[GreenKeys] History - Why voltages andfrequencies are chosen(5v & 3.3v l...

[Keith Mc]

An update / report / summary

In this "issue" - mains voltages and frequencies
(I'll talk about Phonograph record speeds, "Optical Media", and other things later.)

Bottom line - The needs of early Edison light bulbs helped choose the U.S.
"mains voltage" (wall voltage).  The needs of White Appliances (stoves,
dryers, et al) fostered the creation of U.S. split-phase 220 (240).  
A trade-off between the needs of lighting and motors chose our frequency, 
to unify generation while minimizing eye flicker..

MANY cool stories are still coming in, regarding the WHY of engineering choices  
.. BTW... Some people have just said "see wikipedia" for one thing or 
another.  For speed PLEASE add URLs in your replies!  There are literally
millions of wikipedia pages, and they're not always well indexed!

Mains voltage and frequency - several great references:
... http://en.wikipedia.org/wiki/Mains_electricity#History_of_voltage_and_frequency
... http://en.wikipedia.org/wiki/Utility_frequency
For more info, also see "Electrical Energy, an Introduction":
... http://books.google.com/books?id=UokcachsYcYC&pg=PA29
(I'm skipping the "War of the Currents" - Edison's DC vs Tesla's AC...  Tesla won.)

Some of the highlights

Why "110V", and split phase 220 (now 120/240)?   
Short answer:  Edison's first carbon based light bulbs required approximately
100V to work.  So, 110V was chosen, to account for a bit of line drop.
This was carried over when DC went to AC (search for "the War of the Currents").

Cool Side Note - Production batches of early light bulbs were "selected" 
for voltage.  There even was some "secondary markets" for the off-value 
bulbs.  Local generators that could be altered in voltage could take 
advantage of the supply of them.

Voltage Creep (110->115->117->120V)  - Slowly increasing the voltage
over many years appears to be a cheap way to get more utility out of 
the same wiring as they added customers and loads.  Since the more
modern metal filament bulbs can be made at any voltage (and most
motor's speed is most often a matter of frequency, not voltage) 
this wasn't a big problem.

Split phase 220 (240)  
110V (now 120V) is rather low for power "white appliances" (stoves, dryers, etc.)
So a split phase 220 (now 240) was set up, with two legs of 110/120
that add together, for high energy appliances.

Why 230V in Europe?  Again, when METAL filament bulbs first came 
into existence, they could be made for ANY voltage.  So the forward 
looking German company AEG (descended from Edison's original 
European company) "bit the bullet" and converted their ENTIRE 
system to roughly double the voltage.  They helped their customer 
base change over their equipment.  In the end, it turned out that 
customer retrofit was more than offset by the savings in power line 
upgrades, as this halved their line currents.  Unification set the present 
European voltage standard.

What's interesting to note is that unlike split-phase 220/240 in US,
Europe is actually a 3-phase 400V system, giving you the 230V per leg.
3-phase power has a LOT of advantages (which I won't go into here),
and in fact is how the majority of our grid works too.  (Now WHY we 
don't supply it to residential homeowners is a whole 'nother subject!)

Advantage to Makers - This makes home owner/maker 3-phase power 
MUCH more available in Europe, than the US.   Even now, if you want 
3-phase in a US residential area, you often have to pay for a 
"special drop"  The utility runs new wires from your house all across 
your neighborhood to wherever it can be found. YOU pay for that, 
plus an additional, different local transformer, AND more poles should 
the present ones be at their limit for weight.... Not cheap!   

In the past, some US Makers created "3-phase rotating converters" 
out of old 3-phase motors, to save money vs "buying a drop". 
Today, the recent availability of cheap solid state Variable Frequency 
Drives [VFDs)] has changed all of that.  You can now easily create 
3-phase power at home of variable frequency to run a personal mill 
or lathe from 120/240 split-phase rather cheaply, without requiring 
an expensive 3-phase drop any more.  (You also get a speed control
out of the deal, so they are very popular.)

 - FREQUENCY CHOICES - 

Bottom line - There's a trade-off, between what is best for 
motors and transformers, and what is best for lighting.

In the early days, you had two different generating systems,
and they co-existed.

Very LOW frequency (eg 25Hz) WAS most convenient for motors.
It was also first generated by slower moving equipment.  
In fact, the lower the frequency, the better in SOME ways, as 
it minimizes certain kinds of losses.   Many motor systems ran in 
the 16-2/3 (1/3 of 50) to 25Hz range.  There are several low 
frequency systems still operating, for industrial customers 
(some Niagara Falls generators, some electric rail systems, etc).  

