---------------- You would have to have the pole pigs as they are the most efficient (and cheapest) way to change voltage levels. In addition you would need inverter/rectifiers and filters. Couple this with the switching problems and the present AC distribution looks damn good- the reason that AC superceded DC still holds for most situations.
As for 240DC in homes- how much more are you willing to pay for fire or accident insurance?
I know of lots of people, especially in the more remote and desert regions of the US that generate their own power. I am assuming you have checked out homepower magazine, which has been available on the web for years.
Ok, I wasn't sure of your knowledge level. My background is an EE who works more with computers, but I've always had an interest in power systems.
Yes, there are DC Transmission Systems all over the world, some of them work quite long distances and some are in undersea cables. Wikipedia well tell you all about it.
The US even has a few. The one where I live (in Oregon) feeds power back and forth (at different times) from the Oregon side to the Columbia River to and from Los Angeles.
The convertor stations are big, expensive, and contain equipment that is exotic when compared to the typical AC transmissions stations. The used to use mercury arc rectifiers, but now I believe they are all solid state.
Back in Edison's day, there were no sockets, dynamoes, wire, insulation, fuses, switches or meters readily available for sale. They had to practically build everything from scratch.
Yet, they were successful and the first DC Central Station came on line in the crowded Pearl District area of New York City in 1882. The voltage was 110 volts and the maximum distance for customer service was about one mile.
Beyond that, you have the classic problem with DC. Even at this distance the mains had to use extremely large diameter wire... else the lights would dim out to zero output. One mile was about the maximum economical distance for a complete low voltage dc system and that is derived from the laws of physics.
Soon, many towns had "Central Station" service. If you can ever get a copy of one of the early trade magazines, "Electrical World", I believe it is called. It offeres a fascinating glimpse of this era.
This leads to the story of the "Battle of the Currents", AC vs. DC. Tesla, Westinghouse, and the development of the electric chair which I'm not going to attempt to re-tell here, but you can certainly read about it from other sources.
Bottom line, high voltage DC was (and still is) pretty much only good for long-distance transmission for limited point-to-point service. In the pre-electronic age, there was no economical and efficient way to convert DC high transmission voltages to a safe distribution voltage for every customer that needed it in the sprawling wide areas. . New York City offered DC for elevator power distribution to certain districts until fairly recently, I'm told, but, like a central station, most of the customers were clustered in one relatively small area.
What revolutionized AC was the invention of the transformer by Lucien Gaulard and John Gibbs in the early 1880's. Like the radio triode inventor Lee DeForest, these guys didn't really understand their invention very well and kept hooking them up incorrectly (in series) until other engineers stepped in to fix the technical problems. Like the birth of the electric chair, this itself is a fascinating story...
Do you live within a mile of the power plant? Then the answer is maybe yes? Would they be willing to install DC generators just for you?
Or. is your office a few miles from a coal mine or hydro plant? Then you could have point-to-point dc transmission at a higher voltage, but they might make you pay for all of the extra equipment needed.
Beachcomber
Just to emphasize that this was true for old TV sets, but all modern TV's have DC power supplies and stable oscillators. Those of you who remember the old vertical and horizontal sync adjustment knobs on your tv set know that this wasn't always true. Modern TV's have electronics that locks to the incoming video signal and not to the AC power line. Are you old enough to remember when sometimes they just didn't want to lock and the picture would roll?
The early TV's had "tank" oscillators that needed to synchronize with the incoming AC power line, else they would drift. Until the 3.58 Mhz color "burst" oscillator crystals became widely available, this was even more critical for the 1952 NTSC color TV system used in North America.
