How to use TEGs to create power with your coolant heat
 

A little over a year ago, I showed you guys how to use TEGs to create power with your exhaust heat. To make a long story short, you can totally do it but its really expensive. I calculated over $1100 to generate 210W.

This time around, I'd like to see what we can do with using TEGs and engine coolant heat.

First, lets start at the beginning again with a little recap.



https://upload.wikimedia.org/wikiped...iagram.svg.png

What is a TEG / peltier / Seebeck generator?

Wikipedia says:

This time around, we are looking at using engine coolant as our heat source. This should allow us to use TEGs that don't have to withstand as high of temperatures, and thus reduce cost. However, less temperature differential also means less power output. Will it be worth it? Lets continue on to find out.



https://ecomodder.com/forum/attachme...1&d=1538503271

Alright, lets find some TEGs!

We could start by finding some TEGs at the normal online shops like ebay and amazon. However, we already know price is a major concern from our last go about with TEGs. So, I'm going to go for the cheapest thing I can find that will work. For that, I checked out amazon. On amazon, I basically found a TEG that I think would work. It is 40mm x 40mm, and it is rated for higher than 100C operating temperatures.

Gimiton SP1848-27145 TEC 40x40mm


By looking at this, we can get an idea of how much the one TEG module will generate. We know coolant temps are around 190F when the engine is warmed up. I'm going to use 70F as an ambient temperature. Obviously it will get better or worse depending on the environment. Now, lets convert to units the rest of the world uses.

190F = 88C
70F = 21C

So, we have a temperature difference of 67C. Lets round down to be conservative. That means our TEG will put out 2.4V and 469 milliamps, right? Wrong. Open circuit readings put no load on the TEG. As you put a load on it, the voltage will drop. The same thing happens with solar panels, and batteries. The amperage reading I believe we should be able to take as fact because there must be a load for there to be amperage. So, by my previous experience with TEGs, we can take the open circuit voltage and divide by 2 to get an idea of what we'll actually see while its in use. Thus, we will get 1.2V and 469 mA, or .56W per TEG module.



https://ecomodder.com/forum/attachme...1&d=1538503479

How many TEGs do you need?

This all depends on how much power you want to generate. A typical small alternator puts out a max of 80A, or around 1100W. So, if you wanted to generate that kind of power you would need 1100 / .56 = 1964 TEGs. Kind of ridiculous, right? How about we look to achieve something more reasonable like reduce the alternator's load by 10A or 135W. That means we will need 135 / .56 = 241 TEGs. Still seems pretty silly. That is over 4 square feet of TEGs to mount and install somewhere!

And, this is really where the TEGs creating power with coolant falls apart. You're talking a ridiculous number of TEGs to create any sort of real power. 10A won't even power a normal car going down the road.



Cost

Let us look at the cost aspect real quickly. The SP1848 TEGs currently cost just under $2 each shipped from China. If we get 240 we pay $480 for ~134W of generating power. We still need to mount them to something, run coolant to that something, and put a heatsink on the other side. So there is plenty of cost there as well. Just the TEGs themselves cost $3.58 per watt. That is actually an improvement over the exhaust powered TEGs which came in at $5.23 per watt.

For kickers, lets also compare this to solar power. Solar panels can be had for under $2 per watt these days. The wonderful thing about solar is you can generate power no matter if you are driving or not. As long as your solar panel is in the sun, you're generating power. This is a huge advantage (unless its overcast of course) as your solar panel can chug along all day long vs just the 30 minutes you're driving to work. Of course this assumes you're running a deep cycle battery, but I digress. Solar is still cheaper, and still almost nobody uses it.



Conclusion

So, can it be done? Certainly. Is it practical? Sadly, no. Not even close. It seems TEG efficiency will need to increase and/or price must decrease a good amount before this would be viable as a DIY project.

Substitute the word battery for TEG in your conclusion, and it sounds just the the current series of arguments in EV vehicles.

Could you do this instead on the exhaust? You would get a much better hot/cold side differential.

https://www.sciencedirect.com/scienc...14157X14000033

Fig 5 is a Honda prototype:
https://www.sciencedirect.com/scienc...5943111501128X

^^It's literally in the first line of his first post:

https://ecomodder.com/forum/showthre...eat-35394.html

So if you mounted TEGs on both an aluminum radiator (side tanks) and the exhaust, you could just about do away with the alternator, but you’d pay over $2k for it...

Yeah! I think it is pretty awesome that we COULD do it if we either didn't care about money, or landed a nice score of super cheap or donated TEGs...

Use solar panel and teg to do an alt delete.
Use teg so the battery doesn't drain as fast while driving.
Then solar panels can replenish the battery during the day.
The teg would also continue to make some power after the car is turned off.

But if you have an electric vehicle you don't have to worry about any of that.

I would not want to wire that thing.

The cooling would also be a nightmare. You would probably need to redesign the radiator. Maybe make it flat and put it under the car.

solar already done :D

I've been considering both TEG sides .. My ideal (might not end up being doable) would be to combine both (radiator and exhaust) TEG with a Meredith Thrust style air flow over the 'cold' air .. as the 'cold' side air absorbs heat via the TEG transfer , that cold air will heat , and expand .. and if designed right produce minuscule additional amount of thrust at the same time from the same heating of that 'cold' air :D

https://ecomodder.com/forum/attachme...5&d=1538603669

I think the easiest way to do it would be to pipe some coolant up to the front of the car that way you can heat the hot side and cool the cool aide with outside air.

