I'm going to make you put on your thinking caps now. Suppose you were setting out to design an electrical system from scratch.

You have a completely free hand. You're not tied to using any type of connectors for compatibility, or using certain voltages because of existing equipment.

Yes, I know that's not very practical with the millions of installations already in use, but just pretend you were given the job of devising distribution systems for the first permanent Martian colony or something..!!!

What systems would you use for residential services? 3-wire at about the current 120/240V level? Would you prefer that even small appliances run on a higher voltage and maybe adopt a 3-wire 200/400 to 250/500 system so that each house only needed a 2-wire service for 200-250 volts?

If you were sticking with lower voltage, how about 3-ph 120/208 for residential? Would you rather see a British-type system where distribution is 3-ph 240/415V and houses just get one phase? Would you adopt some other standard supply voltage entirely: 175V? 300V? 2 What about commercial supplies?

Frequency: The current 50 or 60Hz standards or something different? Why? (I can't imagine anyone suggesting going back to a DC system, but if you want to, let's hear a good reason!)

Grounding arrangements: Would you like to stay with the current American system, or adopt something different? How about a separate ground wire back to the xfmr and the neutral grounded only at the xfmr? Or perhaps some other arrangement.

You get the idea. Any & all suggestions for devising the ideal system.

Well Paul,
how absolutley and uninhibitedly innovative you've allow us to be! To actually be so unrestrained of the all the powers that be leaves me somewhat at a loss here....
My choices are;
volatge
phase arangement
frequency
grounding configuration

My first concern and choice would be what interfaces best with alternatively contrived systems, and NOT with what the 'grid monger'junkies would have me do

For residential, I quite like the sound of a system they used in parts of Scandinavia (not sure if it still exists).
They had all 220-volt equipment fed from a 1-ph xfmr with a center-tap ground. In other words, very much like a U.S. residential arrangement but without any 120V circuits.
I'd keep what is now your neutral to the xfmr tap and use it solely as a grounding conductor.
As I see this proposed system;
Advantages: 220V power with only 110v to ground at any point. With no neutral as such, a N-G short couldn't go undetected; a short on either line to ground would trip a breaker. No need to worry about polarized plugs (equipment would have to be suitably designed, of course - Most already is, and many European countries don't have polarized plugs anyway).
Disadvantages: Cost of having double-pole breakers for every branch circuit. Awkward to do from a 3-ph distribution system. Any others?
What do you reckon?

[This message has been edited by pauluk (edited 10-02-2001).]

Well, the isolated or singular point of grounding at the X-former ( vs. our dual-usage N&G )would certainly end much squabble here.

A higher voltage would solve some V-drop problems, probably lessen conductor sizes a tad....

I can't figure a good reason for a different frequency???? Higher= more skin effect, lower= possible visual recognition.

Higher frequency = smaller transformers, but it would have to be much higher to make a significant difference (think of switch-mode power supplies), and then there'd be a whole load more problems.
50 or 60 Hz is probably about right. Or am I just being trapped by years of accepting this as the norm?
I'd like to see American/Continental type dedicated circuits for all major appliances, and I wouldn't use the British ring circuit. Too many complications.
.
P.S. Have you ever seen the German "Schuko" plug?

[This message has been edited by pauluk (edited 10-03-2001).]

One reason to dump 60HZ would be that it just happens to be in the range the really messes with the human heart. Higher frequencies, I understand, don't send the heart into ventricular fibrillation(<spelling?( ) as easily.
Nick

[

[This message has been edited by Nick (edited 10-03-2001).]

Good point Nick. I remember hearing that too, though I'd forgotten about it for the moment.

Do you happen to know what sort of frequency we'd need to go up to in order to get out of the range most likely to cause fibrillation?

the German "Schuko" plug?

nope, ain't seen one...but knowing thier engineering it probably works better than it sounds..

the V-fib thing sounds interesting.

Along the same lines,I like the idea of a GFI main for safety, but not for nuisance trips, etc. I would prefer some sort of monitoring/notification system, with shutdown time proportional to the level of ma lost.

