, not to me but to the entire universe of housing and systems engineers and designers.

As was true back in the mid 80's, as the CEBUS council et. all took on the organizational task of standardizing LV Comm devices, (incl. RS232, 485, Ethernet, Gigabit 'Enet, USB, I,II,III, etc.,) ---WE, the end users and hobbyists, are now responsible for forcing the hand of BIG ENERGY to live up to their promises by showing real world acknowledgement of humanity's GLOBAL need to conserve, preserve and share access to all the available resources which so directly relate to the very notion of 'sustainability,' while creating all of the cultural disparities seen in our friend Kolya's message above!

Here's to the 5's, 7's, 9's 12's and 24's. May ye soon reign supreme!



It should be possible to run 'inverter' appliances on DC power, but I'm not clear if you could just feed them 240v (or is it 300v) DC, or if you'd need to bypass their AC/DC stages.

For example, it's much much cheaper to buy a regular inverter aircon, than a specialized DC 'solar' aircon, and I suspect the 'solar' aircon is basically the same device without the AC/DC stage.



I'm surprised that arcing is not mentioned. This is a significant disadvantage of DC power.

I think in the nearest future we will have AC power in our homes because:
* AC/DC converters get more efficient over the years. And cheaper.
* DC/AC inverters get cheaper and more efficient (97% is already achieved in some applications).
* AC switching will always be easier and cheaper than switching DC current.
* New advances in power electronics also relieve the traditional pain of AC power distribution: we will have solid-state transformers soon, HVDC will be cheaper soon.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been saved. Comments are moderated and will not appear until approved by the author. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Comments are moderated, and will not appear until the author has approved them.

Your Information

(Name is required. Email address will not be displayed with the comment.)