Thursday, June 24, 2021

Science Fiction, sort of

Posted with the title "Science Fiction" because I am going to invent things that don't exist and because I am going to suggest that people do stupid-dangerous least by today's understanding.

A proposal to increase the electrical distribution capacity at the retail level


The headlong rush into green energy has many concerned about the capacity of our electrical system. Very roughly speaking, the issues can be lumped into Generation, Transmission (say 50kV and higher) and Distribution (45kV and lower).

This work of science fiction deals with the Distribution.

One overwhelming factor regarding the Distribution system is that there are millions of miles of line and tens-of-millions of transformers and hundreds of millions of poles and billions of 120V/240V devices. The raw materials required to upgrade to support universal battery powered vehicles and electrical heat-pumps (in lieu of furnaces) is potentially staggering.

Minimizing impact:

Electrical power is measured in Watts. Watts are the in-phase product of the current and the voltage. It is possible to have the current and voltage out-of-phase. The out-of-phase parts are wasted as the current still generates heat even when it is not generating useful power. This propensity for certain kinds of load to shift the current and voltage comes into play a bit later.

There are three options for increasing energy delivered. Deliver the same Wattage over a longer period of time. This is a non-starter in a solar powered universe because the sun does not shine at night.

Increase the current. Current creates heat and necessitates more wire/higher gage wire.

Or we can increase voltage of the distribution lines. At this point, increasing the voltage appears to be the single best option.

Painting with a broad brush, the major changes that would be required to increase the voltage would be modifying the transformer(s) to step the higher distribution voltage down to the 120V/240V. 

Even a half-cycle (8-thousands of a second) of over-voltage is enough to fry semiconductor chips. While 0.008 seconds is a long time for computational semiconductors, it is a very short time for mechanical switching equipment. So one of the very critical dilemmas is: Do you shut down all customers on a leg or do you double wire and double transformer the poles and switch them over one-at-a-time?

Another issue that pops up is that 40kV is far more likely to start fires than 13kV. That is one reason why electric fence chargers are restricted to a maximum of 10kV. Before anything else would be done to up-voltage the leg, the poles must be inspected and any marginal poles must be replaced before more weight is hung on them and they are subjected to the forces of crews working near them and bumping them.

Regardless of whether you double-wire/double-transformer or choose some other option the utilities will need to refresh any poles that are the slightest bit dotty.

Size comparison between an 11kV insulator (left) and a 45kV insulator. Because of the length, 45kV insulators are often "hangers".

Then the insulators must be up-graded. 

Then the capacitors must be inspected to ensure they are 40kV capable. "What is a capacitor?" you ask.

Capacitors are a very short-duration energy storage device that are installed on distribution lines to counter-act the tendency of some kinds of loads (motors, for instance) to shift the current and the voltage.

None of what is written above qualifies as science fiction. It is background.

Science fiction parts

This proposal was inspired by the events that were described in the post about the project that went into the ditch.

The proposal is to rehang the 12.8kV lines (for instance) from 45kV capable insulators. Then to install a small second transformer capable of stepping the 40kV (or whatever the utility chooses) to the current voltage near every existing transformer. The auxiliary transformer will have its input and output wired to the line feeding from the distribution wire spaced about 12" apart.

Additionally, if the leg is long and difficult to access quickly, the line could be converted in segments if portable transformers of adequate capacity were tied into the line the same was as shown above.

They will do NOTHING until the link between the step-down transformers are severed.

The inventing/sci-fi part

This would be installed in the region circled in red in the previous illustration. It is laid out horizontally for my convenience.

Imagine that every place the linemen installed a step down transformer also had a device similar to the concept illustrated above. A couple of perf-charges set up just inside the lines to the 3:1ish step-down transformers. The charges are wired together to fire simultaneously.

Presumably they would be fired via a wireless arm-fire sequence and the length of wire and the hardware that held the device clipped to the wire would be propelled/drop into the catching basket where a sensor would verify that the feed wire had not just been severed but that a length of it had been removed.

This sequence would also include the closest large, portable transformer that segmented the distribution line.

At that point, the devices being fed by the 3:1 step-down transformer --> original transformer and the line downstream from the closest large portable transformer would be receiving approximately 40V.

If all the sensors registered that all segments had been appropriately severed, the feed could be stepped up from 12.8kV to 40kV. At that point, the overhead wires would be 40kV up to the first large-portable used to segment the task. The line downstream of the large-portable would still be 12.8kV.

The crew would inch-worm their way down the leg. If the sensors indicated a malfunction, they would boogie to the problem and fix it. It may simple be a case of the wire segment missing the basket. Once the crews had verified all issues were fixed the isolated segment could be flipped to the higher voltage. 

One would hope that the reliability of the perf-charge device would be high enough that manual intervention would be rare. In those cases, flipping from the 40V in-process voltage at the house back to the 120V/240V should be less than a second.

