TEGS - Thermoelectric Generators

[Buster]

I think a new thread is warranted on the subject of TEGS.
I really have no idea of their worth but I'm sure a few of our more electronically minded members will be able to shed some light on their value.

If they can recover some of the heat lost to produce enough electricity and use this to produce HHO it could possibly be worthwhile?

http://www.ieverred.com/TEPGR.html

[Philldpapill]

TEGs generally use devices called "Peltier Junctions". Basically, they require a temperature differential(one side hot, while one side is cool) to produce electron drift. This electron drift creates a voltage and current if attached to a load. They are generally used in reverse, where a current is supplied, and the device CREATES a temperature differential.

However, PJ's are pretty expensive... And horribly inefficient when using them in the energy production side(<10% and usually on the order of about 4-5%). On top of that, they are expensive and bulky. Be prepared to pay at least $2/watt. Doing the math on that, an extra 100W is going to be at least $200, and that's just for the PJ's alone. 100W equates to about 8A, and that is assuming you already have a very hot surface to mount these to, and a fairly cool heat sinking material/system. Chances are, you're going to use about half the energy these things make(if not more) just on driving the cooling system. So, in reality, you're looking at about $4/Watt - best case scenario.

Even if you had an extra 200W(which is pretty cool...) from these TEGs, is it really worth the $400 investment and up-keep? IMO, the technology/Seebeck effect is pretty 'cool', but not practical in this sort of thing. You're better off just using the power off the alternator. Richard was going on and on about MORE CURRENT = MORE HHO, which is true, but there IS a point of diminishing returns... Even if you could produce 100lpm, there is point where more HHO injection doesn't result in much more MPG.

Regardless, these things are fun to play with and dream up ideas.

[Helz_McFugly]

I was looking to buy one for camping just to play with. Maybe run a light or radio off of

[Roland Jacques]

I can only see using them on the exhaust system. Cant see how you could maintain A 150 degree differential on the 195 degree cooling system.

I would like to find out more about them. Like how durable are they?

Like Phil i cant see any return on investment happening. But maybe looking at it as something to play with it would be cool. Maybe slightly more practical than having a solar panel on your truck/car.

[richard_lyew]

hello everybody, i was just about to make the thread when i saw it LOL so ima just copy and paste the links over here if you dont mind.

this is for Helz http://www.marlow.com/power-generators/

this is a good read http://www.nextreme.com/pages/power_...ste_heat.shtml

this is what im planning
just use the hot water coming from your engine to the radiator, pass it trough your TEG system's hot side and for the cold side use a big trans cooler with fans, to circulate the cold side i will be using a pump that is connected to the back of the alternator (that way i wont be using power to drive a pump) i will post a pic of the alternator with the pump on it soon(i changed it on my work truck and i keep the old one to mod my present alternator)

this is what BMW is doing, also look at the links on the side
http://www.youtube.com/watch?v=bkP7v8yYivQ
http://www.youtube.com/watch?v=yY0SAzs1LJw&feature=fvw

here is where i got my idea to use the hot coolant
http://www.espressomilkcooler.com/thermoelectric_generator___stationary_generator___ teg_.htm

i will edit this later and add more info :D

[richard_lyew]

I can only see using them on the exhaust system. Cant see how you could maintain A 150 degree differential on the 195 degree cooling system.

I would like to find out more about them. Like how durable are they?

Like Phil i cant see any return on investment happening. But maybe looking at it as something to play with it would be cool. Maybe slightly more practical than having a solar panel on your truck/car.

look at the link that Buster post in the first post, it will give you an idea about durability etc.. :D

[richard_lyew]

TEGs generally use devices called "Peltier Junctions". Basically, they require a temperature differential(one side hot, while one side is cool) to produce electron drift. This electron drift creates a voltage and current if attached to a load. They are generally used in reverse, where a current is supplied, and the device CREATES a temperature differential.