Beachcomber
On Thu, 09 Nov 2006 05:02:59 GMT Beachcomber wrote: | |>| |>| I think it comes down to a basic knowledge of physics. DC is coming |>| into play again because people are installing solar and wind power |>| systems. |>
|>And maybe I need to consider that direction given the not so great level |>of inverters I'm finding on the market. |>
| I know of lots of people, especially in the more remote and desert | regions of the US that generate their own power. I am assuming you | have checked out homepower magazine, which has been available on the | web for years. | |
One target load is a full size electric cooktop with dual oven. I want to rule out using gas on this (at least for now) because gas is a fossil fuel and it need needs a tie-in to an infrastructure (pipes or trucks) in most cases. Being able to do all my cooking on energy gathered renewably on the property is a goal. Generating hydrogen to do that is a possible other option to explore (and I would need to explore the viability and safety of using hydrogen as a cooking fuel).
| Though I live in a suburb and will probably be tied to the grid for at | least the next few year, I've enjoyed reading about the success | stories that people have had with the German "Sunny-Boy Inverters" and | others. Apparently, you can get into important but exotic subjects | like what is the impedance of your utilities power transformer from | the point of your service entrance.
If you want to sell power back to the grid, those might be a good choice. But even those models come relatively small.
My original thought was to have inverters of a size needed to run the cooking, and just duplicate that for other circuits, maybe 3 or 4 total power segments. But with the low fault current issue, now it looks like it might be better to gang them all together and target a very high end level of current, at least enough to trip any branch circuit breaker. The other extreme is to distribute DC and put inverters at point-of-use without concern for the faults.
But in any case, an inverter must be able to survive a bolted fault, even if it just shuts itself down.
|>So basically, it can be done for special cases, but it's not practical |>for the common cases (at least not yet). |>
| Ok, I wasn't sure of your knowledge level. My background is an EE who | works more with computers, but I've always had an interest in power | systems.
My background is CS but I took some EE classes, none of which were power systems. My interest in power has developed over the past few years based on sime issues that I've had with powering data centers. I found that if I learned the electrical codes, I could specify circuits for large numbers of computers without electricians coming back and saying something won't comply with codes. It has expanded from there.
| Back in Edison's day, there were no sockets, dynamoes, wire, | insulation, fuses, switches or meters readily available for sale. | They had to practically build everything from scratch.
There wasn't a lack of knowledge. Electricity had been around for 80 years by then, and electrical power had been commercially produced for many years before Edison got involved in it, some AC, some DC. What Edison did was make a big market for it. You couldn't just pick up some some insulated wire at the corner store, but it was available from more than one manufacturer. The science of the insulation was, though, still quite young.
| Yet, they were successful and the first DC Central Station came on | line in the crowded Pearl District area of New York City in 1882. The | voltage was 110 volts and the maximum distance for customer service | was about one mile. | | Beyond that, you have the classic problem with DC. Even at this | distance the mains had to use extremely large diameter wire... else | the lights would dim out to zero output. One mile was about the | maximum economical distance for a complete low voltage dc system and | that is derived from the laws of physics.
It was actually a 220 volt system, in terms of how far it could be run for a given amount of total power used.
| This leads to the story of the "Battle of the Currents", AC vs. DC. | Tesla, Westinghouse, and the development of the electric chair which | I'm not going to attempt to re-tell here, but you can certainly read | about it from other sources.
Read that several years ago.
| What revolutionized AC was the invention of the transformer by Lucien | Gaulard and John Gibbs in the early 1880's. Like the radio triode | inventor Lee DeForest, these guys didn't really understand their | invention very well and kept hooking them up incorrectly (in series) | until other engineers stepped in to fix the technical problems. Like | the birth of the electric chair, this itself is a fascinating story...
They effectively invented the current transformer :-)
Series, whether transformer connected or not, was a fundamental problem. It did solve certain problems that make it useful in certain situations even today. For example lights inaccessible to ordinary people can be powered in series at an overall rather high voltage to maintain each at the same current level. Parallel wiring leads to distant lights at a lower voltage, so you can't just tweak the voltage to get them at the level you want.
If you want to keep a string of lights at equal voltage over very long runs, you might consider the method posted about half-way down in this thread:
|>I have no such info. What I am wondering about is, if it would be possible |>or practical to distribute power that stays DC all the way from where it |>is generated to to the home/office. I guess the answers would be "perhaps |>and no". |>
| Do you live within a mile of the power plant? Then the answer is | maybe yes? Would they be willing to install DC generators just for | you?
