New highly efficient thermoelectronic generator

Discuss the technical details of an "open source" community-driven design of a polywell reactor.

Moderators: tonybarry, MSimon

Post Reply
Posts: 2245
Joined: Mon Oct 12, 2009 5:05 am

New highly efficient thermoelectronic generator

Post by DeltaV » ... ronic.html
"Practical thermionic generators have reached efficiencies of about 10 percent. The theoretical predictions for our thermoelectronic generators reach about 40 percent, although this is theory only," noted Mannhart.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Re: New highly efficient thermoelectronic generator

Post by D Tibbets »

A thermal to electrical conversion of up to 40% would be interesting. It would match or slightly exceed the conversion efficiency of the steam- turbine approach.

Even if it worked, there would be many questions. Over what delta T does it work and what power intensities could it handle? If the power levels are small, the technique might be useful for satellite nuclear thermal batteries, but little beyond that.If power capacity is high it might compete with steam generation - turbine systems.
I speculate that the greatest usefulness might be derived if a low delta T is workable. Used in tandem with with a steam plant, otherwise waste heat from the low pressure turbines could be further utilized to generate more electricity and reduce the thermal load on the environment (less cooling towers). If you start with 100 % thermal energy and convert ~ 30% of that to electricity with turbines, that would leave 70 % of the energy. A recovery of a further 40% of that fraction would yield ~ 30% more energy recovery/ conversion. Ignoring system inefficiencies, yields of up to 60% thermal to electrical conversion might be obtainable. Depending on the economic cost of such systems, considerable electrical power gains from existing thermal input would be possible.

This not only appears interesting for coal fired plants, but also for fission and fusion plants. Especially in fusion plants where the Q may be marginal, it would allow for better net yields. It may still trail behind direct conversion like that suggested for P-B11 Polywells and perhaps some versions of DPF or FRC machines, but it might still contribute. Even with P-B11 there will the significant thermal loads that will require cooling. If the cooling flow is passed through this thermo electric conversion instead of merely disposing of the heat the net gains would be significant.
If direct conversion is 80% of the total power out from the reactor, the 20% heat output could be converted at 40% to additional electricity.* This would result in an additional ~ 8% increase in net electrical production for a total approaching 90% efficiency. Assuming that this thermo electrical conversion is simpler and more economical than running a mildly less efficient steam plant would result in a significant change in the energy- profitability picture. Using both thermal conversion schemes would give additional gains as illistrated above, but economics may limit the application (you have to include a steam plant).

Even using Riders assumptions and conclusions, this could push P-B11 Polywell capabilities into the black. Using Rider's peak possible Q of 0.4 for P-B11 fusion results in 1.4 units of energy output for every 1 unit of input.With steam conversion that results in ~ 0.4 units of useful electrical power out for each 1 unit of power (electrical) in. This is a net loss. At 80% direct conversion only this results in closer to a net of only 0.3 units out for each one unit in. Keep in mind that direct conversion only works on the high energy fusion product ions. Adding steam conversion or perhaps better yet thermoelectrical conversion to the remaining thermal energy would result in ~ 1.1 units of heat being converted into as much as 0.4 * 1.1 of ~ 0.44 units of useful electrical output. The net from both the direct conversion and the thermal conversion would be ~ 0.74 . Umm... still not there, but getting close. Time to invoke dilution schemes to push you over the top. The additional benefits of claimed potential well distributions that implies slow electrons in the core (with corresponding decreased Bremmstruhlung losses (ends up as heat)) would push gains to much more profitable margins.

If the x-ray energy converter envisioned by Lawrenceviile Plasma Physics (DPF) works and can be applied, the energy balance/ waste heat management for advanced fuel (essentially P-B11 or D-He3) reactors may be even more manageable.

To illustrate the importance of conversion efficiency a supposed over all efficiency of 80% for both direct conversion and thermal conversion would result in net gain for the above Q example. Energy in = 1 unit, energy out =1.4 units(energy in plus fusion energy). Total useful energy out = 1.4 units * 80% = 1.1 units of useful energy out. Also, the use of the otherwise waste heat may be important. For example, the street heating and building heating in some Scandinavian countries changes the picture.

* I'm uncertain of the 80% direct conversion claims. If the Q is high, then the fusion product ions would contribute the most energy in the system and final conversion efficiencies would approach this number. But if the Q is closer to 1, then the contribution of fusion product ion energy and thus the net direct conversion efficiency would be relatively less., such as the example above. Bussard's use of the above 80% conversion efficiency may reflect only the fusion product ion conversion efficiency, so in low Q yielding systems the net efficiencies is a more complex compromise between direct conversion and thermal conversion schemes.

Dan Tibbets
To error is human... and I'm very human.

Posts: 6114
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Re: New highly efficient thermoelectronic generator

Post by KitemanSA »

I think the 40% is OF the theoretical Carnot efficiency. So if the delta T said 50% Carnot, you would get 20% real world. 40% is low, but not obscenely so. For the simplicity, the lower efficiency can be worth it.

Posts: 428
Joined: Mon Mar 02, 2009 6:36 pm

Re: New highly efficient thermoelectronic generator

Post by ohiovr »

Thermionic devices are interesting things and quite different than what is known as thermoelectric devices though the principle is the same. Its a heat engine but its a vacuum tube. Thermoelectric devices require materials that have a high electrical conductivity and low thermal conductivity. These are hard to find. When you think about it, what could be more ideal in this case than a vacuum tube? Phonons don't travel though a vacuum, but electrons sure can. There have been 2 problems with thermionic converters since their discovery. The first problem is negative space charge which restricts the flow of electrons over too far distances. This is a common design problem in most vacuum tubes. What you have is a stream of electrons don't travel instantly from one place to another and in the process of travel they fill a space, and that space becomes negatively charged by electrons and this makes electrons being thermionically emitted from the cathode not want to leave the surface (like charges repel) and head to the anode. If the emitter and collector are within 10 microns of each other, space charge is not a problem because.. there isn't much space.

They were not able to do this in the past because of the second problem, which is the work function of the materials, the minimum energy required for an electron to emit thermally from the surface of the emitter, was too high, which dictates the operating temperature of these devices have been extremely high. Classic Thermionic devices operated from 2000 k to 900 k. At those temperatures, even refractory metals would evaporate somewhat and collect on the anode. Eventually the cathode and anode would short out after a few minutes of operation. This made 'vacuum' converters impractical. So they got around this problem on practical devices by filling the cavity with a gas, usually cesium vapor. This allowed for devices to have about 10 percent efficiency. Obviously this didn't quite set the world on fire. The best work function available on electrodes at this time was around 2 ev. Check out the richardson dushman equation to see how work function relates to operating temperature and current output. It is very non linear! Cesium vapor's ohmic losses lowers the theoretical efficiency for these kinds of converters.

The innovation in this article is about getting around space charge using electrically charged grids. its not unlike triodes and pentodes.

However If you had a superior work function surface, its a moot point. Operating anything at 2000 k is non trivial to say the least. A super low work function material would eliminate high temperature operation restrictions, and also eliminate the electrode erosion problem and would allow vacuum converters to be practical. In fact, a super low work function material would enable thermionic converters to exceed the efficiency of pretty much any heat engine you could come up with and be far cheaper to manufacture.

Post Reply