However, PJ's are pretty expensive... And horribly inefficient when using them in the energy production side(<10% and usually on the order of about 4-5%). On top of that, they are expensive and bulky. Be prepared to pay at least $2/watt. Doing the math on that, an extra 100W is going to be at least $200, and that's just for the PJ's alone. 100W equates to about 8A, and that is assuming you already have a very hot surface to mount these to, and a fairly cool heat sinking material/system. Chances are, you're going to use about half the energy these things make(if not more) just on driving the cooling system. So, in reality, you're looking at about $4/Watt - best case scenario.

Even if you had an extra 200W(which is pretty cool...) from these TEGs, is it really worth the $400 investment and up-keep? IMO, the technology/Seebeck effect is pretty 'cool', but not practical in this sort of thing. You're better off just using the power off the alternator. Richard was going on and on about MORE CURRENT = MORE HHO, which is true, but there IS a point of diminishing returns... Even if you could produce 100lpm, there is point where more HHO injection doesn't result in much more MPG.

Regardless, these things are fun to play with and dream up ideas.

yes Phill i know they are expensive in the western side of the world, this is why i buy most if not all my stuff from china. in the USA you could get 12VDC@each AMP for about $4 but in China i found places that sell the same 12VDC@ each AMP for $1 or less!!!!!!!!!!!!! so USA price for a 12VDC@10AMPS=$40 to $50 but in China the same would cost $10 to $15!!!!!!!!! that is a very very big difference. dont take my word for it, look at the link Buster post then try to find a place in the big mighty USA to beat that price. just try to beat this price http://www.ieverred.com/TEPGR.html then i will post a lower one. :D

[Roland Jacques]

Yeah those prices look much better than what I saw in the past. $22 for a potential 30amps :)

Richard, I dont see how a trans cooler with fan going to get it's temp any lower than the outside ambient temp. So you will have to use exhaust heat.

The question is still how would you effectively dissipate enough heat for the cold side? radiator trans cooler with pump would work.
Would normal computer type heat sinks work?
What happens when the differential temp is 300 degrees is there such thing as too much differential temp?
Can you give these things to much heat and melt wires...?

[richard_lyew]

Yeah those prices look much better than what I saw in the past. $22 for a potential 30amps :)

Richard, I dont see how a trans cooler with fan going to get it's temp any lower than the outside ambient temp. So you will have to use exhaust heat.

The question is still how would you effectively dissipate enough heat for the cold side? radiator trans cooler with pump would work.
Would normal computer type heat sinks work?
What happens when the differential temp is 300 degrees is there such thing as too much differential temp?
Can you give these things to much heat and melt wires...?

im planning on using cooling blocks like these http://www.espressomilkcooler.com/thermoelectric_generator___stationary_generator___ teg_.htm
and im modding my alternator to use it as a pump too and the hot side withh have copper blocks not aluminum. as for wires, im waiting for a quote from a few companies right now for custom TEGs 12VDC@10 to 20AMPS with the lowest AWG wires they can fit them with. as soon as i get my quote i will share :D

[richard_lyew]

http://www.youtube.com/watch?v=pgIOUXKyzFE&feature=related

[Philldpapill]

Let me get this straight..... Richard, you plan on using these TEGs to generate power, yet you even say you will require fans to keep the other side cool.

Basically, you're going to produce MAYBE 5W, and use up 4W to power the fans/pumps. Genius.

Also, Richard, for the love of god, please post your replies in ONE post. It clutters the forum when you do 500 replies. The forum software is set up to display 10 posts per page, so your habits tend to take up half a page when you could have easily said it all in ONE post.


As for your actual idea... It won't work. I can tell you right now. You just aren't going to get 12V@10A out of a small module. You will need about 1 sq. meter of junctions to get anywhere near that much power out of them. You do realize that is 120W, right?

EDIT: After looking at the fine print, you're right Richard - you won't find anything that beats those prices. The reason, is that you didn't read it right... They modules DO NOT put out 12V when you provide a temp differential. They have a Seebeck coefficient of about 0.036 to 0.042 V/K. LOL I don't think you understand these things all that well... The Seebeck coefficient is basically how much electron drift occurs per unit of temperature difference. Basically, if you had a 200degree C temperature difference, you would get around 4V or so out of the unit. That's 200 degrees C... Nearly 400F. Your wiring will melt. On top of that, that is 4V with an internal resistance of 0.35 ohms(model TEC1-12730T200). That means that the maximum power that this thing can crank out is 22.85W with an output voltage of about 2V(maximum power transfer theorem... Basic EE stuff).

