Naked Science Forum
Non Life Sciences => Technology => Topic started by: CliffordK on 22/02/2013 20:06:51
-
Ok, I've been looking at welders lately, and puzzling over the difference between transformers and inverters.
As I understand it, both are essentially a transformer, but the "transformers" run at 60 hz, and the "inverters" run at a much higher frequency, say 10,000 hz or so.
I think the issue is that say you consider a DC coil, it creates a static magnetic field, but doesn't induce current flow in another coil. The current flow is induced by the changing field. So, the higher the frequency, the more field changes, and better efficiency. I'm not finding the estimates, but inverters are somewhere around 10% to 20% more efficient despite the additional electronics required.
So...
Why don't we just use high frequency from the power companies.
Apparently the high frequency has greater transmission losses (http://en.wikipedia.org/wiki/Alternating_current#Effects_at_high_frequencies) due to the skin effect (http://en.wikipedia.org/wiki/Skin_effect), and perhaps capacitance so 10 KHZ would be impractical for utility distribution.
The reason we use AC in the first place is the ease to change the voltage with transformers, however this may have been largely historical, as a century ago most frequency converters used an inefficient rotary frequency converter requiring routine maintenance.
There are HVDC power lines (http://en.wikipedia.org/wiki/High-voltage_direct_current) in some places now with the advantages of lower resistance and capacitance that plague the AC.
Anyway, it would seem to me to be beneficial to start rebuilding the electric grid. Change all the local distribution grid to DC, and replace the telephone pole transformers with high frequency inverters. Even household current might benefit from low voltage (120/240) VDC, or less, although one would likely have to build micro-inverters for many appliances.
Gigawatt inverters are still somewhat complex, but the circuitry is quite simple on a small scale of a few watts up to say 100 Amps (10 KW) or so. Perhaps not as bomb-proof as the telephone pole transformers, but getting there. A single house inverter would tend to be reasonably easy to build.
How much power could be saved? Perhaps not that much on an individual basis. But, potentially significant amounts on for an urban or national grid.
-
There was an early debate in the USA over the use of DC (promoted by Edison) and AC (promoted by Westinghouse). This debate got nasty, with reports of Edison secretly funding the development of the electric chair - powered by AC! In the end, the AC system invented by Tesla won, due to the ability for transformers to generate high voltages for low-loss, long-distance transmission: http://en.wikipedia.org/wiki/War_of_Currents
Selecting the frequency of AC is a tradeoff:
- At lower frequencies, the iron magnetic core "saturates" easily, so you need much more iron and copper, ie larger and more expensive.
- At higher frequencies, the iron core has higher losses due to "eddy currents". This can be reduced by making the core in thin plates, but it is more expensive.
- Lower frequencies were mechanically better for rotating generators and motors
- Higher frequencies were better for driving high-voltage lighting (at or below 40Hz some people see flicker)
- US ended up on 60Hz, Europe & the rest of the world on 50Hz; this range of frequencies is suitable for both machines & lighting, and really quite efficient (about 90% efficient delivery from power station to home)
- some countries like Japan use both 50 & 60Hz in different parts of the country (which contributed to their blackouts following the tsunami)
- Minimum weight is critical in aircraft, and cost is less so; they tend to use 400Hz power
- http://en.wikipedia.org/wiki/Utility_frequency#History
Today, with cheap high-voltage semiconductors that can work at 100-400V, it is possible to use very compact, high-frequency power converters
- Many countries are introducing regulations for minimum efficiency of power supplies for computers, smartphones, cameras, etc. These new rules can only be met by electronic converters
- The electronic converters are smaller & lighter, which is a bonus for our modern mobile lifestyle
- The compact high-frequency transformers use ferrite cores (http://en.wikipedia.org/wiki/Ferrite_(magnet)#Soft_ferrites); Unlike iron, ferrites are electrical insulators, so they don't suffer from eddy currents
- An increasing number of homes are adding solar cells, electric cars, and even some fuel cells. These can all benefit from having an onsite inverter which can take power to or from the power grid.
- Probably the main efficiency benefits would come from coordinating these regional & local generation and storage systems via communications (a "smart grid (http://en.wikipedia.org/wiki/Smart_grid)").
Some challenges:
- It is hard to see how to change over the existing street wiring from AC to DC without major power disruptions
- Converting DC to AC is more complex than DC to DC
- But most of our appliances, and our electricity meters are all optimised for AC
- Transistors operating at 400V do not take kindly to lightning surges that can reach 1,000-10,000V
- In wet or salty conditions, DC conductors are more susceptible to electrochemical corrosion than AC conductors.
- The most likely short-term applications of DC are likely to be isolated network segments like undersea electricity transmission, interconnect between non-synchronised networks, solar cells and powering data centers.
