High power DC (AC) rotary generator

[Salvador12]

I know back in the day many places where high power was needed used a rotary machine whether DC or AC. This has been largely superseded by semiconductors and other solid state technology. But I do wonder though would there still be a need or a market for a high power DC generator for example which could output if needed in the megawatt range with controllable voltage and a pure DC output (no AC ripple) ?
Or for that matter a generator that can output both DC as well as AC with a variable frequency as needed. Where in the AC case the output frequency and waveform is not related to the shaft RPM but can be controlled separately like in an amplifier.

I tried google for various power supplies for stuff like water to hydrogen electrolysis or aluminum fabrication or others but I can't find what they use there nowadays.

 

[KlausST]

Hi,

I don't know any rotating DC generator that produces no ripple..

Klaus

 

[D.A.(Tony)Stewart]

It seems the most efficient hydro or turbine generators are AC, for some reason. But if you can push a Tesla Car in reverse , you might get DC

 

[FvM]

Or for that matter a generator that can output both DC as well as AC with a variable frequency as needed. Where in the AC case the output frequency and waveform is not related to the shaft RPM but can be controlled separately like in an amplifier.

They don't exist. There's no known electrical machine principle to achieve this.

But if you can push a Tesla Car in reverse , you might get DC

Modern electric cars use exclusively AC machines driven by inverters.

I understand that the original post is talking about rotating converters. They are legacy technology since some 10 years.

 

[Easy peasy]

Why would you want such a device ? but yes - the answer is such machines are still made because of the "ride-through" performance and high peak overload capability for short periods - often backed by solid state inverters.

 

[Salvador12]

Well I hate to break it to you guys but this time you are wrong, there is one machine that is known for being capable of creating pure DC without any ripple whatsoever , that would be the Faraday disc/homopolar generator where current is created not by induction but by continuous action of Lorentz force on the conducting electrons within the disc.

On the same token if you use an excitation coil supplied by AC current with the disc then the current output of the disc will resemble the AC waveform supplied to the excitation coil. The disc would then have to be slotted due to eddy current generation in changing B fields but still.
So if one can overcome the problem of the sliding contact wear and high resistance then one has a very low impedance , extremely high current generator capable of operating even under a short circuit condition (because the gen itself is almost a short circuit) so you can have DC as well as AC and can control the AC frequency and waveform.
The only thing I'm not sure about is whether there is a use for anything like that.

 

[FvM]

For a good reason, said Faraday disc appears in physics but not in electrical machines text books. It's of no practical relevance.

There'a long way from demonstrating a physical effect to designing a generator that can at least supply a few W.

 

[RCinFLA]

There are high voltage DC power transmission lines in western U.S. It feeds between the separate large area (unsynchronized) power grids.

 

[Easy peasy]

A commutated DC generator with minimal LC filter provides very good DC out with low ripple .... Salvador12 ...

 

[Salvador12]

@Easy peasy Ok I get your point but what if the homopolar gen can be made with a brushless non wearable contact that is of the same resistance as the generator itself, then one can have no ripple whatsoever and needs no commutation, just a liquid metal contact to take off the current from the rotating shaft.

 

[Easy peasy]

that's the difference between mouse power and real power ....

 

[BradtheRad]

Rotary motion and closely-machined parts are better suited to produce AC rather than DC.

Thomas Edison struggled early to promote his DC power. Equipment to generate and distribute it was cumbersome. On a visit to Menlo Park (Edison's famous facility), I saw a preserved sample of brushes used on his generators. It was made from several plates of flexible copper clipped together, becoming shaped by long wear against a current-carrying rotating shaft.

Seamless DC appears to come most efficiently from electro-chemical cells (batteries), photovoltaic, atmospheric differential, geologic, etc.

just a liquid metal contact to take off the current from the rotating shaft.

From olden days experimenters have constructed experimental motors using pools of mercury. The ingenuity they demonstrate is remarkable. Perhaps some have made DC generators to operate on similar principles.

 

[Kajunbee]

When I was reading through Joseph Henrys works he talked about the mercury switches. If I remember correctly they would have a layer of oil that covered the mercury. This allowed them to pull the wire out of the mercury without producing and arc. I will have to let the more educated members explain the importance of a clean disconnect with no arc.

 

[c_mitra]

This allowed them to pull the wire out of the mercury without producing and arc.

