Meeting the cooling conditions for TEGs

Hello forum. My first posting here.
I've recently moved to a 400 years old house, 15 minutes drive from a small village. Then winter came and I realized something has to be done to improve heating before next winter. Here temperatures can reach -30C (-22F). The sun diapered behind a near by mountain about 15th of November. This property produces more dead wood than I'll ever be able to make use of.
So I started searching for ways heat with wood without making myself a full time firewood worker and discovered the rocket mass heater, and then I discovered this forum, which has given me a lot of great reading for weeks. Then I got to think that this stove should be able to produce electricity as well. So I looked into variations of sterling engine and realized they were too small to make useful energy. Next up was steam engines, which scared me off. I touched TEGs and learned they were expensive. So I figured woodgas was the most realistic way to go - until today.

Today I came across cheap TEGs - less than $.8 / watt on ebay. These are small modules. 4x4cm, cost $2.5 and claim to produce 3.2w. 100pcs of these would generate quite some useful electricity. The catch is the conditions required to do this: Delta-T of 100C (212F), and max temp of 120C (248F) on the hot side, min temp of -60C (-76F) on cold side. If delta-T is dropped to 60C (140F) the electric power is cut to a third. I'd guess that delta-T > 100C would make quite a bit more electric power.

I assume that it's not a good idea to aim for the max temp using these TEGs. I'm thinking about 95C (203F) on the hot side could be reasonable for the lifespan of the TEG. When making a RMH to produce hot water, 95C (203F) could be a reasonable output temp for looping water. Then the cold side is wanted below -5C (23F), -20C (-4F) would be a dream. On the coldest few days the outside temp here should be able to cool a heatsink on the TEGs fairly well. But I'd like this thing to work even in the warmer days of winter, say in temperatures around 5C (41F). I've seen gas powered refrigerators (which I assume can be converted to rocket powered), but never freezers. Is it possible to meet the TEGs cooling needs of the TEGs with the energy of the firewood?

You can run hot water on the hot side and cold water on the cold side.
Where will this cold water come from?
Just a coil of piping that goes outside to your -5C garage. or a water+glycol mix that piped to outside.

I am not too sure how much energy the cold water pump will use up.
Please not that the 300W will only be produced while the stove is lit not all 24hrs of the day.
So closer to a solar panel 4hrs/day vs a hydro-electric 24hrs of production per day.
$300w*4hrs will give you 1200WHr/day - pump and other system lost.

You should build it. I might even buy a module from you once you have it up and working.

I'm planning to have a spring loaded feeder tray to push the firewood in. That way I'm hoping to have it burning maybe 10 hours per load. Burn time should just be a matter of length of the tray.

The thing with water is that you can't really get the 100C range without freezing or boiling. Glycol will help. But then, how do you cool the water when outside temp is above freezing? Even though outside temps can reach -30C here, most days has a temperature around -1C - +3C.

I feel that you haven't prioritised what you want.

The original idea is to heat your house with wood, then shifting to get electricity from heat using TEGs, and finally to actively refrigerate the TEGs with ice cold water.

The final idea is completely at odds with the first idea.

I can sympathise with you as I've been through a similar thought process myself. I spent some time looking at TEGs. I realised that nobody is using them in any serious capacity. There is probably a good reason for this.

This seems very complicated.
I would go back to wood gas.
It is pretty perfect for co-generation.
Cooling the IC engine heats water which you store for space heating needs.
A coil in the  storage tank could provide DHW or at least preheat that water.

How is the wind there?
Any running water?

Peter B. Onde wrote:The thing with water is that you can't really get the 100C range without freezing or boiling

That's only true at constant pressure. Pressurise the hot side and you're sorted. A pressure cooker might be adaptable?

Teg chips and modules come in an array of operating temp ranges. The ones i am looking at have a do not exceed temp. of 850 deg.F! That make the balancing act alot easier by allowing a wider hot to cold side min/max.

Operating a hot side at 450 deg.F, there is much more room for incidental temps over and below the target.

William..... 'tegs are complicated, try an internal combustion engine and its wood gasifier apparatus'

!

We are talking a brick or two of tegs, a circulator (3-5 watts) a radiator and a couple fittings and tubing.

TEGs can be quite simple and having low to no moving parts, no oil changes and other constant maintenance no  noise is a plus on some homesteads.

These are great for avoiding low batteries during long duration cloudy weather. Could be used as primary power, but usually are incorporated to augment other sources.

A single 100w teg averaging 30-75 watts 12 hours per day would have totally covered us for the last 4 weeks of low sun availability! Inverter standby and lighting loads are about 600Whr/d at our house.