I think that's a "low voltage" issue rather than a DC issue. If there is a _practical_ and _safe_ way to step DC up to high voltage and then step it back down, then "DC over one mile" is not an issue.
This is about low voltage ... and about DC not really being practical for common voltage changing (yet). DC itself is not inherintly the cause of this limitation; the fact that it can't be economically transformed is. Solve that tranformation issue and DC works.
| Or. is your office a few miles from a coal mine or hydro plant? Then | you could have point-to-point dc transmission at a higher voltage, but | they might make you pay for all of the extra equipment needed.
AC is still the practical means to transmit and distribute power. That may change some day in the future.
| b)Potential or actual frequency differences (even of a fraction of a | Hertz)between systems where an asynchronous tie is of benefit in terms of | system stability Some DC ties are back to back rectifier inverters because | of this (Japan-50-60Hz, Eel River New Brunswick Canada -first solid state | converters in use 60-60 Hz )
Why does a frequency conversion need to be AC-DC-AC? Why not just switch pulses directly out of each phase in an appropriate balance so that you are drawing power from each steadily? Afterall, the power-sum of all three phases is flat across complete cycles.
| Both of these are of concern in the Western DC link from Oregon south to | California and 4 corners region of US ( where rapid control of power | transfer aids in stability of the parallel AC system) and in ties from | Northern Quebec to New York State. | c) long underground or underwater lines - a rough rule of thumb is that 30 | miles of cable is equivalent to 300 miles of overhead line in terms of | capacitance and associated problems.
Looking at the transmission lines as an RF problem, going between overhead and underground or underwater cable, sure looks like a change in characteristic line impedance.
| 2)In addition, there is a problem with circuit breakers and switching. This | is far more costly and often beyond present technology for DC. With AC, | circuit breakers do not normally "break" the current but simply take | advantage of natural current zeros - preventing re-ignition of the arc. On a | small scale look at simple switches. Open a knife switch 1/4 inch on AC at | 120V, 10A and there is a small spark. Do it on DC under the same situation | and there will be a sustained arc flaring out a good half inch or more from | the contacts. (proven good for lighting cigarettes and if done carefully, | less dangerous than smoking the cigarette- I am a reformed smoker). | Why is this important? Simply put, a grid system depends on circuit breakers | and transformers. Hence DC for "grid" use is, technically and economically, | not a viable option except as indicated above under a)b)c).
This sounds like the real issue for DC.
Sounds like I need to get 10 big marine batteries, a bank of light bulbs, and a knife switch.
"Beachcomber" wrote
They still do supply DC in NYC. but I think they are trying to phase it out. As a first step:
----------- Great -try switching 1000A backed by a 500KV source at any part of the cycle? Try it- attempting to switch at any point in the cycle will mean that actual switching occurs at current zero which is not the same instant in each phase. In back to back, the AC-DC-AC works well. In long distance transmission the use of pulses doesn't eliminate the problems that favour DC transmission.
----------- True- but the main problem is that cables have a much higher capacitance/mile/phase and a corresponding increase in charging current and charging MVAR as well as increased Ferranti effect.
------------
---------- It is a very big issue in terms of "grid" flexibility.
|> | b)Potential or actual frequency differences (even of a fraction of a |> | Hertz)between systems where an asynchronous tie is of benefit in terms |> of |> | system stability Some DC ties are back to back rectifier inverters |> because |> | of this (Japan-50-60Hz, Eel River New Brunswick Canada -first solid |> state |> | converters in use 60-60 Hz ) |>
|> Why does a frequency conversion need to be AC-DC-AC? Why not just switch |> pulses directly out of each phase in an appropriate balance so that you |> are |> drawing power from each steadily? Afterall, the power-sum of all three |> phases is flat across complete cycles. | ----------- | Great -try switching 1000A backed by a 500KV source at any part of the | cycle? Try it- attempting to switch at any point in the cycle will mean that | actual switching occurs at current zero which is not the same instant in | each phase. In back to back, the AC-DC-AC works well. In long distance | transmission the use of pulses doesn't eliminate the problems that favour DC | transmission.
