Is There Still a Market for Mechanical High Power DC/AC Generators?

AI Thread Summary
The discussion explores the current relevance of mechanical high power DC/AC generators, particularly in light of advancements in semiconductor technology. Historically, motor-generator sets were essential for applications requiring variable frequency or pure DC outputs, but many fields have transitioned to solid-state solutions. Despite this shift, there is still interest in mechanical devices for generating high power DC, especially with potential upgrades to homopolar generators for increased efficiency. The conversation highlights challenges such as the maintenance of mechanical systems and the generational gap in knowledge as new engineers focus on modern techniques. Ultimately, while the market for mechanical generators may have diminished, specific applications still warrant their consideration.
artis
Messages
1,479
Reaction score
976
Hello, back in the day motor-generator sets were far more frequent before the advent of high power semiconductors etc. So they were used for many applications where a variable frequency high power was necessary or a purely DC output was needed. There were even some special generators that could be called "Faraday discs" as they worked on the homopolar generator/motor principle that gave a pure (no ripple) DC output for special purposes like large pipe welding machines or probably hydrogen generation due to electrolysis etc.

Now I am interested to know about the current situation as of now, some high energy fields have been covered purely by semiconductors while others have retained their non-solid state of affairs like high power RF for example.
How it is with high power DC (particularly where MA of current are needed) or variable frequency AC at high powers , is there still a market and need out there for a mechanical device which could meet the requirements?

I am asking this particularly with DC in mind as I thought about an upgrade that could be done to a homopolar/Faraday generator to make it close to 100% efficient (also have no fast degrading slip contact) which could generate mega amperes of current continually or in pulsed fashion.
 
Engineering news on Phys.org
There is still mechanical friction, and the maintenance of rotating machinery to overcome. Increasingly, there is a problem with familiarity as new graduates are schooled only in the most modern techniques.

Also, Moore's Law does not apply to mechanical machines. In the 1970s, ASEA sold a solid state frequency converter to the Itaipu dam in Brazil. The price tag was approximately $1 billion. It would be vastly less expensive today.
 
  • Like
Likes sophiecentaur
anorlunda said:
there is a problem with familiarity as new graduates are schooled only in the most modern techniques.
That not only applies to new graduates. It also applies to newly imported young bosses who will often re-invent the wheel as a result of an idea that 'came to them in the shower'. That idea was tried and failed ten years before.

It's a modern trend that encourages frequent job changes. It can benefit some individuals but organisations can suffer as the know how leaks away.
 
  • Like
Likes DaveE and anorlunda
@anorlunda I do have two questions wit regards to what you said, first of all why would a hydroelectric dam need a frequency converter? Do they not already produce the frequency of the grid matched at their generators just step it up through a step up transformer? At least that is how the ones I know work,

secondly with respect to More's law , I would tend to think that More's law also doesn't hold for power semiconductors as it does for the ones used for information processing. Because for power semiconductors one has to clear the device from the heat generated as well as have the minimum thickness for certain amperage and insulation for certain voltage. One cannot shrink the conduction region with respect to amperage I think.
Or did you think purely in terms of how much they cost?
 
  • Like
Likes DaveE and sophiecentaur
Thanks @artis. You gave me an excuse to spin an old tale. Here is the gossipy story I heard about Itaipu.

The Itaipu dam is on the Parana river which makes the border between Brazil and Paraguay. In the 1970s, it was the biggest hydro power dam in the world. (14 GW today).

Brazil (60 Hz) was growing and very thirsty for power. Paraguay (50 Hz) was a poor agrarian society. The total electric load in Paraguay was only a few MW. But by treaty, Paraguay owned 50% of the dam. The plan was to sell Paraguay's share of the power to Brazil. But then there was a ¿Quién es más macho? dispute between the Presidents of the two countries. Paraguay said, "Our 50% of the generators will be 50 Hz."

Q: How can you sell 7GW of 50 Hz power to a 60 Hz country? A: With a frequency converter.

Brazil hired my firm to design the transmission system. (which is how I got the gossip feed) Then they ordered the $1 billion frequency converter from ASEA. (That would be $5 billion in 2021 dollars) ASEA engineers posed holding up the SCRs to be used. Each was the size of a large pizza. [That is where Moore's Law comes in. Modern high-power SCRs and transistors are very much smaller than that, more effective, and much easier to manufacture.]

The converter was to be located right at the dam next to the generators. A monument to male stubbornness.

But before anything got built, diplomacy triumphed. Brazil agreed to convert all of Paraguay to 60 Hz, buying new appliances for every consumer in the country. Mexico did the same a few years before, converting to 60 Hz. [juicy gossip from that story too] Orders for half the generators were changed to 60 Hz. The order for the frequency converter was cancelled. My firm got paid to design the transmission system a second time with all 60 Hz. :smile:
 
  • Like
Likes DaveE, davenn, berkeman and 2 others
Well it would be very dumb and also inefficient to have a gen producing 50hz just to then convert that to 60hz.
@anorlunda How did those frequency converters work? Did they take in incoming AC and rectify it to DC and then made an arbitrary waveform purely out of switching a DC line with switches to form something resembling AC (step AC) which then gets attenuated to something more like real sine wave due to transmission line impedance? Which is the way I see HVDC substations make AC from the incoming DC line.
 
artis said:
Well it would be very dumb and also inefficient to have a gen producing 50hz just to then convert that to 60hz.
@anorlunda How did those frequency converters work? Did they take in incoming AC and rectify it to DC and then made an arbitrary waveform purely out of switching a DC line with switches to form something resembling AC (step AC) which then gets attenuated to something more like real sine wave due to transmission line impedance? Which is the way I see HVDC substations make AC from the incoming DC line.
Sorry, I don't know how they work in detail. Perhaps one of the others can explain it.

But it must be more than just a simple rectifier plus inverter. For one thing, the power into and out of the device is critically dependent on the phase angles of the voltages. For another, conservation of energy must be satisfied at every instant. That suggests to me that some kind of sub-cycle temporary energy storage is required.
 
  • Like
Likes DaveE
artis said:
Well it would be very dumb and also inefficient to have a gen producing 50hz just to then convert that to 60hz.
@anorlunda How did those frequency converters work? Did they take in incoming AC and rectify it to DC and then made an arbitrary waveform purely out of switching a DC line with switches to form something resembling AC (step AC) which then gets attenuated to something more like real sine wave due to transmission line impedance? Which is the way I see HVDC substations make AC from the incoming DC line.
Japan is a mixed 50Hz/60Hz country.
1631939282105.png

https://en.wikipedia.org/wiki/Shin-Shinano_Frequency_Converter
 
  • Like
  • Wow
Likes anorlunda, davenn and sophiecentaur
Back
Top