My vote would be similar to what is used on aircraft and shipboard systems--400 Hz 3 phase.

Primary advantages:

Smaller, lighter transformers/motors.

No visible flicker from fluorescent lighting.

Higher ripple frequency means smaller, lower cost filtering components (chokes and capacitors) in AC/DC power supplies.

The above mentioned cardiac fibrilation issue.

3 phase power available to residences would allow the use of lower cost motors for things like furnace blowers and well pumps, as well as doing away with failure-prone start/run capacitors and centrifugal switches. Also makes for easy reversal of rotation.

The standard US 3 phase voltage of 208Y/120 seems fine to me, but the plugs/receptacles would have to change. The German Schuko would be a good starting point. I also like the British "fuse in the attachment plug" idea for appliance protection.

A GFCI main breaker set at around 100 mA would provide primary protection against "bolted" ground faults, with 20-30 mA units (in the breaker panel) handling individual branch circuits as needed for personnel protection. Individual single-phase branch circuits protected at 10, 20, or 30 amperes for lighting, general receptacles, or special purpose receptacles. Circuits for large appliances (ranges, A/C, etc. would be 3 phase.

Breaker panels would have a combination of gas tube and MOV surge protection right at the service entrance, where it can do some good.

No. I looked for info but came up empty so far. I can't remember where I heard this. I'll keep looking and if I find something I will post it.

>My vote would be similar to what is used on aircraft and shipboard systems--400 Hz 3 phase.
What is the voltage?

Sparky:
The "Schuko" plug is probably the most widely used type in Europe - not just in Germany but also Austria, The Netherlands, Denmark, Sweden, Finland, Norway, and many more.
Its a round plug with two round pins for line and neutral and fits into a recessed receptacle. The ground connection is made by strips on the side of the plug which mate with two springs in the receptacle.
Have a look here for some pics: http://www.king-cord.com/ourpro/europe.html http://www.powercords.co.uk/pc107.htm
The only possible drawback to the plug is that it is reversible, so on a system with a grounded conductor it's no good where polarity is important. For the balanced system mentioned above, that wouldn't be a problem of course.
NJ:
As someone who has had to design and build power supplies to give outputs like 12V @ 30A+, I certainly agree that a 400Hz frequency would make things much easier in the filter capacitor department.
I take the point about 3-ph making it easier for motors as well. Most residential services in Continental Europe are 3-phase already (220/380V or 230/400V), although it seems to be done more for balancing single-phase loads than anything else.

Main GFI: OK if it's a higher-rated or time-delayed type solely to protect the panel and any sub-feeders and all branch circuits then have their own more sensitive GFI.

The systems here where the only GFI is the main are a nuisance. One ground fault anywhere and the whole lot goes off!

DS:
I think there are several different voltages used at 400Hz. Some of the ferries operating from the U.K. to Europe certainly use 220/380V systems.


[This message has been edited by pauluk (edited 10-04-2001).]

Aircraft/shipboard power is generally 400Hz, 3 phase 200Y/115V. 28 VDC is very common for smaller loads, as well.

>Aircraft/shipboard power is generally 400 Hz, 3 phase 200Y/115V.

Do you know the maximum ampacity of a circuit?

What I was thinking of is that the skin-effect is going to limit conductors of 400 Hz to about 0.260" diameter which is 2 AWG. Heavier conductors would have to be tremendously derated.

2 AWG would limit circuits to around 125 A. So we would have to be allowed to parallel them. The idea of having many of these paralleled is unappealing.

Perhaps the final transformer could be placed right ahead of or after the meter?

Equipment drawing over 100 A would have to run from a higher voltage - which isn't really a bad trade off.


[This message has been edited by Dspark (edited 10-04-2001).]

What formula/method are you using to calculate the skin depth/conductor size?

AFAIK, skin effect only becomes a real issue at tens/hundreds of kilohertz or higher. At these frequencies, conductors are often tubular copper (transmitter tuning coils, etc.) or a specially fabricated (and expensive) conductor material called "Litz Wire", which is a woven conductor comprised of hundreds of separately insulated strands, joined at the ends.