Sure, you can switch the AC at zero crossover easy enough. But what are they doing with the DC? Either the DC is a bunchof half-cycles at twice the incoming frequency if no filtering is used, or it is filtered and relatively flat, in which case the thyristors doing the switching won't have any zero crossovers to switch at anywhere. If they can switch that much power in DC, they can switch as much power in mid-cycle AC.
Whatever they can use to switch the DC with, I suggest switching the AC with that as described before, three phase 50 Hz to three phase 60 Hz or the reverse.
|> | Both of these are of concern in the Western DC link from Oregon south to |> | California and 4 corners region of US ( where rapid control of power |> | transfer aids in stability of the parallel AC system) and in ties from |> | Northern Quebec to New York State. |> | c) long underground or underwater lines - a rough rule of thumb is that |> 30 |> | miles of cable is equivalent to 300 miles of overhead line in terms of |> | capacitance and associated problems. |>
|> Looking at the transmission lines as an RF problem, going between overhead |> and underground or underwater cable, sure looks like a change in |> characteristic |> line impedance. | ----------- | True- but the main problem is that cables have a much higher | capacitance/mile/phase and a corresponding increase in charging current and | charging MVAR as well as increased Ferranti effect. | ------------
The far end of the transmission line changes its impedance different than the characteristic empedance of the line. It still looks like an RF problem ... on a very large scale.
Buffalo NY has these, or at least had them in the early '80s. What is a transformer with a "mechanically floating secondary windings", and why is such useful for that type of circuit? How does bypassing a blown bulb work, did they parallel each bulb with a device similar to a Zener diode that conducts if the voltage exceeds a certain level, keeping the voltage drop across it constant?
Of some interest, what they seem to use to power the lights was a setup on a pole with a large transformer can, a small can and something else about the size of the small can. I don't know what each component was.
Reading, Berks, UK had such a system. There were a number of series circuits, all driven from the basement of the old town hall, at least up to the 1970's.
The circuits are constant current, and the series streetlamp bulbs came in a few different power ratings intended for constant current operation. (I don't recall what current was used, but I think it was in the region of 6A.)
It was paralleled with a device which shorts out when a high voltage appears across it, just like series fairy lights. With the supply being a constant current supply, the remaining lamps continue running at their rated power, but the total voltage supplied to the series chain drops. Each chain has a volt meter on it, and you can look at them and tell how many lights are out on each chain by how much the chain voltage is reduced. You could also do this my monitoring the counter-weight position on the floating transformer windings, although I never actually saw the transformers driving the system in Reading.
Also available was control gear for 400W mercury vapour lamps to run from the series chains, and this was actually what was in use by the time I saw the system in Reading, although it might have started off with filament lamps. There's a lamp failure bypass mechanism for these too, but I can't now remember how it worked.
wrote in message news: snipped-for-privacy@news1.newsguy.com...
------- Thyristors turn off at current zero or when the next phase takes over. This is true for inverters as well as rectifiers-the current in the individual thyristors is AC but not sinusoidal. The total from a bridge is relatively smooth DC- better than a bunch of half cycles- with notches. (Turn on is not the problem as there inductance works for you to mitigate the rate of rise of current. Interrupting current by any means is more difficult.) * Note that there are no circuit breakers involved on the DC side. The thyristors can be used to interrupt at current zeros in case of a fault on the DC side. For the AC side breakers are satisfactory. However, for grid switching and system configuration changes, this is not satisfactory as the whole DC link must be shut down at all sources (inverters as well as rectifiers require an AC source for frequency and voltage control as well as reactive needs) rather than simply switching in or out a line as done with AC. There have been HVDC circuit breaker attempts in order to get the switching flexibility of the AC grid- many introduce a resonant circuit to cause oscillations and current zeros -then an AC breaker cuts in. Generally not practical. Note that the cost of filtering at a DC station is quite high (filters on both AC and DC sides. Now consider the increased cost of such filtering for a chopped wave- if you could chop it successfully. Also consider that such filtering would be needed everywhere you have such a link. Someday there may well be a satisfactory DC breaker but not today.
There are many references available on HVDC converters and their control.