BUT, low frequency causes perceptible flicker in lighting.
So for LIGHTING apps, you are mainly driven by the needs of the EYE, 
and want HIGHER frequencies. Some generating systems for lighting
went all the way up to 140Hz.  

Having two separate frequency systems got expensive in infrastructure.
Rotating frequency converters COULD up the frequency from the 
"motor supply" 25Hz up to 60Hz.  But, that isn't nice mathematically.
The "common denominator" was large enough that it required very
large machines to do it.  (30->60 is easy had THAT been chosen, 
but 25Hz was entrenched. Hmmm...)

So... To make motor and lighting systems compatible without requiring
"frequency converters", you want to choose "the LOWEST frequency 
that eliminates perceptible flicker".  That means about 100-120 flashes/sec 
minimum, or 50-60Hz of full cycle waves.  (50Hz = 100 half cycles/sec, 
right around the flicker rate of MOST people's eyes,  60Hz = 120 half 
cycles/sec, a bit above most people's eye flicker rate.  Some Americans 
can notice the European 50Hz flicker when they travel there...)

50Hz vs 60Hz  - Lots of different stories abound.  But from what I've found:

In the early days, there were two BIG camps in play - Edison & 
Westinghouse (Tesla) in the west, and Germany's AEG in Europe.

Around 1900, there were lots of frequencies in operation, ranging from
16.667Hz up to 140 Hz in operation.   But the market eventually settled 
on two - 50Hz in Europe, and 60Hz in the west, so I'll only talk about them.

Big supplier AEG in Germany chose 50Hz, and Europe eventually went to it.

Some thought arc lighting looked a BIT better with 60Hz, and it worked 
better with clocks. 60Hz transformers were also SLIGHTLY lighter than 
50Hz transformers.  So, big supplier Westinghouse moved from higher 
frequency systems down to 60Hz.  Eventually, the west standardized 
on it, for compatibility.(eg Mexico changed over from 50Hz to 60Hz).

Interestingly, some places (like Japan) are still split, based on early purchases 
of equipment long ago from either US or Europe.  The western part of Japan 
(Kyoto and west) uses 60 Hz. The eastern part (Tokyo and east) uses 50 Hz. 
There is even a line of massive high voltage DC converters at the boundary 
line, trying to connect the two systems, and keep them compatible.

BUT... This is NOT STABLE, even today!

The US used to make sure the total count of cycles/day stuck very
close to 60Hz, so that AC driven wall clocks would stay stable.
But with the advent of GPS synced clocks, and "crystal controlled 
everything" (micros, etc), the US is playing around with dropping that 
requirement to make shuttling power from zone to zone easier to 
accomplish.  (One generator "leads" or "lags" another, to transfer load 
and power between them. But this can SLIGHTLY change the line 
frequency.  If you are "counting cycles" to keep time, this can become
a problem.)

A list member from Kuwait also wrote they are using 240/415VAC 
as the nominal single phase and three phase power line voltages.
But the local ministry isn't keeping them well adjusted.

He wrote:
> At home, we had up to 257VAC phase to neutral for a few weeks, 
> and I was going through incandescent lamps quite rapidly.   
> Line voltage is now much closer to the 240VAC nominal.
> 
> I can tell that the power ministry doesn’t do well on frequency 
> control, as the power line time base driven clocks in the stove 
> and the microwave oven aren’t very accurate.  The stove clock 
> is about 30 minutes ahead of real time now.

A few years back I too was once in a subdivision, that could jump from 
90V one day to 135V the next in the summertime.  My early Intel 
development system wouldn't boot at 90V, so I added a Variac on it 
to help "keep it in spec".  But one morning I forgot to check the line 
voltage before powering up the computer.  Though the weather was 
similar, line had made a HUGE jump up from the day before. 
Unfortunately, my Variac now compounded the problem.  
It ended up putting well over 150V into my computer, and fried some 
VERY expensive early disk controller boards.  That shut me down for 
several days (and cost me a big chunk of change, to fix...)  
Needless to say, my "morning rituals" were vastly improved after that!

So even today, things can vary widely from place to place.

I hope others found this as interesting as I did...  

BTW... I've also found info on "why 78, 45, and 33-1/3 for records" 
(and other similar trivia).  But this was long enough. I can post 
that information at another time.

- Keith Mc.