@Matt, William
What I want? Get useful energy to the household using the resources available on the property - wood.
Sure there are other resources as well. There is a lake 300m away. They say the ice on this lake is never safe. I assume this means the ice never gets thick. So far I think of this more for fishing than energy.
There are two SMALL rivers here. Comparing weather statistics with the flow I see, tells me they should be able to produce the power needed for the household over a year. Problem is that when I first moved here they were dry. And I bet they will freeze in January. So using these as a source of energy will require a lot of buffering.
Wind is to be found on a top 400m away or by the lake.
Sun is useful April - October.

My needs are:
Heating the house October - April without being a fulltime firewood handler. I think a rocket mass heater combined with a large tank (5000L?) of water and radiant floor heating will be the solution. This will require water pumps.
Electricity to fridge, freezer, water pumps, laptop, ligths. Solar covers this April - October.

So, what my plan does not yet cover by the resources available at the property is electricity October - April, and my heating plan depends on this. This is why I started the search for ways to make rocket powered electricity. As stated, I've been looking at sterling, steam, and TEGs before I figured woodgas was the way to go. Then I came across these "cheap" TEGs that seems to solve "everything" if I can figure a way to cool them below -5C when the outside temperature is around 0C. I'm thinking there may be a way to do that makes the TEG-approach better than woodgas. I just don't know how, so I'm asking for ideas.

@Steve
Pressure can move the boiling point, but it takes a major leap in pressure to move the freezing point significantly. When the desired temperature range is from -5C (or less) to 95C, antifreeze is a better approach than pressure change.

@Frank
Sure there are TEGs that handles higher max temp. What do you find for cost / watt and life span of these high temp TEGs?
As mentioned, at first I wrote off TEGs as too expensive. Then I found these low temp modules for less than  $1/watt. I don't know much about these modules. Actually ebay is the most scientific source of information in this case! They claim to work with 150C delta-T, in the range from -60C to +125C, which is 185C, way more than 150C.
Claimed output:
delta-T : generated power
20C : 0,21w
40C : 0,66w
60C : 1,12w
80C : 2,00w
100C : 3,21w

So the output seems to be exponential to increasing delta-T. If it continues like that all the way to delta-T = 150C one could actually put an TEG powered compressor/heat pump to cool/heat the TEG and would end up having a machine that powers it self and creating energy from nothing! Who dare to say this is not a fantastic machine? (Never mind this machine only works in theory based on guesswork based on the data provided by a seller on ebay who probably has no knowledge of the product sold.) I ordered a dozen of them. 60 days of shipping.


Well. The topic for this thread. How to use energy from wood/air/ground/compost/.... to provide the best possible conditions for this TEGs when outside temperature is around 0C? Hot side is easy: water heated to 95C by a rocket. What about the cold side? (Ground temperature here is about +4C.)

I think your best cooling options are the lake or the ground.  Both would get you close to 5C with tons of capacity.  Long geothermal loop of antifreeze water in either heat sink could work.

To get the hot side to 95 without pressure or "boom-squish" you could have the hot side be an open water tank that is heated nearly to boiling with fire.  If it did boil, it's open so it won't blow up.

Close to 5C seems easy, yes. The goal is at least 10C colder than that < -5C

frank li wrote:
William..... 'tegs are complicated, try an internal combustion engine and its wood gasifier apparatus'

The OP will need to refrigerate his cold side to get the items they described  to work.
There is no off the shelf solution for this,much less one powered by heat.
Propane powered refrigeration is mentioned, but it would need to be converted to wood heat.
The wood heat would need to be consistent.
Could it work?
Maybe.
Charcoal and wood gasification works.
IC generators work and can be bought ready to go.

TEG generators are out there,ready to go.
This one is about $11.00 a watt at 100 watts.

https://www.tegmart.com/thermoelectric-generator-systems/100-watt-water-cooled-teg-sytem

They seem to water cooled, if they also need refrigeration,  we will have the same problem.

I just think the need for refrigeration, powered by heat, is one complications too many.

A simple closed loop of coolant through an electric freezer, better still through a tank or body of water, seems reasonable.

I don't fully understand the propane fridge. My understanding is that it needs a very high temperature at one small area to run. With a RMH this could be achieved by placing this area on top of the barrel. The rest of the cooling system will probably need to be as cold as possible - be exposed to the outside temperature rather than the RMH.

To my understanding a propane fridge can be placed in a 20C environment and will create 5C inside the box. That is a delta-T of 15C. IF that means it will be able to maintain the same delta-T when temperature outside the box changes, it may work for my application. To reach my goal in a +3C environment my "fridge" just needs a delta-T of 8C. From there it should just be a matter of scaling to reach a sufficient

If it's so that the propane fridge will aim for a temperature of 5C (thermostat disabled) independent of environment temperature, maybe another refrigerant in the system could be the solution?