You can try this if you want, but I'm telling you - it's a total waste of money and time. I did the math on this, man... Take my word or do what you want. Whatever.

[Helz_McFugly]

his fans run on love phill. phill I think Im going ot make you a sweat band with a TEG built in it and a string of red LED lights all the way around. when you get :mad: the lights will light up. There will be a photoshoot involved as well.

Man Dicky Poo is over in his attention thread wundering why Buster isnt getting cussed at because of his claims of 100% gains, while Dicky Poo claims 1000% gains an expects everyone to take his word for it. he doesnt understand the difference between logical/credible and prevarication. Im done with "woo hoo everyone look at me" thread because its no longer based on the original system which turned out to be fictitious anyways. :D

[Philldpapill]

I don't think he even cares about the issue anymore. If you'll notice, he bumped the thread 5 consecutive times when no one replied. There are many comments in those posts that are clearly trying to incite more "cussing". He's trolling the thread and the forum now. Yuck.

[richard_lyew]

Let me get this straight..... Richard, you plan on using these TEGs to generate power, yet you even say you will require fans to keep the other side cool.

Basically, you're going to produce MAYBE 5W, and use up 4W to power the fans/pumps. Genius.

Also, Richard, for the love of god, please post your replies in ONE post. It clutters the forum when you do 500 replies. The forum software is set up to display 10 posts per page, so your habits tend to take up half a page when you could have easily said it all in ONE post.


As for your actual idea... It won't work. I can tell you right now. You just aren't going to get 12V@10A out of a small module. You will need about 1 sq. meter of junctions to get anywhere near that much power out of them. You do realize that is 120W, right?

EDIT: After looking at the fine print, you're right Richard - you won't find anything that beats those prices. The reason, is that you didn't read it right... They modules DO NOT put out 12V when you provide a temp differential. They have a Seebeck coefficient of about 0.036 to 0.042 V/K. LOL I don't think you understand these things all that well... The Seebeck coefficient is basically how much electron drift occurs per unit of temperature difference. Basically, if you had a 200degree C temperature difference, you would get around 4V or so out of the unit. That's 200 degrees C... Nearly 400F. Your wiring will melt. On top of that, that is 4V with an internal resistance of 0.35 ohms(model TEC1-12730T200). That means that the maximum power that this thing can crank out is 22.85W with an output voltage of about 2V(maximum power transfer theorem... Basic EE stuff).

You can try this if you want, but I'm telling you - it's a total waste of money and time. I did the math on this, man... Take my word or do what you want. Whatever.

wow this guy follow me around this forum and try to tell me how much stuff will not work bla bla bla, ok phill i will use 8 CPU case fans that use 0.25AMP each so that will be a total of 2AMPS for the fans and for the pump i will be using the pump on the back of the alternator(almost zero drag). as for the TEGs i was not hoping to get full power out of them but if i can get half the amount of power each TEG all i would have to do to get what i want(200AMPS) is just add more TEGs also its a very good thing the coolant/water temperature is not up to the top temperature of the TEGs because that might lower their life span. phill relax man, this plan is full proof its just that people couldnt get TEGs at a low price before. i plan to buy a potential 400AMPS at about $0.50 per AMP so that would cost me $200 and i am sure i can get 200AMPS out of them with the temperature difference i will provide. so now phill 200AMPS worst case scenario and just 2AMPS for the fans for a total setup cost of about $300 and it works better than an alternator that's why BMW start using it plus it helps your car to save on power as the coolant WILL lose heat when pass through the TEG system so the radiator fan wont work as hard(it wont turn on as often to cool the radiator) saving power, plus adding power to the car's electrical system as the cells wont be using the full amount of power that the TEG system will be making because when the car is driving the cooling will be better than when its just parked. so no matter how you dont want it to work for me, it WILL :D it might never work for you but it WILL for me!!!!!! :D

[richard_lyew]

oh yea one more thing, phill this is not your thread or forum and im not using your computer so why do you think you can tell me what to do? :D i will post and you will have to love it and live with it for the rest of your life :D and i can use this smiley as much as i like :D take that and this :D LMAO!!!!!!!!!!!!!!!!!!