-
Aside from skin-effects reducing the effective cross-section of large cables, you should also consider the wavelengths of higher AC frequencies... lamba = c/f so for 400Hz, l = 750km, or for 4kHz, l = 75km (actually reduce these by about 0.9 for the 'velocity factor' in cables). I reckon this has issues that long-distance transmission cables would start to act like antenna and radiate away significant power - but also you'd struggle to keep the country in phase with itself - which would be a grid-management nightmare. You'd also struggle to send power down long cables as the effective impedance would rise.
Higher frequencies (than 50-60Hz) also tend to be perceived to be much more audible (see A-weighting http://en.wikipedia.org/wiki/A-weighting ; @1kHz same amplitude sound will sound 30dB louder than 50Hz), so breakthrough into audio systems, or buzzing transformers, or metal next to high power cables, would make much more annoyance.
Substantial DC grid with electronic 'inverters' is all very well, but you have to consider reliability. Electronics reliability can be pretty good, but compare an inverter circuit-board with a hundred-odd components to a simple transformer with just a half-dozen connections, and I think the latter will prove to be far more reliable. To keep maintenance costs down (and customers happy) you need a very reliable network.
Don't forget that switches for DC use need to be much bigger due to the tendency of DC to arc when you break the circuit.
While I can imagine long-distance DC trunk cables becoming more prevalent, I can't imagine substantial moves toward more local DC distribution, or higher-frequency AC distribution.
-
My research indicated that long distance high frequency power transmission is impractical.
However, it wasn't quite as clear if local grids would benefit from higher frequencies. Many airplanes, of course, have 400 HZ AC power.
There are, of course, the big utility transformers that would either work at a higher frequency, or could be designed to work at a higher frequency. Every "wall wart", Telephone, TV, microwave, and computer also includes transformers. So, the losses are not only at the utility, but also at the end devices.
As it is, it is quite possible that one could build small transformers as inverters, and potentially save both power, as well as save on the cost of copper, potentially paying for the change. But, it would be even simpler to run them at a higher frequency if the utilities supplied high frequency power. Of course, vacuum cleaners and such might not like it.
-
The problem with running existing power transformers at higher frequencies is that eddy-current losses increase rapidly with frequency.
Present transformers are cost-optimised for operation at 50Hz or 60Hz, with an iron core made of thin sheets. An optimal design for higher frequencies would have much thinner sheets, but would need a lower mass of iron, overall. Running a 60Hz transformer at (say) 120Hz would increase the idle current, and become less efficient overall.
A transformer's maximum power output is limited by the resistive power dissipation in the windings. This is fairly independent of frequency, so you would not get any more power out of the transformer by operating it at a higher frequency.
Perhaps the simplest efficiency improvement for countries currently operating on 110VAC would be to install more appliances operating off 220VAC (ie opposite phases of the 110V). For an appliance requiring a fixed number of kiloWatts, doubling the voltage halves the current drawn, and reduces resistive losses by a factor of 4.
-
The American Three phase/Biphase system can be confusing to technicians not familiar with it as the voltages of the three phases (star configuration) as we know it are not all equal relative to the ground whereas the voltages of the three phases relative to each other (Delta) are.
-
While inductive losses, skin effect losses, radiative losses, and resistive losses are all lower with DC power transmission, there are losses in the rectification end and in the inverter end. Because of these losses, DC transmission is only an efficiency win over AC power transmission for power lines greater than about 300km. In addition to these advantages, DC transmission lines make it possible to intertie grids of different frequency, thus AC grids of 50Hz can be tied to AC grids of 60Hz via a DC intertie. Additionally phasing is no longer an issue. On some long distance AC transmission lines, their capacity is limited by sag. If the line warms up, it sags more, the sagging makes a longer path and causes a phase shift which then causes more current to flow between the two points that isn't going to a load, heating the line up even more. DC transmission eliminates that problem. Lines can carry more current because they aren't as limited by sag, and they can carry more voltage because when they are carrying AC, on average the line is only carrying .707 times the peak, with DC, it's carrying the peak voltage all the time. DC lines also eliminate the subjection of humans near to low frequency radiation which has been associated with a slight increase in leukemia risk. DC transmission lines are also immune to space weather. So there are many good reasons to use them but they make the most economic sense and give the greatest efficiency improvements if they are 300km or longer and thus it doesn't make so much sense to replace the local distribution system with DC lines.
-
Please provide peer reviewed evidence for leukaemia caused by low frequency ac fields I consider this just a myth spread by the greenies who don't like the view from their thatched cottages spoilt by pylons.
If there are any medical effects due to living close to pylons it is possibly due to an excess of Zinc in the environment dissolved from the Galvanised steel structures or Ozone generated by the high voltages.
-
I'm not convinced by "DC transmission lines are also immune to space weather."
I can see the fault scenarios are a bit different, but I suspect 'immunity' is rather strong.