Of course, put that many TEGs on the cooling system and you are turning so much coolant heat into electricity that the cooling effect on the engine could adversely effect fuel economy by keeping the engine quite a bit colder than desgined. You'd need to adapt the radiator size, too, I think. More cost. Better to stick to exhaust waste heat.

To make 1kw you would need 3,412 BTUs if 100% efficient. So maybe 25% efficient so 13k BTUs.
If I remember correctly they aren't really efficient.
I think 100w is more realistic so you would only lose 1.3k BTUs. Which is a fraction of the heat you lose using the heater with the windows and the system on recirculation.

Plus if you build your teg unit up front and pipe the coolant to the unit you could control coolant flow with a valve and not open the valve till the engine warmed up.

Yeah, TEG efficiency is pretty horrific actually. Wikipedia says:

https://en.wikipedia.org/wiki/Thermo...tor#Efficiency

But in combustion engines it is basically free energy

How durable are they ?

What would be roi? Or do they fail before roi?

FWIW, it's been done before, and researched.
https://link.springer.com/article/10...664-011-1580-6
I bet they meant 'alternator'. But 75 Watt ain't enough.

But one could simply buy a liquid to liquid type generator: https://thermonamic.en.alibaba.com/p...EG600_48V.html
600 Watt, that's more like it. 48 Volt, ah well that needs a buck converter.

Pity it weighs 75 kilograms. No free lunch :/

Put a switch on it, and you could have a control-alt-delete!

If an ICE throws away around ~70% of fuel energy .. Just ~8% of that ~70% would be about ~5% of total original fuel energy .. such a ~35% ICE+TEG combination is a ~16% increase over such a ~30% ICE alone .. Such a ~16% increase to a ~50MPG driver would mean ~58MPG , same ~16% increase to a ~80MPG driver would mean ~92MPG, etc .. Although I myself suspect around ~10% increase over ICE alone is probably a more reasonable/realistic limit of DIY feasible for such waste heat recovery.

Let's say this is for a liquid cooled stationary backup generator and size, weight aren't really a problem.

Also a MPPT charge controller might be a good idea, since the delta T power curve looks kind of like what you get out of solar panels at different light and weather conditions.

With an mppt controller you could get any where from 120% to 200% or more the power compared to just wiring them straight to a battery. Also you won't have to worry about over charging the batt.

Any idea how long thy last?

Wow
I use TEG's similarly, but on my home furnace to get 50 free miles per day for milk and eggs runs!

Cost?

Also How do you get the 300c necessary to fire these up at full power?

Exhaust?

I would be using it at full power.
Probably heat it with liquid coolant.
If I use 300C the wires insulation would melt.

From what I've read .. baring some type of mechanical failure .. the semi-conductor part itself of most TEGs .. Roughly the same rate of degradation as most solar cell semi-conductor degradation .. about ~0.5% to ~1% per year (+/- context and material specifics).

Mechanical failure can result from allot of sources .. vibration , temperature changing cycles of expansion/contraction , sheer , impact , etc.

Soo , in ~20 years of use , it would be roughly 'fair' to assume you will drop down / loose somewhere around ~15% reduced power output (all else being equal).

I get the over 300c to fire up our TEG thermalelectric boiler by burning wood pellets, then through a grid tie inverter we put the energy from the wood pellets into our Nissan LEAF to the tune of 2 free round trips to town per day. Meanwhile the heat produced makes hot showers and goes I to our home hydronic heating loop.
Total system efficiency is near 100%

Do you have a picture of your setup?
How many modules?

105 40x40 mm TEG modules in this boiler from hastyheat.com
https://youtu.be/oTijbSC8y6A

Interesting video series on burning a TEG up by overspec thermal loading...
Synopsis if you have a good cooler you won't melt the solder out, so pour the heat to them...
https://youtu.be/Ya40vwpTCf4

So you have to have a solid cooling system.

I went down the TEG rabbit hole again today in my mind and did a little reading. This is an especially tantalizing quote from an article with some good images and research summaries, too:

"The Honda system used a simple design of a thin flat rectangular box with TEGs placed on the top and bottom surfaces. Liquid cooling was used in this design. The system consisted of 32 30 mm × 30 mm TEGs and produced a maximum of approximately 500 W. The claimed fuel consumption reduction is 3%."

SOURCE: B. Orra, A. Akbarzadeha, M. Mochizukib, and R. Singhb, "A review of car waste heat recovery systems utilising thermoelectric generators and heat pipes," Applied Thermal Engineering, Vol. 101 (2016), 490-495.
https://www.sciencedirect.com/scienc...01128X#bib0060

An example of someone bringing a 100w TEG system to market. An ICE is kinda like a stove, right? Quite pricey at +$1000.00 USD, especially without TEGs included:

"Introducing the first plug & play TEG Water Cooled System with an accessory solar input, battery charger and patent pending prioritization for active cooling. Our Patent Pending Tegpro Multi TEG & Solar Battery Charger with Pump Priority is a mulit-input charge controller that prioritizes power to the active cooling system onboard. It also displays voltage and current while charging your battery system.With the factory settings the unit can charge up to 14.4 VDC (can be set at factory set from 12 - 28.8 VDC, so please ask tell us!) and will accept up to 30 V DC inputs. ..."

https://172.98.64.46/thermoelectric-...oled-teg-sytem

Cheaper 100W one 'made in China' .. plumb the coolant fluid flow and apply heat.

https://ecomodder.com/forum/showthre...tml#post583882