My understanding is that the 3:1 transformer can be relatively small because it is handling relatively low amperages. 


  1. In the words of of Commander Montgomery Scott. "The more you overtake the plumbing the easier it is to stop up the works".

  2. Interesting concept, problem is there aren't enough transformers or correctly sized wire to actually do this.

  3. The output load rating of each pole pig (13kV to 240V transformer) is another limitation. You can't double the capacity to the houses without doubling the number of pole pigs/end-stage connection circuits. It will probably be more efficient to install new circuits as 43KV to 240V from Day One instead of trying to rebuild the wings while the airplane is over the ocean.

    In new developments with all 200A house panels this probably isn't an issue, the low load density problem arises with older houses with much smaller entry breakers/panels. The other change is end-user solar systems reverse the expected power flow while the sun is out. My PV/Battery system runs a commanded 5KW discharge every day which is 5x my max load and probably takes care of the load for all the other houses on my 240V branch...

    MostlyCajun is a high power electrician/engineer and should have a lot more insight.

    1. Mostly Canjun is a very high-power electrician/engineer and he was one of the possible readers I pictured when I decided to call this science fiction.

      He might, for instance, say you cannot just have a 3:1 transformer with 12.8kV inputs (and grounds) without creating hellish current loops. If that is so, then some kind of switch needs to be installed on the auxiliary transformer. This is where my electrical limitations start to show.

      My take on the 200A vs 100A service is a little bit different. Most of the time, homes run very, very below the maximum. The problem will arise when every person plugs in their Prius at 4:35 PM and pegs the service. Incidentally, that is when they throw the pizza in the oven and drop the thermostat on the AC five degrees.

      Solar might be adequate, especially if the panels are in western Kansas and the load is in Illinoose.

      That is when the lines start sizzling.

  4. Via email:

    The process of growth has been going on since the beginnings of electric utilities.

    The original Edison 120/240 Volt DC system in downtown areas was overlaid with AC systems circa 1900, until some DC systems were shut down as late as 1950s.

    AC systems started out at 4800 Volts (Detroit Edison, others many different). Each Investor-owned utility or municipal system chose their own voltage. Many 3 phase systems ungrounded, so one conductor could be touched by a tree for example and not spark or interrupt service.

    When Rural Electric Administration came along in 1930s, serving distant, scattered customers, they chose higher primary voltage, typically 12kv to 13 kv, grounded system. If a conductor is touched by a tree in grounded system, fuse blows and circuit goes dark.

    As time passed and electrical demand increased, new substations were built to serve local load, fed by sub transmission voltage, typically 24 kv to 40 kv and higher.

    In areas of higher electrical demand, new substations were built for 13 kv distribution voltage, back in 1970s and continuing today. Local circuits reconstructed for 13 kv, eg to serve new subdivision, possibly underground. Away from substation, pole-top transformers dropped voltage to match older circuit to serve existing customers. For example, my subdivision was once served at 4800 volts from a substation about 4 miles away. When a new big box store was built about 2 miles away in another direction, a new substation was built at 13 kv. The service lines and equipment from the substation was reconfigured for 13 kv, and pole top transformers installed outside my subdivision to maintain service in my neighborhood at 4800 v.

    In summary, electric utilities saw demand double every 10 years through 1980s, so are well- versed with dealing with growth. Replacing poles, insulators, conductors, transformers, etc to upgrade capacity is standard practice, and costs are built into rate structures.

    The unknown is impact on charging of electric vehicles. Night time off peak charging is ideal, filling in demand curve with no impact on daytime demand. My Chev Volt at 8 amps had no impact, but time will tell.

    My one comment on current utility practice is what is called "Net metering". My neighbor just put a big solar system on his roof. Such customers sell electricity back to utility at retail rate while sun shines, then buy back at retail rates when no sun. I consider it a RIP off to the rest of us conventional customers, since the wholesale price of electricity is 1/2 to 1/3 the retail cost. Regulators and other do-gooders support such schemes as encouraging solar and other schemes, while the rest of us pay.

    -David Smith

    1. David, I have a solar/battery system and am on a Net Energy Metering Time of Use billing plan with So Cal Edison (so ymmv with your electric utility). Our peak rate power cost is $0.36 to 0.48 per KW-hr when I am powering the grid, buying back at night. Last full year my bill showed a $570 energy credit on net 466 KW-hr I provided to the grid. My actual bill credit was based on the $0.02577 per KW-hr, a $12.00 credit against my next month's $11.58 connection charge. Now, if I hadn't fully offset my demand I would have owed balance due payment. If SCE's practice is standard your neighbor won't get a huge check at the end of the year, If he is a net supplier he will see a small credit but not the hundreds of dollars.

      I admit I was confused and quite angry at the end of my first operating year when I got ~$30 credit for my 625 KW-hr of net generation.

  5. I went from ~$1,000 per year in electric bills to about $50.