Mains power voltage drop to reduce usage? (2023 Update)

No a split along the middle of the main island and a DC interlink between them (which turned out not to be high enough capacity after the Fukushima disaster took out so much plant on the NE Pacific coast).

The map of Japan is a bit erm.. schematic.

Kyoto went with Westinghouse US electric kit and Tokyo went with German generating kit so that both were present. This schism persisted after WWII when the country was occupied by (mostly) US and British forces.

More of the story here:

I think they probably could increase the low frequency bound for domestic and modest sized solar or wind farms not to drop off grid. It is preferable to stress a few motors and transformers for a few minutes when compared to the major cost of dropping large segments of load.

Otherwise you get the cascade failure mode that afflicted the UK after a freak lightning strike took out a fairly small electric plant and caused a cascade of failures that took down mains for London and much of England.

It didn't help that electric trains required an engineering reset before they could be moved from where they ground to a halt when the mains went down.

There is scope to modify the rules a bit so that losing 0.5Hz in grid frequency doesn't cause all local generation to drop off grid. Even in this rather bad case the generating companies got back to nominal conditions within 5 minutes but by then they had shed a lot of load.

WOW!

I had no idea.

Yikes. I wonder if any of the converters are motor-generators.

I have an ancient (1967) GE SCR manual, which is full of still-interesting circuits, like a frequency changing cycloconverter.

All the big new ones are state of the art converters by Toshiba I think. More info here:

But I wouldn't put it past them to still have running some arcane elctromechanical kit lovingly maintained by a national living treasure.

It always used to amuse me going into one particular world class semiconductor plant that at the right time of year there would be a rice farmer ploughing an adjacent field with two oxen and a classic plough.

I visited the HP minicomputer operation in Cupertino CA, well, long ago. The offices had big tinted windows and outside there were people on ladders picking oranges.

We went to the old mill in Maynard MA too. We picked PDP-11 over the klunky HP 16-bit thing.

I would expect 5/6 gearing to spin the motor/generators are the correct rpm.

I was only aware of SCRs hitting power conversion, DC/AC in the 80s?

Are you sure it was Oranges? Plums or Cherries would have been far more likely. Oranges grow much better further south.

They said they were oranges and they sure looked like oranges.

Some types of oranges and lemons grow well here, even in San Francisco, which is a lot colder than Cupertino.

The HP machine was basically a 16-bit PDP-8, a page-addressing machine with a couple of accumulators. HP also created a really slow, klunky hyper-CISC machine that wasn't successful. HP was stunningly not-creative when it came to computers.

None of the minicomputer companies survived, sort of like none of the tube or early semiconductor companies are still around, or at least still in the device business.

Intel is running on inertia now.

Or have a different number of poles. But the converters are mostly electronic now.

SCR based dc/ac converters were fun, when one SCR forgot to commute.

In commercial use (just) in Norway for Gotland Island by 1970. Prior to that they were using mercury arc converters. MOre detail here

And even more here: I can vaguely recall Ferranti having some R&D on it.

Yes, I have an orange tree in my back yard in San Jose. But, it's not optimal conditions as it does drop below freezing a few times each winter. Perhaps less so up the peninsula.

When I moved to Sunnyvale, there were still cherry orchards west of el camino.

HPIB interface was pretty good for scientific and engineering and their HP200 computers using 68k were OK. IBM PC's were very rough by comparison and their graphics truly abominable.

NEC9801 was pretty cool! HPIB and up to 1120 × 750 graphics in up to 16 colours (or 256 if you had wads of cash). Monitors with that resolution were ludicrously expensive back then. Our ARGS On the VAX was 512x512.

It was a long while before the 7220 was surpassed. I quite liked TI's

9918 video chip with support for hardware sprites also ahead of its time.

Their latest chips are still pretty good given the architecture. Pipeline stalls are much less frequent than they used to be.

sqrt on the i5-12600 somehow benchmarks faster than divide on mine! (it must be a quirk of how the thing pipelines instructions)

The smallest number is counted in pole pairs. 1 pole-pair is 3,000rpm. 5 pole-pairs is 600rpm, for the other side to have 6 pole-pairs for

60/50Hz conversion.

The motor will be 5 times larger than a 50Hz single pole-pair running at

3,000 rpm. 6 times larger than one running at 3,600 rpm. I can't see the economics working for using larger, less efficient and more expensive low-speed machines.

When they worked they were very good. I recall a small self-commutating converter in a TV. It was quite reliable.

The article mentions IGBT, an entirely different beast, and far better behaved than a SCR.

A better article for the case in point, it goes to show the differences between SCRs and IGBTs.

I suppose its friends would remind it very quickly, probably in a spectacular fashion. Peer pressure.

Jeroen Belleman

I think HPIB was abominable, but it was the only thing that worked at the time. Its only merit was that there were lots of electronic instruments that had that interface.

Has anyone read the original HPIB spec? There is a state diagram that will give you nightmares. And words that have to be translated into concepts like "logic high".

There was a hardware ack line in the cable. When SCPI began to be used over RS232 and Ethernet, ack was forgotten somehow. It's post and pray.

You meant "steal".

So they don't have enough. Oh dear. Canadia and Mexico should shut them off.