[Philldpapill]

*slaps palm against forehead in frustration*

Richard, go start your own thread or something. I don't want to repeatedly tell you WHY it won't work, and then have you come back with some sort of non-thought out, jury rigged plan that makes no mathematical/engineering sense.

The CPU fans are insanity. Did you not read a word I said? You just aren't going to get the needed power. You will need so many TEGs that the sheer weight of it all will start cutting into your gas mileage before you get any kind of gains! THAT is how ludicrous this is. Using one CPU fan on a single PJ may get you a 50-75 degree differential, MAYBE. At those levels, you're lucky to get 2-3V open circuit out of one PJ. If the internal resistance is 0.35 ohms, the maximum power you can get out of one of those is between 5.7W - 12.8W. Remember, this is the BEST CASE scenario. If you are using a CPU fan that is 12V, 0.25A, that is 3W being used up right there alone. Now, your net power is down to 2.7W - 9.8W. That upper limit SOUNDS pretty good, right?... That's at 1V - 1.5V. I SUPPOSE you could put some of these in series, but that isn't a good idea as some PJ's will be at different temperatures and this will cause problems which I don't want to go into.

Now, do you see that even with the BEST CASE, you MIGHT get a net of about $2 per watt? In reality, you're probably looking at more like 5W for one of those $20 units(the best ones in your link). Unless you are talking about buying 500 of these TEGs(about $10,000 at $20 a pop)... You know what - I take it back. It's just stupid. Plain and simple.


EDIT: Richard, you do realize that you can get 1,000A from a 0.1V source, right?

[richard_lyew]

ok here we go, this will end it. ok phill you are right it wont work, its not working for Nasa, BMW and none of the 100,000,000+ people who have buy them so it wont work for me eider, what was i thinking. you like this post phill? :D now get off my case, there i said its not working and will never work :D i hope ur happy now LOL. :D :D

[richard_lyew]

now back to the thread, i know for sure that an transmission cooler can get cooler than ambient air because i have worked on turbo charged cars and after driving the intercooler gets cooler than the ambient air and thats without and fans(just the breeze from driving) and even though a intercooler cools air or charged air, the same laws apply to both the transcooler and the intercooler.

here is some info as to how they work
http://www.gnttype.org/techarea/turbo/intercooler.html

i will add more links soon :D

[Helz_McFugly]

why dont you just get one already and stop yapping about it. Ide say prove what you say but we all know how that goes. You'ld think with all the money youre saving by getting 300MPG you would have already bought a small one to start testing.

Hey Richand. If you fill up your car with gas, how far can you drive it before you run out of gas? try not to beat around the bush, I know you say you never use more then a 1/4 tank so you should know how far your entire tank will get you. just answer the quesion with a number. :D

You would think, being that you are doing this for you, to save and make money like you made very clear, you would get a job as a courier, a taxi driver, a transporter of some type. you would rake in the $. I just wanted to through another log on your fire.

[HurstOlds]

Richard, your second link in Post #5 of this thread is not set up correctly, can you please fix? Thanks

[Buster]

There have been some pretty convincing arguments against this technology being much use to us. I must admit I have my doubts that it would be worth the effort of trying it out. I think, though, that Richard's enthusiasm will give it a good try, and you never know, maybe he can find a way to make it work effectively enough to be of use. I think he should be supported in the effort no matter how unlikely it may appear that it will be successful. Sometimes unlikely things can work in a certain way if we can find the angle on it.
Don't put a man down for trying. It's all good, even if just to make certain that that way won't work.

[Helz_McFugly]

Ide love to see something like this work, even though the math doesnt add up to much, but with Richard, he has a lack of showing that anything he speaks of working outside of his brain in the real world. just sayin.

[Philldpapill]

Yeah, you're right, Buster. I know from experience that sometimes things SHOULD NOT WORK, yet for some reason, they do. More commonly it's the other way around(with LOADS of hair pulling) where something SHOULD work, yet it doesn't.