-------------
--------- There is a relationship in that the telegraphers equations are valid but the operating regime is quite different. For example, a rather long 60Hz line without compensation may be about 1/8 of a wavelength. At that distance, an open circuited line will have a receiving end voltage which is about 40% higher than the sending end voltage and the charging MVAR level will be high and may cause other problems. To hold the voltage to a decent level, there has to be a reactive source at each end and often in the middle. Even at this length, there will be shunt reactors used and possibly series capacitors as well -simply to ensure a reasonable power transfer capability. Quebec Hydro's 735KV lines handle (old data) about 12 GW and require 25GVAR compensation. Hence there is a point where the AC line costs exceed the costs of going to DC.
For a 500KV underwater cable (near my home) of about 30 -40 miles, there are shunt reactors at each end and on an island in the middle. A 40 mile
500KV line is considered extremely short. In RF circuits, impedance matching is important as you are dealing with multiple wavelength lines as well as deliberate 1/4 wavelength lines In power systems these don't occur and there is no attempt to match impedances. Impedance mismatches are of importance when considering lightning switching surges (not that any attempt is made to do such matching) but not for 60Hz. There is a ballpark measure called surge impedance loading which is based on the characteristic impedance. For some shorter lines the loading can exceed this and for longer lines it must be considerably lower. No power line is operated anywhere near the maximum power point. - too inefficient and generally unstable.
The secondary coil was free to move on the core and was counterbalanced. If the current was too high , the coil moved away from the primary and the coupling weakened a bit. If low, it moved closer to the primary. Whether there is a core gap, I cannot recall. In effect it was a kind of constant current device. As to the bulbs- it was much simpler than a zener diode. If a bulb blew, then the total line voltage appeared across it and across a small disc in parallel. This would break down and act as a short circuit across the bulb. A similar thing is built into Xmas light bulbs. Cheap, simple, one shot.
I don't know what was in the small cans. one was probably a photo sensor and the other may have a contactor used to close the main circuit.
Don is of course correct, as usual, but there is an even older piece of the story. The 6.6 amp series circuit originated prior to incandescent lighting in the days of arc-lamps for street lighting. Arc lamps, with their negative dynamic resistance, can not be run in parallel. The series connection was necessary to force the lamps to run at equal currrents. When incandescent lamps became practical, a lamp with a 6.6 amp rating was designed to permit retro-fitting the arc systems without re-wiring them.
The cost/benefit ratio of series street lighting escapes us today, but it was absolutely correct at the time of its inception.
| As to the bulbs- it was much simpler than a zener diode. If a bulb blew, | then the total line voltage appeared across it and across a small disc in | parallel. This would break down and act as a short circuit across the bulb. | A similar thing is built into Xmas light bulbs. Cheap, simple, one shot.
All the Xmas lights I have seen are voltage sourced. That is, the whole series string is driven by a constant voltage. So if a burned out bulb results in a device shorting out with less resistance than the filament had, the whole current level will rise. At some point it will begin to cascade through all the bulbs. There would need to be a fixed resistance in each to prevent that from happening. And these are very cheap lights.
On Fri, 10 Nov 2006 20:27:47 -0500 BFoelsch wrote: | On Fri, 10 Nov 2006 07:09:11 GMT, "Don Kelly" wrote: | |>- |>At one time street lighting circuits were series. This required special |>transformers with mechanically floating secondary windings (rather exotic |>design but practical) to keep the current constant as well as overvoltage |>bypasses as in series Xmas tree lights which allow continued operation of |>the string as a whole if a lamp fails. |>Compared to parallel systems, there is no contest in terms of cost and |>safety. Note that present practice is to simply run the lights within a |>local area from a local supply so voltage drops are not a great |>concern-local photocells provide the control for the local lights. | | Don is of course correct, as usual, but there is an even older piece | of the story. The 6.6 amp series circuit originated prior to | incandescent lighting in the days of arc-lamps for street lighting. | Arc lamps, with their negative dynamic resistance, can not be run in | parallel. The series connection was necessary to force the lamps to | run at equal currrents. When incandescent lamps became practical, a | lamp with a 6.6 amp rating was designed to permit retro-fitting the | arc systems without re-wiring them. | | The cost/benefit ratio of series street lighting escapes us today, but | it was absolutely correct at the time of its inception.
