Lessons in building a 120-Watt thermoelectric wood stove

Guest Blog: We are reposting a 2012 blog from Instructables by Tecwyn Twmffatt at Goat Industries. It describes an early effort to build a thermoelectric wood stove.  This blog is part of a series of blogs providing information for the 2018 Wood Stove Design Challenge.

Introduction: Thermoelectric Power Generation (TEG) 

These videos document my first attempts at generating electricity from a thermoelectric peltier device in 2012. The TEG that I used is a high powered unit able to withstand high temperatures and specially made for electricity generation. In terms of instructions, I don't think many people would want to build the 10 TEG system as it was ridiculously expensive, so I'm putting in a section for creating the 1 TEG arrangement, which is relatively easy and low cost.

Step 1: Part 2 of 3 


A ten unit Thermoelectric generator system is shown being constructed and then fitted to a wood burner. The theoretical maximum output is 200 watts. The video shows how the generator was put together and how the wood burner was modified to get maximum heat through the TEGs. The TEGs themselves are able to withstand a constant 325 degrees C on the hot side and require plenty of heat to get the 20 watts that each of them are capable of producing.

Step 2: Part 3 of 3 


In part 3 we successfully generate a significant amount of energy from the woodburning stove. In the first session, a circulation pump, a fan and 10 x 10 watt flood lights are powered up. In the second session, we attempt to get a more balanced load wired up to the tegs and measure a noticable increase in power output. The 10 tegs are wired up in 2 parallel strings and, from the manufacturer's specification, the optimum output voltage is 14.4v . The nearest that we manage is 13.8v, at which we generate 120 watts. The specifications suggest that 200 watts is possible when the load is matched.

Step 3: Full Playlist

31 Minutes of Thermoelectric video heaven!

Step 4: Creating the 1 TEG Generator

Here we are going to build the single TEG generator shown in the first video.

Step 5: Tools and Equipment  

Thermoelectric power generator TEG module (GM250 449 )
...... buy direct from China at: www.thermonamic.com/
Aluminium block 102 x 115 x 20 mm
Steel block 102 x 115 x 10 mm
1/4" BSP blanks x 6 of
1/4" BSP male stud push fit pneumatic fittings for 10 mm pipe x 2 of (See photo above)
5 mm Hex bolts x 40 mm x 2 of
25 litre water butt
OD 10 mm ID 8 mm nylon pneumatic pipe
12V water pump
12V LEDs, 1 watt x 20 of Tools:  1/4" threading tap
5 mm metric coarse threading tap
Drill 11.5 mm
Drill 5.5 mm
Drill 4.2 mm
Drill press
Torque wrench
MIG welder
Plasma cutter / Grinder with cutting discs

Step 6: Drilling and Tapping the Cooling Block

Use the engineering drawing to produce internal coolant passage ways in the aluminium block. I ended up drilling all the way through to the other side and using more of the 1/4" blanks. Connect the 1/4" pipe fittings to the block and plumb in the pump. Add antifreeze to the water in the water butt if it's likely to get cold at all. To create a 'sandwich' with the hot block (steel block), the TEG and the cooling block, drill and tap holes in the steel block for the 5mm bolts. Weld the hot block into the side of the wood burner and recreate the TEG sandwich, tightening the bolts up with a torque wrench (see attached file). Connect up LEDs on the TEG, turn on the pump, light the wood burner and off you go!

Step 7: 10 TEG Layout

If you really must build the 10 TEG generator, the photo above shows what is involved. I have got CAD drawings, PCB drawings etc. If anybody is interested. Not for the faint hearted! PCB 03.pcb
PCB 01.zip
CAD files 02.zip Heated Up!

Could a Thermoelectric Wood Stove Pay for Itself?

By Ken Adler, AGH Senior Technical Advisor

Payback calculations are common in the residential solar photovoltaic industry where homeowners want to know how long it will take for them to recoup their initial investment. If you purchase panels outright, payback periods depend on a variety of factors including a utility’s price for electricity, tax incentives, and amount of daily sunlight hours. A range of 5 to 8 years is possible however, it can be as wide as 3 to 15 years.[1]

Answering the payback question for thermoelectric wood stoves is one of the objectives for the 2018 Wood Stove Design Challenge. In the meantime, there are several ways to begin answering this question with information already available. It is also useful to look at how use of a thermoelectric wood stove in combination with another energy-saving system, i.e., solar, could prove beneficial to the homeowner and thus both industries as well. For example, in northern states and Canada, a thermoelectric wood stove could reduce the number of residential panels needed and thereby save the homeowner thousands of dollars in panel costs.