However, I don't think this is one of those things where I say it won't work, and it just might. This isn't so much a clash of theory and science, as much as it is a clash of purpose. What I mean, is that we are talking about getting, what - 500W TOPS out of this TEG system? That's a decent gain(maybe 3.4% increase in available power, assuming an average power usage of 20hp or 14.92kW), but at what cost? Even if you could get these TEGS at $1/Watt, that's a $500 investment for 3.4% gains...

How long would that take to break even? Assuming you get 30MPG, and you drive 1200 miles a month, that's 40 gallons of gas. At $2.50 a gallon, that's $100 a month in gas. If you save 3.4% in gas due to the TEG, you'd be saving $3.40 a month. With the $500 investment, it would take over 12 YEARS to pay for it.

Sometimes, you just have to do a little math and think things through. Richard, you're more than welcome to try it out, but you may save some time and money by focusing on some more lucrative ideas. It's your time/money, so it's not costing me anything, so I'm happy. :)

[Buster]

Yeah, I think you're probably right on this one. It may need the technology to develop further before it is ready. A lot of stuff is like that, it just needs a bit more development before it's useful. I vaguely remember some new technology which took years before someone figured out a use for it. At least we have some use for this one if it becomes efficient enough.
Personally, I still think it's putting out the fire after it's burnt the house down. I mean, it makes more sense to me if we can utilise the excess heat to expand water vapour inside the combustion chamber, as many seem to be having success with doing, via water injection systems.

[Philldpapill]

Agreed!

On a different topic... Buster, WHAT is your avatar a picture of? Maybe it's my dirty mind, but it LOOKS like a boxer with it's junk in it's mouth... or at least trying to get there. LOL

[Buster]

It's of a stupid dog just about to bite it's own b####cks off.
Very painful it looks too!! Definitely not recommended.

[richard_lyew]

Richard, your second link in Post #5 of this thread is not set up correctly, can you please fix? Thanks

here http://www.nextreme.com/pages/power_gen/power_gen.shtml
sorry about that :D

[theegb]

Phil - My thermodynamics professor once told me that understanding units was the difference between an ME and an EE. I decided to get both. Richard has some fundamental flaws in some of his reasoning, but let me back up the TEG a bit (as I have a masters degree in Thermo acoustics, it's kind of my baby). The seebeck coefficient is in volts per Kelvin. So, using your example of S=.036, R=.35 and dT=200C (473K):

V=.036v/k * 473K
V= 17V not 4V (the unit advertises 15V)

V=IR
15V = .35I
I = 43A (unit advertises 30A)

P=IV
P = 30A * 15V
P = 450 W (unit advertises 289W)

I think if you apply your maximum power transfer theorem, you will find they are pretty accurate in their advertisement.

[loreoff]

I think if u attach the Tegs onto the headers u might beable to get a deferential great enough to use a simple radiator in computers

[Philldpapill]

theegb,
The voltage you are getting is based on the absolute temperature of only the hot side(right?). That would be accurate IF the cold side were near absolute zero... However, I was under the impression that the temperature we are interested in is the difference in temperatures of the hot and cold side. In the example, you don't have 473k difference - you only have 200k, which would yield about 7.2V OPEN CIRCUIT. That voltage, of course, drops off as you draw more and more current. The device can be modeled as a voltage source in series with the 0.35 ohm internal resistance. Granted, there are some discrepancies due to the non-linear seebeck coefficients over a broad range of temperatures, but they are fairly negligeable when compared to the differences in our calculations.

In your power calculations, you multiplied the open circuit voltage and the short circuit current... That equation is meaningless because the open circuit voltage won't be the same voltage when you short circuit the device - it will be 0V.

Now, using MY numbers:

Open circuit voltage = 0.036V/K * 200K = 7.2V
Using maximum power transfer theorem:
Pmax = (V/2)^2/0.35 = ~37W

I'm looking at the company website again, and I don't think the 289W is refering to output power... Since it's "Qc", it's refering to ACTUAL heat flowing through the TEG. That number should NOT be used in any kind of reference to the available power output, unless you are computing efficiency of the device. If this is the case, then the TEG is indeed fairly efficient at about 12.8%(37W/289W). Not TOO bad, but Richards idea about using it is still insanely stupid and worthy of the government giving him a disability check each month.

