HVDC may be more efficient, but we have transmission lines running at a wide range of voltages from 11kV up to 400kV.  So there would have to be DC/DC converters to replace all the current AC/AC transformers.  Plus there would have to be a DC/AC inverter to replace every local substation to generate the 400/230V 50Hz that houses and businesses currently run on.  That's an awful lot of new infrastructure to build.  And hence a huge cost.  Meanwhile, transformers are nice simple things that can run for decades with no maintenance.


Why would you want to charge a battery only once a week?  Where batteries are currently used, it's to handle peak loads when the grid is under stress.  The battery would be recharged again as soon as demand (and wholesale price) drops back again.  A battery that could power a village for a week would be enormous and expensive.  And why bother when you have already installed the supply to the village?

Hi Simon , yes it would cost quite a lot of money , but it might pay for itself from the energy gains/ losses , 4% in MW terms is a bi deal , I think UK peaks and lows are about 20,000MW to 45,000MW , so that's something like 800 to 1600MW every hr , which is pretty much like a new power station , thrown in for free.

The energy use problem of efficiently generating power and low carbon thinking, wind and hydro are mostly in DC ,as is energy store in batteries (car or Grid scale), and water electrolysis are all DC systems and surely its best to optimise in DC before inversion into AC. We do have pretty efficient transformers and DC requires rectification and only relatively recently has the engineering made HVDC systems possible , siemens I think has managed 1000kv DC , but that was for a super hydro project .I think they use 400KV DC in the EU , so they must have efficient rectification devices, and I guess the cooled superconductor is pretty near (were at -60oC) so we really could have a system with much lower transmission losses, but as you say not much we can do past the interconnector as most domestic and business use is for AC . That's what worries about me about the car chargers being planned they are all  Dc to car battery but AC in feeds , so putting one in every house will add up ,, one calculation I saw put at least a 20% in crease  on MW required for the electric car to replace fossil fuels , so we really going to need an efficient transmission system sooner or later .

Until you do the costings, everything is just speculation.  You would need to work out what the losses are now (in cost terms) and what the losses will be after the upgrade.  You would need to know how much it will cost to do the upgrade (lots of money), and the effect on maintenance costs after you've done it.


I suspect the upgrade would never be cost-effective, and it would be cheaper to put up a few more windmills and live with the losses.

mmm not sure about that I mean from a costs point of view it beats windmills as its a continuous efficiency, or another way of viewing it is it makes every windmill feeding it seem to produce more energy . Transmission losses are well documented I think up to and interconnector and post transformation but it is quite a lot, but no denying efficiency it would attain over current system , depending on where you placed your new power stations.

Yes, renewable sources generally create DC, or if they create AC, it is not at anything like 50Hz. But the DC voltages generated are not very convenient, or even constant.


The current strategy of then inverting to a constant voltage AC, and then stepping up by transformer to something high for transmission is hard to beat.  Losses in a line carrying the same DC voltage are indeed slightly lower than the same line carrying AC - but that is the wrong comparison, you need to factor in the inversion losses. (and the reliability ) Unless you are under the sea so losses are too high, or the lines are transcontinental, so phase shift is an issue, then a higher AC  transmission voltage is the easier way out.


For your isolated village, if you have no inertial generation at all, (or indeed less than about 30% inertial) the inverters need to be signifiantly oversized, as you will not have the luxury of a frequency falling with (over) load, and the automatic load shedding action this gives when driving induction motors - this allows you to ride out short duration transients associated with inrush.

So the transmission network is designed in AC purely for motor powering ? so howcome most new netwroks are HVDC?

HVDC tends to be used for underwater cables, and really long cable runs.  For your average run between a new power station or wind farm and the National Grid, it's not worth the effort.

oh I agree at over 1000km other non electrical transmission systems start to be more efficient , but at under 1000km (remembering we now have much more design efficiency to think about with DC and moving and storing DC ) it should win out , for instance if you have a 100kw low head hydro , it has a higher availability than a 100kw wind turbine , both will generate in DC and if you want to store electricity as electricity , then its a DC battery , that is the most efficient electrical system , to feed say a small town , in turn the new demand for electric cars will be the biggest use of electricity from domestic supplies and the conversion losses will be important over time , so for a small village/town using a renewable it may be better if they charge there cars from the large battery . What I mean is with the electric car we will have very large use of DC electricity so it starts to change thinking on grid design and if we want to create Hydrogen from water electrolysis , this will also be using high amounts of DC electricity , for example a hydrogen car will do 400miles on 5kg of H2 , so at 60 kwh for 1 kg of hydrogen , 350 kw/hr (note conversion loss of electrolysis cell efficiency and efficiency of Hydrogen fuel cell, which are rarely quoted , which electric car companies see as the end game with Hydrogen as battery to electric motor is 80%), then a town of 1000 cars (doing 200 miles each) will need 175,000 kwh of new DC electricity each week , if a house uses 50 kwh of non car charging electricity each week , you can see that this new demand for the electric car is of a different order and well worth thinking about grids differently. Electrolysis DC electricity requirements are so high that efficiency becomes really important .

I'm not sure that most new networks are DC, or even a modest fraction. Are you thinking of some specific cases ?


Generation at DC tends to be at voltages that are too low - hundreds of volts for solar panels, and a few kV for wind turbines, and both vary variable with the weather.

This is not a suitable source for transmission over any distance more than perhaps a few hundred metres.

To step up a variable DC to a higher DC for transmission, perhaps a few hundred kV and stable in voltage to perhaps 5%  involves generating an AC first, so it is generally easier and more efficiant to stop at that stage, rather than to rectify, transmit and invert back to AC again.

Yes I agree the problem of lack of transformability is most cited obstacle , but my understanding came from an article on a big hydro project in china where siemens had built a 100kv HVDC system and it gave considerable savings , Most hydro I think generates in DC as does solar and wind so we are already losing some inversion losses , which I haven't seen any decent papers on for the UK although siemens has made a recent DC system for a wind array and that has worked really well in electrical efficiency improvement, The actual collection points for cables on DC wind arrays/cables has a device for "cleaning" all the mixed outputs into one. The efficiency is in the transmission itself , but looking to the future all battery and water electrolysis systems use DC and will be using a lot of DC , in some models if we make of Hydrogen , more than actual demands we measure at the moment , so if your taking about sending large amounts of power to different places in the country and you also have large regular battery use demands , you need a HVDC transmission system  , to the points off distribution/interconnector and then convert to AC . Large inverters have been developed to do this. Actually generating in DC is difficult , the gensets are usually larger and heavier than AC, and most electrical engineers give up on it , but I think it can be done if you use say a 6 or 12 pole genset , brushless 3 phase DC has already been done by a clever bit of shunt wiring , but not at say 500MW as everyone is stuck on AC for simplicity. The efficiency of HVDC is proven in transmission losses , the transforming or up stepping to HVDC from DC generators is not quite perfected , and I myself don't fully understand the equipment that is used , recently there was some talk of an idea by Nikoli Tesla on a transformer , which I think was called a Farday transformer, but no technical efficiencies were ever shown or proven. Once we have a large output DC generator it all fits together much better ,then I have asked GE about it , but they are a bit shy about discussing technology , but they are looking at HVDC transmission .