What about using a series of HID lights at whatever current they need, controlled by a single current regulator (that can deal with a range of voltage)?
| Thyristors turn off at current zero or when the next phase takes over. This | is true for inverters as well as rectifiers-the current in the individual | thyristors is AC but not sinusoidal. The total from a bridge is relatively | smooth DC- better than a bunch of half cycles- with notches. (Turn on is not | the problem as there inductance works for you to mitigate the rate of rise | of current. Interrupting current by any means is more difficult.) * | Note that there are no circuit breakers involved on the DC side. The | thyristors can be used to interrupt at current zeros in case of a fault on | the DC side. For the AC side breakers are satisfactory. However, for grid | switching and system configuration changes, this is not satisfactory as the | whole DC link must be shut down at all sources (inverters as well as | rectifiers require an AC source for frequency and voltage control as well as | reactive needs) rather than simply switching in or out a line as done with | AC.
If you have DC coming in to a DC-to-AC system, how that system converting the DC to AC and creating the zeros if it doesn't cutout the DC? Is this with a resonant tank on the DC side of the thyristors?
overvoltage
Even simpler than that. Across each lamp is a pair of spring loaded contacts that touch each other and short circuit the lamp when they are in contact. Between the contacts is placed a prepared paper disc, about the size of a quarter. The paper has been partially nitrated so it resembles guncotton in its flammability. When the lamp is good, the voltage across the contacts is low and the paper disc prevents the flow of current. When the lamp fails, the entire supply voltage appears across the disc which is unable to withstand the high voltage. A spark results, which ignites the disc, which very quickly disappears on a flash of flame. The disc having vanished, the contacts touch, short circuiting the lamp. Due to the constant-current supply, there is no effect on the remaining lamps in the circuit. When the lamp is replaced a new disc is installed.
|> >At one time street lighting circuits were series. This required special |> >transformers with mechanically floating secondary windings (rather exotic |> >design but practical) to keep the current constant as well as | overvoltage |> >bypasses as in series Xmas tree lights which allow continued operation of |> >the string as a whole if a lamp fails. |>
|> Buffalo NY has these, or at least had them in the early '80s. What is a |> transformer with a "mechanically floating secondary windings", and why is |> such useful for that type of circuit? How does bypassing a blown bulb |> work, did they parallel each bulb with a device similar to a Zener diode |> that conducts if the voltage exceeds a certain level, keeping the voltage |> drop across it constant? | | Even simpler than that. Across each lamp is a pair of spring loaded contacts | that touch each other and short circuit the lamp when they are in contact. | Between the contacts is placed a prepared paper disc, about the size of a | quarter. The paper has been partially nitrated so it resembles guncotton in | its flammability. When the lamp is good, the voltage across the contacts is | low and the paper disc prevents the flow of current. When the lamp fails, | the entire supply voltage appears across the disc which is unable to | withstand the high voltage. A spark results, which ignites the disc, which | very quickly disappears on a flash of flame. The disc having vanished, the | contacts touch, short circuiting the lamp. Due to the constant-current | supply, there is no effect on the remaining lamps in the circuit. | When the lamp is replaced a new disc is installed.
That would work for the constant current series lights. But what about the Christmas light series strings?
They work the same way. There's a piece of wire wrapped around the lead-in wires which has a layer of thin insulation, which breaks down when mains voltage appears across the lamp.
The circuit is not constant current, so the loss of a lamp increases the power of the remaining lamps, and reduces their life. This can create a run-away effect as more die and the remaining ones burn at higher power. There's normally a fuse lamp in the set, which is a lamp which does not short out when it dies. This is to prevent a few lamps running at excessive power and igniting the Christmas tree, or the whole set shorting out. In the UK, we do get typically a couple of house fires a year caused by Christmas trees, and this maybe when people have replaced the fuse lamp with a shorting lamp.
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.