Early Thoughts on Payback

The retail price of a thermoelectric module is around $ 57.50 for a 22-watt module, or $ 2.61 per watt.[2] One critical point to make here is that the power output of our 22-watt module assumes an optimal hot-side temperature of 300 C (572 F) and cool-side temperature of 30 C (86 F). This ideal temperature differential is very difficult to achieve in real world conditions, so the real-world cost per watt for thermoelectric modules will be higher. However, cost should decrease and efficiency improve with widespread adoption of thermoelectric modules, similar to what happened in the solar industry. For example, DOE estimated that the installed cost of a solar panel declined from $ 7.06 per watt in 2009 to $ 2.93 in 2016, a reduction of 60 percent.[3] If we go back to 1977, the cost of a solar panel was $ 77 per watt. It is not unreasonable to expect a decline for the cost of thermoelectric modules as economies of scale reduce production costs.

Of course, when a thermoelectric module is placed into a wood stove there are other associated costs. The primary cost by far is the heat exchange system. As I’ve discussed in a previous post, to generate at least 100 watts of power, it’s likely that a water-cooled heat exchange will be needed. The current retail price for a 100-watt water cooled thermoelectric generator, which includes eight thermoelectric modules, is $ 599, or $ 5.99 per watt. One question the competition will attempt to answer is how much this heat exchange will cost when it is integrated into the design of the wood stove.

Secondary cost considerations include the price of the wood stove, its installation, and fuel costs. The price for a larger size 50,000 BTU wood stove can range from $ 900 to over $ 4000, and the average consumer spends about $ 2,500. Since a thermoelectric wood stove would be providing both heat and electricity, it is difficult to separate out how much of the cost of the stove is for each function. The more crucial point for now is that many larger size stoves, which can generate up to 50,000 BTUs and meet the 2020 EPA NSPS standard, are available for as little as $ 1,300. While this does not include the cost of installation, it does suggest that the wood stove portion of the costs should not be a major obstacle.

The cost of installing a thermoelectric wood stove into a home should not necessarily be that much greater than the cost of installing a traditional wood stove. One additional cost will be attaching the power outputs from the thermoelectric wood stove to an inverter. However, if we assume that early adopters will already have or are planning to get a solar PV system (more on this below) the cost of the inverter would not be a major obstacle.

Finally, one can assume that the fuel cost for a thermoelectric wood stove is essentially zero because the wood stove is already being used to heat the home. A thermoelectric module will convert only 3 to 6 percent of the heat from a woodstove into electricity, while the remaining 94 to 97 percent passes through the module and is released as heat into the home. In other words, the module is only using a very small percentage of the heat generated by the stove to produce electricity.

Value in Combining Technologies

While more in-depth analysis is needed, it’s possible that a thermoelectric wood stove could help reduce the size and cost of solar PV systems in northern climates that have limited sunlight/solar radiation in winter. For example, a typical 5000 watt solar PV system in Vermont produces 6,280 kWh of electricity per year, while the same system produces 7,913 kWh in Los Angeles.[4] Most of this difference is due to the low winter time output in Vermont between October and February: For example, the Vermont system produces 239 kWh in December, as compared to the Los Angeles system’s 473 kWh. If the Vermont resident wanted to generate the same amount of power as in Los Angeles, they would need to increase the size of their solar PV system from 5000 watts to approximately 6300 watts. At the current cost of approximately $ 3.36[5] per watt installed for residential solar, this could cost the Vermont resident an additional $ 4,368 for additional solar panels.

Alternatively, instead of purchasing extra solar panels, the Vermont resident could invest in a thermoelectric wood stove to boost their winter time power output. As we mentioned in our previous blog, a wood stove with a 150 to 200-watt thermoelectric generator operating 16 – 20 hours per day could generate 93 to 124kWh of electricity per month, which would be a good boost to the Vermont output of 239 kWh in December. And, at 0.16 $ /kWh for electricity in Vermont, the thermoelectric wood stove could save the homeowner an additional $ 15 to $ 20 per month.

