Next Generation of Nuclear Power Ideas -- Opinions?

[Dlhill13]

Recently, I have been looking into more effective methods of harnessing nuclear power and making an attempt to move away from conventional steam plants and the use of turbines, thus reducing heat losses, maintenance and construction costs, etc.

That being said, my current concept uses a reactor to transfer heat to a coolant, which is moved through piping via a pump. The piping is coated with a thermally conductive material that is not electrically conductive. Around this layer are multiple thermoelectric generators, all connected in series to increase voltage output.

On the outside of the TEG's, a refrigeration unit would keep the outside of the units around -10 to -20F. In areas with cold climates, outside air may be used as a heat sink for the condensing unit of the refrigeration system instead of a body of water. The temperature difference (around 350K) will be enough to generate small voltage in each unit, then the series connection adds all of the voltages. This output is then run through a step-up transformer to get it high enough to be used for electrical distribution. Waste heat due to I^2R losses from the transformer will be removed via another coolant system, which dissipates that heat into a regenerative set of TEG units, cooled by the same refrigeration unit as the main units.

To compensate for the drop in current while stepping up the voltage, multiple transformers would be used in parallel to achieve an appropriate amount of current. A single reactor would have between 8 and 12 coolant loops, each loop having multiple TEG piping units. All units would connect to 2 different outputs to supply electricity to the power grid. Please share any opinions, suggestions or questions you may have. Any input from engineers would be greatly appreciated.

 

[berkeman]

Recently, I have been looking into more effective methods of harnessing nuclear power and making an attempt to move away from conventional steam plants and the use of turbines, thus reducing heat losses, maintenance and construction costs, etc.

That being said, my current concept uses a reactor to transfer heat to a coolant, which is moved through piping via a pump. The piping is coated with a thermally conductive material that is not electrically conductive. Around this layer are multiple thermoelectric generators, all connected in series to increase voltage output.

On the outside of the TEG's, a refrigeration unit would keep the outside of the units around -10 to -20F. In areas with cold climates, outside air may be used as a heat sink for the condensing unit of the refrigeration system instead of a body of water. The temperature difference (around 350K) will be enough to generate small voltage in each unit, then the series connection adds all of the voltages. This output is then run through a step-up transformer to get it high enough to be used for electrical distribution. Waste heat due to I^2R losses from the transformer will be removed via another coolant system, which dissipates that heat into a regenerative set of TEG units, cooled by the same refrigeration unit as the main units.

To compensate for the drop in current while stepping up the voltage, multiple transformers would be used in parallel to achieve an appropriate amount of current. A single reactor would have between 8 and 12 coolant loops, each loop having multiple TEG piping units. All units would connect to 2 different outputs to supply electricity to the power grid. Please share any opinions, suggestions or questions you may have. Any input from engineers would be greatly appreciated.

Welcome to the PF.

So you are trying to add in a bit of thermoelectric generation on top of the standard steam/turbine conversion? Are you aware of the thermoelectric efficiency versus the Carnot efficiency? Can you post those numbers for us?

 

[Dlhill13]

Based on the temperatures that were standard in the reactor plant I used to work at, efficiency (of the steam plant alone), was around 35% in an ideal situation.

(let n represent Carnot Efficiency)

n = [ (T_h - T_c) / T_h ] * 100%
n = [ (449.8K - 289.8K) / 449.8K ]* 100% = 35%

I'm going to do some math on the thermoelectric efficiency when I get home.

 

[gmax137]

My guess is that the refrigeration unit that keeps the cold side cold would use more power than the TEG generates. In other words, wouldn't the TEGs just heat up to a uniform temperature? I'd have to see a quantitative analysis to be convinced otherwise.

Also, what is the cost of the TEGs, how long do they last, etc.

 

[QuantumPion]

If your main concern is maximizing the thermodynamic efficiency of a reactor, the most efficient would be a once-through very high temperature brayton cycle (gas turbine). If simplicity is your goal, I think the best approach would be a stirling engine design. Thermoelectric conversion is low efficiency and is only useful for low power applications requiring no maintenance/moving parts such space probes.

 

[Download]

Recently, I have been looking into more effective methods of harnessing nuclear power and making an attempt to move away from conventional steam plants and the use of turbines, thus reducing heat losses, maintenance and construction costs, etc.

That being said, my current concept uses a reactor to transfer heat to a coolant, which is moved through piping via a pump. The piping is coated with a thermally conductive material that is not electrically conductive. Around this layer are multiple thermoelectric generators, all connected in series to increase voltage output.

On the outside of the TEG's, a refrigeration unit would keep the outside of the units around -10 to -20F. In areas with cold climates, outside air may be used as a heat sink for the condensing unit of the refrigeration system instead of a body of water. The temperature difference (around 350K) will be enough to generate small voltage in each unit, then the series connection adds all of the voltages. This output is then run through a step-up transformer to get it high enough to be used for electrical distribution. Waste heat due to I^2R losses from the transformer will be removed via another coolant system, which dissipates that heat into a regenerative set of TEG units, cooled by the same refrigeration unit as the main units.

To compensate for the drop in current while stepping up the voltage, multiple transformers would be used in parallel to achieve an appropriate amount of current. A single reactor would have between 8 and 12 coolant loops, each loop having multiple TEG piping units. All units would connect to 2 different outputs to supply electricity to the power grid. Please share any opinions, suggestions or questions you may have. Any input from engineers would be greatly appreciated.

You realize the refrigeration will defeat any gains you make from the lower temperature, right?