How does the oil prevent an arc being formed? The arc is the result of excessive high voltage that causes the local insulation to break down. I agree that the arcing can be reduced but it appears difficult to eliminate.

 

[Kajunbee]

Yes sir, and I’m sure you are correct. It most likely as you say only minimized the arcing. But I guess this was the best option they had at the time.

 

[D.A.(Tony)Stewart]

It seems the most efficient hydro or turbine generators are AC, for some reason. But if you can push a Tesla Car in reverse , you might get DC

I was joking about pushing a Tesla in Reverse to get AC

--- Updated Sep 22, 2021 ---

Mercury switches did have high surface tension with an inert gas fill so the make/break resulted in hysteresis from the surface tension, unlike solder flux from pine tree sap , rosin (sp?) with lowers surface tension rapidly and spreads out solder.

The old furnace thermostats would rotate the angle of the glass tube between fixed electrodes and would attract rapidly on contact and break rapidly on disconnect, thus no bounce . Then some had a preheater to raise the temperature to reduce the hysteresis.

I used them in the late 70's with 16 ch mercury switches HP mux cards for high voltage A/D mux because they had no bounce for 128 conversions /second.in a loop to 16x8 channels. I suppose the inert gas supports the low arc flash. High power contactors use a vacuum which also raises the BDV/mm

 

[Kajunbee]

 

[FvM]

Except for conventional DC machines, which are still used in some applications as EasyPeasy showed, most of the stuff discussed in this thread would better fit a (not yet existing) Technical Museum forum.

--- Updated Sep 24, 2021 ---

Also mercury wetted reed relays are in production, but very remotely related to the original topic.

 

[KlausST]

Hi,,

a word about pictures and their file size.
edaboard is a worldwide forum. Many places in the world still suffer from low internet bandwidth.
Thus one should take care not to use unecessary huge pictures/files.

In post#17 there are two picutres with rather low quality but huge file size. They were was about 6MBytes in total.
Now I modified the post to see some low file size preview pictures.
If interested press on the picture to see it in full size.

***
Another way is to reduce pixel rate and color quality:
In below picres I reduced from 2400 pixel (in X) to 800 pixels and stored it as JPG with 30% quality grey scale.
now the total file size is 0.083MByte -> 1.4% file size or 70:1 file size.

Klaus

 

[D.A.(Tony)Stewart]

For Museum purposes...

Joseph Henry was born in 1797, the pioneer professor of Induction. He mentions oil in the article, but oil is never used today. This would contaminate the interface with oxidants (insulators) from the bi-products of wetting of the plasma gas of the arc to the conductor interface and breaking down the oil into many different hydrocarbons depending on the activation energy.

Plasma is the 4th state of matter that combines solid, fluid, and gas.
Here is IBM's 1977 mercury switch patent (with no oil)

It may be interesting to some that “wetting” of most dry contacts rated <=2 Amps is not required, if those are based on the modern practice of gold plating over some type of metal alloy. But anything rated > 2A will not be gold plated which would burn off immediately . However, non-gold plated contacts must have small arcs of wetting current at least 10% of the rated current to burn off the oxide insulation. But no more, for 1 million cycle life, as this just burns off carbon insulation and more will burn the silver-copper-palladium-nickle-steel combination “metal alloy” contacts and reduces switch life. The more energy in the arc, the shorter the cycle life.

I suspect the term wetting the contacts of the air-metal surface comes from the analogy of liquid mercury and the property of plasma.

I learned this from a failed experience 45 yrs ago when I designed a box with 100 power relays for sounding rocket umbilical cables. Each power relay had a dual sensor switch for detecting active state feedback read by TTL. When I tested the box by sequentially toggling each relay when sense matched the state.

It sounded like a misfired rattling Gattling gun.

I later that day fixed it, my reading about plasma de-oxidation and "wetting" and used a 10uF cap across the "non-gold plated" sense contacts with a 5V 10k pullup resistor. Then it performed like a well-oiled machine. unlike mercury switches which needs no oil to get a proper surface tension, just high purity which Joe Henry might not have had in his day.

--- Updated Sep 24, 2021 ---

Hi,,

a word about pictures and their file size.

FWIW, If one has Windows, I have always used Irfanview which can display and convert any image format and any size.