While a real payback calculation for a thermoelectric wood stove will need to wait until prototypes go through more testing and we get results from the 2018 Wood Stove Design Challenge, the available information suggests thermoelectric wood stoves could help reduce the cost of residential solar installations, and potentially save homeowners thousands of dollars.

[1] http://solar-power-now.com/the-typical-solar-panel-payback-period/
[2] See our Resources page for a list of thermoelectric retailers.
[3] NREL. U.S. Solar Photovoltaic System Cost Benchmark. September 2016. In 1977, solar panels cost $ 77 per watt.
[4] NREL PVWatts Calculator
[5] EnergySage. Solar Marketplace Intel Report. April 2017.

Heated Up!

Hybrid Residential Solar and Thermoelectric Power Generation

by Ken Adler, Senior Technical Advisor at the Alliance for Green Heat
Some of you may be wondering about thermoelectric wood stoves and why we decided to include them in the 2018 Wood Stove DesignChallenge, which will be held in November 2018 on the Washington Mall.  Our goal of this competition is to support development and commercialization of a revolutionary thermoelectric wood stove that produces electricity equal to 50 percent or more of the winter time output of a residential solar photovoltaic system. By combining a thermoelectric wood stove and a residential solar PV system and home battery, like the TESLA Powerwall, we can support residential and grid-based distributive power goals, and incentivize greater investment in solar power. 
Specifically, thermoelectric wood stoves can help solve the problem of low winter time solar PV output in northern climates, where useful solar radiation is limited to 2 – 4 hours per day.
While a thermoelectric wood stove may sound revolutionary, the technology behind the stove has been used since the 1980s in oil and gas field operations, where methane gas provides a low-cost source of heat to power the thermoelectric generator. Wood stoves, like waste methane gas, can provide a free source of heat for the thermoelectric generator.
Alphabet Energy Thermoelectric Generator
Thermoelectric generators are like solar panels, however, instead of turning light into electricity they turn heat into electricity. To generate electricity, one side of a thermoelectric module is heated by the wood stove while the other side is cooled with either an air or water-cooled heat sink. For applications above 100-watts, water-cooled heat sinks are the most common approach because of their ability to extract greater amounts of heat from the thermoelectric module.
60-Watt Water Cooled Thermoelectric Generator
In northern climates like New England, Canada and northern Europe, low winter time solar radiation increases the cost and reduces the efficiency of solar PV systems, and the cost-effectiveness of battery storage systems like the Tesla Powerwall.  According to NREL, solar radiation in northern areas like Vermont peaks at 6.0kWh/m2 in June and declines to 1.7kWh/m2 in December. This means that an average 4,000-watt residential solar system will go from producing 571kWh in June to 191kWh in December–a 66% reduction is solar power output.  This project will demonstrate how a thermoelectric wood stove can cost-effectively supplement a solar PV system.
Building on our experience from 3 previous Design Challenges, we will work with wood stove manufacturers, universities and others to build and test 100 to 200-watt thermoelectric wood stoves that could effectively increase by 50% the winter time output of a 4,000-watt residential solar PV system.   

Thermoelectric generators are currently sold as accessories for wood stoves; however, these accessories are limited in size and efficiency. By integrating a thermoelectric generator into a wood stove we can achieve far greater power output, efficiency, and lower cost. For example, a wood stove with a 150 to 200-watt thermoelectric generator operating 20 hours per day could generate 93 to 124kWh of electricity per month, which compares favorably with the December solar PV output of 191kWh in Vermont.

Russian Thermoelectric Wood Stove 
(not certified for sale in the U.S.)
There are several reasons why now is the time to consider thermoelectric wood stoves. First, the price of the thermoelectric modules, which are a component of the TEG, has dropped substantially because they are now being mass produced in China.[1] Second, the EPA’s recent wood stove NSPS regulation is helping to make new wood stoves cleaner and more efficient and, coupled with cordwood testing and automated features, a new generation of cleaner stoves could also generate electricity. Third, thermoelectric wood stoves can produce electricity up to 24 hours per day eliminating load management concerns common with solar and wind power. Lastly, the stoves are powered by local wood supplies, making their fuel low carbon and locally sourced.
The 2018 competition on the Mall will demonstrate the role thermoelectric wood stoves can play in promoting solar power, energy storage systems and biomass energy, while also reducing energy costs, supporting climate change goals, and increasing distributive power.   