William that TEG looks nifty.  I didn't see the temperature specs for it in there. Do you happen to know the min for the cold, the max for the hot and the ideal delta T?

Peter B. Onde wrote:@Steve
Pressure can move the boiling point, but it takes a major leap in pressure to move the freezing point significantly.

Sure, so use a pressure cooker to get 115C on the hot side and use ambient ground or lake temp for 15C on the cold side. Sorted.

The devilwatt and firevolt teg modules seem to be close in spec. No thorough data yet on firevolt. Devilwatt publishes an inlet temp ('cooled') of 86 degrees F. and outlet temps around 92 degrees with a hot side temp of about 450 degrees F. This with a 5W circulator (not sure gpm) and an unidentified radiator.

Still, i have seen chips (module) that are 400 or more watts! Kinda cool, even if run at 35%, they are cheap enough to use in a larger array and still meet a moderate load, low temp or not.

Mike Jay wrote:William that TEG looks nifty.  I didn't see the temperature specs for it in there. Do you happen to know the min for the cold, the max for the hot and the ideal delta T?

No idea, just found it looking for the commercial solution.

Steve Farmer wrote:

Peter B. Onde wrote:@Steve
Pressure can move the boiling point, but it takes a major leap in pressure to move the freezing point significantly.

Sure, so use a pressure cooker to get 115C on the hot side and use ambient ground or lake temp for 15C on the cold side. Sorted.

And when you don't want to be that close to the modules max temperature? (with durability in mind)

----

When it comes to those 230C (450F) max units I haven't fond any data telling their electricity production. They cost 15 times as much as the 125C max units. Do they produce 15 times as much electricity in a real world application? What I consider real world application is cold side kept somewhat above outside temperature (about 10C), and hot side kept as high as possible without reducing the units lifespan.  There are also some 180C (350F) max units out there. They cost 5 times the 125C. Do they produce 5 times as much electricity?

From the information I've read, I get the impression that the differences between these modules are the material used to seal the unit and cable isolation material. If that's true, their performance should be much the same at any given delta-T. From the data on the 125C unit, I find a formula for electric power produced to be:
P = 0.0004672339 * dT^1.915394

That makes this table:
cold - hot - dT - power
-20 - 95 - 115 - 4,14
-10 - 95 - 105 - 3,47
0 - 95 - 95 - 2,87
10 - 95 - 85 - 2,32
20 - 95 - 75 - 1,82
30 - 95 - 65 - 1,39

-20 - 115 - 135 - 5,62
-10 - 115 - 125 - 4,85
0 - 115 - 115 - 4,14
10 - 115 - 105 - 3,47
20 - 115 - 95 - 2,87
30 - 115 - 85 - 2,32

-20 - 150 - 170 - 8,74
-10 - 150 - 160 - 7,79
0 - 150 - 150 - 6,88
10 - 150 - 140 - 6,03
20 - 150 - 130 - 5,23
30 - 150 - 120 - 4,49

-20 - 170 - 190 - 10,82
-10 - 170 - 180 - 9,76
0 - 170 - 170 - 8,74
10 - 170 - 160 - 7,79
20 - 170 - 150 - 6,88
30 - 170 - 140 - 6,03

-20 - 180 - 200 - 11,94
-10 - 180 - 190 - 10,82
0 - 180 - 180 - 9,76
10 - 180 - 170 - 8,74
20 - 180 - 160 - 7,79
30 - 180 - 150 - 6,88

-20 - 210 - 230 - 15,60
-10 - 210 - 220 - 14,33
0 - 210 - 210 - 13,11
10 - 210 - 200 - 11,94
20 - 210 - 190 - 10,82
30 - 210 - 180 - 9,76


As can be seen, if you get 30C on the cold side, a hot side of 210C makes 7 times more electricity than 95C on the hot side.
If you get -20C on the cold side, a hot side of 210C makes 3,76 times more than the hot side of 95C.

To me it seems like a bunch of the cheap ones will produce way more electricity than the expensive ones for the same money. And the colder you can make the cold side, the better value for money the cheap modules get.

Another significant point to consider is the thermal resistance of the heat sinks, one on each side of the TEG module.  As I recall, the better cpu heat sinks have a thermal resistance of 0.1 deg.C/W.  
The common SP1848 27145 TEG reportedly has a thermal resistance of about 1 deg.C/W.  Adding these up gives a total thermal resistance of 1.2 deg.C/W.  Note this is 20% higher than the TEG module alone.  This means that the temperature difference across the TEG would be only 1/1.2 = 80% of the temperature difference across the entire system including 2 heat sinks. Basically what this means is that heat sinks using liquid as a heat transfer fluid will in theory perform much better than just using air.