Anyway, I stand by my calculations. This thing doesn't put out anywhere near 450W @ 200K difference in temperature.

EDIT: That website is flakey... I've thought about it some more, and they are CLEARLY advertising these things to generate electricty, yet the numbers they give are not for generating juice, but instead, they are for creating a temp differential(i.e. using them to cool something like a CPU, etc.). Also, their pricing seems to be for very low volume customers(price break is for 10 units? Really???). That says they aren't interested in long term buyers, or volume buyers because those customers would see through their shady advertising. Shady, shady, shady... I wouldn't buy from them..........

[richard_lyew]

on my old wet cell system the TEC was working as a TEG because the wet cell would get so hot. i think recovering energy from wasted heat is very good. try not to be wasteful phill, dont waste current in your big PWM. i bet your PWM has a FAN but when i put a FAN on my cell you said i was wasting power but you have a FAN in your system too and you are wasting more power than i am through your PWM. think about that.

by the way everybody i might make a new thread soon. i have a new setup, drycell with no holes, great MMW, direct inject of HHO to engine, vacuum meter added, high AMP alternator added, 20LPM HHO output, inverting 13VDC to 120VAC then bridging it back to DC to spread across plates in cells. i find by doing this i get more volts to work with and pull the same AMPS or less. im no longer cooling my cells because now i am using 5 n plates and the cells don't get hot at all. im now directly injecting my HHO to the vacuum at the main intake valves. i am now at -50 at idle and -40 while driving and yes that means i can turn off my fuel pump at idle.

i am very very busy with other stuff so i will post the videos when i can. :D

[Philldpapill]

Actually, Richard, MY PWM doesn't use a fan at all. It can handle ~30A per Power Module Board(an area of about 2"x2"), and doesn't need a fan to stay cool. It gets slightly warm, but that is due to the natural ohmic loss of the conducting material, but nothing major. Each module dissipates just under 1W of heat, at the fully rated 35A.

Also, MY PWM is even more efficient than probably any other on the HHO market. It uses a high frequency switching, and power inductor/capacitor to demodulate the power into a very smooth output voltage/current. By doing this, you avoid the high amplitude peak currents that contribute heavily to power loss. In other words, it's about as efficient as physically possible.

Suck on that. BTW, I didn't read past the first paragraph. I'm sure it was junk.

[theegb]

Phil - You are correct. There is no way to produce the temperature difference needed to create the voltage they advertise. I also thought about it more, and signed on to correct my post when I saw your reply.

[Philldpapill]

Good to know, theegb. I was questioning my thinking, but I'm glad you agree now... Makes me feel more sane. :)

That kind of stuff seems to happen sooo much. Some companies will advertise stuff like that. It's either a marketing team with very little scientific/engineering background, or it's a knowingly deceptive claim. Either one of those is bad, and incredibly misleading. I was shopping around for some small solar panels on ebay, recently, and came across one seller claiming the cell puts out like 15V, and 1A, therefore "15W of solar power!!!1!"(I kid you not... they actually had a "!!!1!" in there...). I looked at the datasheet and those were open circuit, and short circuit values. The guy selling these was clearly misleading in this case because next to the wattage claim, he had an asterisk... Following the asterisk to the bottom even says that these values are OC and SC and "your actual power output may vary". The actual power output was probably around 10W, maximum, or only 66% of the claimed output. Pure BS.

Most people buying this sort of stuff don't know the difference. Nothing against the buyer, but that kind of example is why I chose engineering over marketing...

The goal of engineering: Design a device or process that performs above and beyond specifications.

The goal of marketing: Design a way of selling a device with specifications that are above and beyond how it performs.

Sorry if that offends any of you marketers. This TEG seller that we are referencing is a prime example of my personal disgust with the profession in general. Not that ALL marketing is bad, or anything... Just people that knowingly hide/distort specss.