[1]The cost of a thermoelectric module has fallen below $ 2 per watt (uninstalled), compared with $ 3.50 per watt for solar panels (installed).

Heated Up!

Thermoelectric Wood Stove, Solar Power, and a Floating Cabin!

Guest blog post, by Margy Lutz

Finally this winter, our thermoelectric wood stove generator is fully operational. Following our test runs, we placed the pump to recycle cold water down in the lake water under the cabin. In winter, it gets about 5 degrees C (41 F). That’s plenty cold for a good differential between the 300 degrees C on the hot side.
Most system owners don’t live in a float cabin four feet with a constant cold water source under the floor. The typical user has to use a recycled liquid (usually including a water/antifreeze mixture) that runs through a radiator for cooling.
In addition, a charge regulator/controller is used to protect the batteries and prevent overcharging. The model that came with our system has lights to let you know the status of the charging process.

Wayne likes to know more about the charge we are getting. He installed an ammeter and a volt meter. The switch in the middle controls the water pump down below the cabin. To maximize the charge to our cabin battery bank, we’ve installed a separate solar panel and two six volt batteries wired in a series to run the pump.

Living off the grid has its challenges, but having an alternative power sources has made our winters much brighter (pun intended). Do you generate power? What are some of the solutions that have worked for you? — Margy

Postscript by Ken Adler, AGH Technical Advisor: 

Congratulations to Wayne and Margy on their thermoelectric wood stove. In a follow-up communication with Wayne, he reported that they are no longer using the system because the thermoelectric modules failed. Wayne doesn’t know why they failed, however, the most common reason for failure is overheating.  The modules can also fail if Bellville washers are not used to allow the module to expand and contract during heating and cooling. Wayne also reports,

Even when I was partially (marginally) operational, I produced less than 2 amps at 12V DC (23 watts) to recharge my cabin battery bank. This would have been enough to put a top-off charge on my cabin batteries (normally recharged via my solar system), particularly valuable in the winter when solar power is minimal and my wood stove is operating nearly 24-7. The primary reason for the low amperage was the need for a 1,8 amp 12V (21.6 watts) water pump to feed the cold side of the modules. In many ways, I reside in the perfect test location for this thermoelectric system, since 

I have a nearly infinite supply of very cold water 4 feet below my wood stove. I live in a floating cabin on Powell Lake BC, and the lake is extremely deep and very cold in all seasons. What an opportunity to serve as a source of cold water through the cooling system! The pump only needed to pump the cold water up 4 feet and then outflow back into the lake. Even with this tremendous advantage, I couldn’t get everything fully operational.

Does this make me a non-believer in thermoelectric from a wood stove? Absolutely not — I still believe this is an important future source of electrical power in my cabin, since even a top-off voltage during the solar-depraved Canadian winter would be worth the price. I’d be one of the first in line if a recreational property thermoelectric system was available, and I’d be quick to try again. Thus, I wish you all of the best with your preparation for the 2018 conference. I’ll be following the results closely.

In an earlier post, Wayne reports that he is using three 25 watt thermoelectric generators for a total rated power of 75 watts of output, however, he’s only getting 23 watts of power for his battery. Part of this is due to his pump, which is drawing almost 22 watts of power. If you are interested in building your own thermoelectric wood stove, there are a few improvements that you may want to consider. First, TEG suppliers (see our resourcespage) now sell more efficient lower wattage pumps. Second, consider starting with a thermoelectric generator rated for 100 to 200-watts. While this is more expensive, if you go with a smaller system much of your power will be consumed by the pumps and/or fans you need to cool the modules. Third, Bellville washers are critical for allowing the modules to expand and contract.

If you are interested in designing a thermoelectric wood stove for our 2018 Wood Stove Design Challenge, please visit our web sitefor more information. For more information on Wayne and Margy’s life on a floating cabin, please visit their blog at Powell River Books Blog.

For an overview of the potential of thermoelectric wood stoves, click here.

Heated Up!