Lily Power Pods, TEG thermal electric module

Thermal Electric Generator from concentrated solar, designed to float in water, small scale power anywhere in the world.

Similar projects worth following
This project is a submission for the energy harvesting challenge. My local area is heavily surrounded by over 600 miles of water ways. This device combines three separate ideas. First being solar oven absorption , Second Concentrated solar and Third " TEG" Thermal Electric Modules using Seebeck Effect current generation. Having power harvesting in a pretty package means it can exist in places electronics and wired power normally cannot go. Lakes, rivers, hunting, off grid camping trips and even float these in the water fountains or in pots of water by park benches.

The inner workings of the Lily Power Module

2018-07-14 14_07_06-3D design 12 petal no supports _ Tinkercad.png

topview 3d representation of the head collector and the heatsink on the bottom. I will not draw the details as it will take forever so no good reason.

Portable Network Graphics (PNG) - 243.62 kB - 07/14/2018 at 18:12


2018-07-14 14_06_25-3D design 12 petal no supports _ Tinkercad.png

3d representation of the head collector and the heatsink on the bottom. I will not draw the details as it will take forever so no good reason.

Portable Network Graphics (PNG) - 96.28 kB - 07/14/2018 at 18:12


2018-07-14 14_04_43-3D design 12 petal no supports _ Tinkercad.png

3d representation of the head collector and the heatsink on the bottom. I will not draw the details as it will take forever so no good reason.

Portable Network Graphics (PNG) - 201.51 kB - 07/14/2018 at 18:11


24 petal lily reflector.stl

This is the newest model as of 7/11 . It has three sets of petals and varrying agle to reflect light at the center about 4-6 inches above the center depending on the angle of the sun.

Standard Tesselated Geometry - 4.81 MB - 07/11/2018 at 14:06


12 petal flower.stl

18 petals, 12 reflective to center collector. Print in any color but the inner face must be ccovered in mylar reflactive film

Standard Tesselated Geometry - 2.20 MB - 07/11/2018 at 13:54


View all 13 files

  • 5 × TEG Thermal Electric Generator Seeback module this uses hot / cold temperature transfer to generate voltage.
  • 18 × PLA 3d printed reflectors These are curved reflectors I am using to focus sunlight heat energy onto the center heat collector
  • 1 × DC to DC converter regulate voltage, step up from low voltage to 5 volt and usb output.
  • 1 × CPU heatsink, $10 on amazon this is the cold end that is submerged into water to attract heat transfer through the seeback modules.
  • 4 × 1/4" aluminum plates heat transfer medium

View all 9 components

  • 3d model loaded into files.

    Josh Starnes7 days ago 0 comments

    I am loading the 3d model into the files and even making a version with the cutouts for the heatsink and heat conductor already in the print. I needed to draw a beefy float to keep the flower above was as well that can also be 3d printed if you do not have 2 inch thick foam available for this project. The black box represents the heat conductor. I am considering using many copper 12 gauge wires in a bundle as the collector , but they will not be as good as a solid block, so this is a decision between look s versus performance. If you repeat my project this is a choice you will need to make yourself. The bottom box represents the CPU heatpipe cooler to handle the cold side of the TEG.

  • Multistage seeback information

    Josh Starnes07/13/2018 at 18:50 0 comments

    12.0 Description & Modeling of Cascade Thermoelectric Modules

    12.1 A standard single-stage thermoelectric cooling module is capable of achieving a maximum no-load temperature differential (DTmax) of approximately 72°C. It is possible to obtain DTs of up to 130°C by mechanically stacking modules on top of one another whereby the cold side of one module becomes the hot side of another module mounted above. This stacking arrangement is called a Cascade or Multi-Stage module configuration. Cascade modules usually, but not always, have a pyramid shape thereby the higher stages are physically smaller than those below. Regardless of the physical shape, however, lower stages must always have greater heat pumping capacity than the higher stages. although cascade configurations of up to six and seven stages have been constructed, practical cascade devices usually have from two to four stages.

    The principal factor that limits cascade module performance is related to the temperature dependent properties of the thermoelectric semiconductor materials. The performance of Bismuth Telluride alloys used in most thermoelectric coolers generally peaks near 70°C and performance falls-off appreciably at lower temperatures. Consequently, cascade modules exhibit a condition of diminishing returns where, as successive stages are added, the increase in DT becomes smaller.

    Figure (12-1) 

    Performance Graph of a Typical Cascade Module

    12.2 MODELING OF CASCADE MODULES: Modeling of cascaded or multi-stage thermoelectric coolers is somewhat more complicated than for single-stage devices. With multi-stage coolers, the temperature between each stage is critically important and module performance cannot be established until each interstage temperature value is known. With a two-stage cooler only one interstage temperature must be determined but, as more stages are added, the thermal analysis becomes increasingly complex. Manually calculating multi-stage module performance is extremely laborious, yet with a computer, the required calculations can be performed with little effort.