>What formula/method are you using to calculate the skin depth/conductor size?
I would ask the same of you.

>AFAIK, skin effect only becomes a real issue at tens/hundreds of kilohertz or higher.
Not so.
At 60 Hz conductors are nearly maxed out where you see them end... 600 kcmils. You wouldn't want larger conductors since harmonic distortion has a swallower skin and can saturate the outside of a conductor.

Skin depth is inversely proportional to the square root of the frequency (7.75 * 7.75 = 60).

I use the simple formula 2.6 divided by the square root of the frequency. That gives the skin depth in inches. Multiply by two to get the diameter. Multiply by three to get the practical limit (>98% of current) - which works out to about a one inch diameter for pure copper.

But alas the discussion was about 400 Hz.
2.6 / 20 = 0.13. That's where I got 2 AWG. And since I note that only the outside surface of a conductor dissipates heat in long runs, the underutilized inner copper is of no use on continuous loads and larger conductors would required huge deratings.

This is still a lot thicker skin than the depths of just a few atoms seen when the frequency has a dozen digits rather than two or three.

Another way of thinking of it is that the skin at 60 Hz is only 1000 times thicker than at 60 MHz but 10,000 times thicker than at 6 GHz.

Simple!: Use Superconductive Elements!!!
I would promote the studies and experimentations of Superconduction Theories, then apply it to local - per occupancy power generation and systems.
Voltages and system types would be as described below under "Conventional" power systems.
Each occupancy would generate their own power, plus apply this to Superconductive circuit elements when possible.

Since that's simply a pipe dream / wishfull thinking, below is more understandable to actual human beings of planet earth - as opposed to mad quack scientists from other planets [like me ].

Here's my input for "Conventional" Power Systems

Residential:
200 VAC 1 phase 3 wire. Center tap grounded at transformer, plus grounded conductor from center tap brought to all services, bonded to local GES and all EGCs, but not an active circuit conductor.
Voltage to ground will be 100 VAC max.
Incandescent lighting would [should] be Quartz Halogen, or if normal Tungsten fillament / Argon filled envelopes are used, 4 position fillament supports.
This system might eliminate the problems revolving around common grounded conductors which are active in circuits [noise, open "neutrals", "hot grounds", overloaded commons, etc.].

For Commercial projects:
200Y VAC 3 phase 4 wire Wye [with 115 VAC to ground, but no L-G circuits used] for small occupancies and normal low voltage circuitry.

480Y277 VAC 3 phase 4 wire for larger occupancies, with SDS for 200 VAC systems.

For Industrial projects [low voltage only]:
200, 480 or 600 VAC 3 phase systems.
Grounding the system is still up in the air to me. Pros and Cons to the methods of either type [would like some input from others on this one! I have several ideas either way]. Systems point more towards the Ungrounded types - but will still be grounded at the transformer.

As for Frequency, I would lean towards 400Hz. The Skin Effect problems could be adjusted by designing conductors differently [multiple and separate tubular layers, Aluminum materials, Heat sinked exterior insulations, insulation with low Capacitive losses - possibly with an inert layer of pure water].

With higher Frequencies, the Efficencies [Q, core loss, etc.] of Induction type machines would benifit. Transformers, Reactor-Core Ballasts and Induction Motors fall in this category.

Transmission circuits will be effected the most. Multiple stacked parallel feeders would be one possible remedy - although the extremely large Induction problem with these "series" feeders would create choking and heat problems.

Possible to Transmit at 60Hz, then Invert to 400Hz - or Transmit as DC, then Invert to 400HZ. Either one will have Inverter losses and wastes [Transistor heat, Rectifier losses, etc.].

About the only system type I would really like to see changed would be the ones where untrained personnel can easilly mess with things! These would be Residential and small Commercial occupancies.

Frequencies would be better up higher for the end user, but remain low for transmission.

I still would like to dump everything in favor of Superconductive Elements!!! That's my idea of the ultimate systems!

Scott SET - the soapbox is now empty and available for the next speaker

Aw Man, I was gonna say something about Super Conductors earlier in this thread, but my ignorance of providing any detail to said system stopped me!

Do go on Scott!

(I hope y'all don't find it annoying that I edit almost every post, I'm a perfectionist sometimes...)

Q:

Assuming we can't make the entire system out of superconducting (i.e. "room temp") materials, where would the less exotic (sub 0ºC ) superconductors be best employed?

Liquid nitrogen in Xformers to cool them to superconducting temps (with appropriate materials)?

My "no holds barred" setup would begin with a Dyson Sphere\" ...



[This message has been edited by sparky66wv (edited 10-06-2001).]

if and when i do decide to build a house, i want to have my own private transformer vault.. id have the PoCo provide a primary loop to my vault, 3ø 13.2kv ohh yeah.. then id set it up with a large 100 kva 480Y/277 transformer bank. id have all kinds of amp meters and power quality meters.. really high tech.. for the house, id step it down to 120Y208, and id bring 480 in to my shop, where all my power tools (milling machines etc) would be 480, all the shop lighting would be 277 too. id also step it down for small tool circuits... im happy with 60HZ, though i would like to see all recepticles be of the Hubbell twist lock type, and for the bigger plugs, the hubbell ones with the pins.. the huge ones.. lol id also have a stand by generator.. oh what fun id have if i won the lotto...

-m

Wow, I seem to have started something here! Superconducting liquid-nitrogen cooled conductors, Dyson Spheres..... Anyone for a Flash Gordon style energy beam??!!

A little off the subject of an ideal system, but a few years ago I heard about someone who lived in the shadow of a powerful shortwave broadcast transmitter (it was in Switzerland if I recall correctly). He fitted out his roof with dozens of coils and got enouh energy to run all his lighting on HF power. I suppose it was ideal from the point of view that once installed the cost per kWh was zero.

I've always liked the idea of having my own private transformer(s), though I'm not sure I'd go as far as having a 13.2kV (or British 11kV) supply!

My reasons are twofold.

First, the xfmr would provide a certain degree of isolation from all the harmonics, switching transients and other hash present on the public supply, especially if used in conjunction with suitable filters. This is important to me because of the sensitive radio equipment etc. I use at home.

Second,
I could arrange my neutral & grounding system as I want it, not as is dictated by the local conditions. This is tied in slightly with the above. I want a low-impedance ground path, but our PME system which ties the ground to the neutral which can introduce some of the same hash into the house grounding system.

In general, we have big xfmrs which feed a large number of houses - the two xfmrs in my location feed about 300 homes via a 3-ph network.

As far as I could make out from studying lines & poles while over there, you seem to use a greater quantity of smaller xfmrs for a given number of homes.

How would you feel about extending this to one house per xfmr throughout? Greater capital cost for equipment I guess, and possibly higher losses overall due to the smaller xfmrs, but good in other respects.

Hey, if the whole lot is going to be fed from energy from a Dyson Sphere, we'd have plenty to spare anyway, right?!

here in miami, the power system is very diverse, with about an equal mix of overhead and underground... most older neighborhoods are overhead, 7200 to ground or 13,200 to ground (different areas different voltages) at my house, we have 3ø feeders (336 kcmil maybe larger) that dead end on out property line, we have a single 25KVA transformer feeding our house and the one across the street.. were the only ones on our block fed from the front.. all others are fed from poles in the rear, which have a single phase line and a transformer ever 2 or 3 poles, between the tx poles there runs a 120/240 "bus" which house services are tapped off of. in some neighborhoods, they use 50-100kva's to feed a lot of large houses, i have seen transformer condition sheets (we get them when we do secondary faults) and FPL loads a lot of them up to 150% or more of their rated capacity.. in underground neighborhoods, there are about 2 transformers per block (depends on block size) they are usually 75kva or 100 kva, each tx feeds around 8 or so houses.. miami is an interesting city power wise.. i think the rigging here looks nice, a lot nicer than in other places ive seen..

-m