    The most common method for computer-modeling cascade modules involves carrying out an iterative series of performance calculations beginning with assumed interstage temperature values. Using this technique, the performance of each stage is repeatedly calculated until the difference between successive interstage temperature calculations becomes very small (typically 0.1°C or less). When this point is reached, each of the relevant module performance parameters can be ascertained. Note that the temperature-dependent value of SM, RM, and KM must be converted as explained in paragraph 11.2.4 to reflect the number of couples in each stage together with their optimum TE element currents. The following paragraphs describe the calculations needed to model two and three-stage cascaded thermoelectric modules. Four and greater-stage modules may be modeled in a similar manner by expanding the three-stage calculation routines to include terms for each additional stage. Calculations of the various parameters should be performed in the order shown.

    12.2.1 TWO-STAGE MODULE CALCULATIONS: A typical two-stage thermoelectric module is illustrated in Figure (12-2). The following new terms will be used in the module performance calculations:

    TM12 = the interstage temperature between stages 1 and 2 in °K  SM1 = the Seebeck coefficient of the 1st stage in volts/°K SM2 = the Seebeck coefficient of the 2nd stage in volts/°K RM1 = the resistance of the 1st stage in ohms  RM2 = the resistance of the 2nd stage in ohms  KM1 = the thermal conductance of the 1st stage in watts/°K KM2 = the thermal conductance of the 2nd stage in watts/°K

      Figure (12-2) a) The interstage temperature (TM12)...

    Read more »

  • update

    Josh Starnes07/11/2018 at 14:01 0 comments

    I printed the new flower model I put together, the edges ofcourse on a couple tips are wonky because I pushed the model too close to the edge of the bed and the print carriage hit the end stop. Such as Life.  I did recieve the step up voltage converter and regulator for 5 volts 1 amp hour and the heatsing I planned to use on the bottom of the flower. On the other hand however I have NOT recieved the TEG modules I ordered, they missed my location and went to another city, they are on their way back now and I expect them today. Lets cross our fingers :)

  • Reflective surfaces

    Josh Starnes07/10/2018 at 03:18 0 comments

    When you look directly at the flower you can see the angles of the petals reflect light. Once Mylar reflective film is covering the leaves this should reflect heat to the center like a mini solar oven.

  • Version 2 printed

    Josh Starnes07/10/2018 at 03:15 0 comments

  • Version 2 lily misprint but some good news

    Josh Starnes07/10/2018 at 02:06 0 comments

    the second version of the lily was a misprint totally my fault. I moved the print head when changing filament and I should not have unlocked the stepper motors. Sigh.. anyways it did get far enough along that I could verify the center is 100 times stronger than the first version. The first version was a piece of a model I found on thingiverse but has been scrapped, the new model has three times the petals and mimicking a dish shape to reflect heat at the center collector. My fiancé even volunteered to help paint it a pretty purple like a living wild lily shown in the pictures on the page profile.

    This is the new center, the heat collector will come up through the center of this and be all metal with flat black coating, there will be a silicon barrier separating the metal and plastic from touching and also a radiant barrier as well.

    You can see in comparison to version one center and petals version two is beefy, but still looks like a flower.

  • 2nd version designed and printing

    Josh Starnes07/09/2018 at 05:12 0 comments

  • 1st test pieces printed

    Josh Starnes07/09/2018 at 05:01 0 comments

    ok so the lily pad needs to be bigger so I scaled it to 1.5 and started agian. I liked the section of petal reflectors, but the center was poorly designed so I made the center solid so the petals have plenty of support.

  • 3D printing version 1 today

    Josh Starnes07/08/2018 at 22:44 0 comments

    in 7 hours I will have the main pieces printed. I will need to determine scale and go from there. If I like how it looks I will print at least two more sets of petals.

  • Cutting the lily model into peices

    Josh Starnes07/08/2018 at 15:36 0 comments

    OK so first I want to say the original model I found on thingiverse. I am not using the whole model and the parts I am using will be modified so It will really not look the same in the end.  I downloaded 123d to play with the parts and make what I need. I need to make a base, the petals and chop them up for printing a larger scale than I can get all in one print.  I will print the first pieces tonight. I am off tomorrow to work on some of this. The petals will be lights with silver to reflect heat towards the center which will be fabricated with copper wire.

View all 13 project logs

Enjoy this project?



Josh Starnes wrote 07/07/2018 at 19:43 point

I am talking about Fahrenheit  not Celsius 

  Are you sure? yes | no

Martin wrote 5 days ago point

Then it is best to write this explicitly (F or C) because in many parts of the world reading only "degrees" is understood as Celsius.

So you have not been at a volcanic spring :-)

  Are you sure? yes | no

Martin wrote 06/26/2018 at 08:42 point

How/where do you get rivers with about 60°? The hottest river I experienced was something like 28° in a really hot summer, like 35° air temperature.

But anyway, I would try to use a heatpipe from the cold side into the water instead of active cooling. That saves power for the pump.

  Are you sure? yes | no

Josh Starnes wrote 7 days ago point

I took your advice martin and purchased a amazon basics heatpipe heatsink. I think it will work well.

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates