TEC - 12076 - module

Hi guys. Am just wondering if someone of you knows "lifetime" of those Chinese TEC peltier modules. Would this work for two-three years under constant usage ?

Reply to
en2
Loading thread data ...

What's the failure mode supposed to be? Has anybody had one fail?

--
Bill Sloman, Sydney
Reply to
Bill Sloman

I imagine thermal cycling would be the most likely trigger of most failures - solder joint fractures. There are hundreds of individual solder joints. Could be deferred by feeding them a relatively smooth 'DC' current.

Of course there are probably lots of idiots who overheat them in muching about with them- like the guy who complained that the Peltier device was defective because both sides were getting hot, just one was way too hot to touch and the other was a bit less burny.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward" 
speff@interlog.com             Info for manufacturers: http://www.trexon.com 
 Click to see the full signature
Reply to
Spehro Pefhany

I seem to recall one TEC going open circuit at a place I worked. I think they were slowly PWMing it to control the temperature, which I have read is not good for them.

Condensation is also probably not good for them in the long term. Supposedly the silicone encapsulation lets water vapour in and won't let the liquid water out again (according to a TEC manufacturer who uses some other epoxy sealant!).

I would like to figure out a way of stopping the condensation as I would like a tiny thermal chanber for testing circuits in, possibly built using a cheap stainless steel vacuum flask as the insulation, with two stacked TECs and a water block instead of the stopper of the flask. I guess I could very slowly bleed in some dry gas (e.g. from my argon cylinder) but that is a nuisance. I wonder if someone makes hermetic TECs.

Chris

Reply to
Chris Jones

nt

they were slowly PWMing it to control the temperature, which I have read is not good for them.

It's actually a classic beginner's mistake. The thermoelectric heat transfe r - which you do want - is linearly proportional to current.

The resistive heat dissipation in the junctions (which you can't avoid, but can minimise by keeping the current tolerably stable) is proportional to t he square of the current.

Slow PWM is consequently a very bad idea. My 1996 paper does described a PW M system, but the PWM was fast-ish - repetition rate 17.4kHz - and the chop ped current was fed to the Peltier junction via big LC filter which blocked better than 99% of the 17kHz and higher ripple.

My boss had very vivid memories of somebody else's improvised PWM drive (wi thout filtering) which had got a Peltier junction hot enough to melt the so lder joints holding it together.

Reply to
Bill Sloman

TECs basically die from thermal fatigue, as Bill and others mentioned, but they are also vulnerable to cracking due to shear stress, either from shock or differential thermal expansion.

Bismuth telluride is pretty friable stuff.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs

For a moment I thought you were talking about "TECHs", I know some that think they're on the brink of destruction!

Jamie

Reply to
Maynard A. Philbrook Jr.

i have one additional question. Let's say we have controlled environment. On one side we have heating plate that constantly heats up to 50 degrees Celsius. On the other side we have some water cooler. Difference in the temperature would be around 30 degrees. How to calculate electricity output of that particulate module ?.

I have made some tests with candle and cooler and i got around 1,5V /

0.4A but the problem is that this kind of cooler is not adequate for such heat so...

To return to my question, is there some formula how to calculate electrical power when you have controlled environment and DT around 30 degrees ?

Reply to
en2

Dunno, since you didn't specify the usage (i.e. the working environment). I suspect the lifetime is limited by the number of thermal cycles, which might crack the ceramic. See: Note that the solder used is low temp 138C BiSn.

While not a thermo-electric cooler, I did find a great way to destroy three Peltier devices in succession. The devices were Melcor CP2-127-10 series devices, now owned by Laird: No lifetime or max number of thermal cycles specified on the original data sheet. What I was trying to build is a self powered fan for my woodburner: The way these work is the Peltier device sits between the 300C heat from the wood burner top, and the much cooler fan area (my guess is about 90C). The Peltier device is positioned for maximum differential temperature, but not so hot that the solder will melt. At 50C, the data sheet says 77C max differential temp. The trick is positioning the Peltier for similar operating temperatures.

My construction was similar to those in the photos, except that my base was shorter and thicker. That raised the temperature of the Peltier junction. The design also relies on the fan being operational to provide he necessary temperature difference. Stop the fan and eventually the temperature of the heatsink becomes uniformly too hot. My fan was too large, which had too much air flow in the Peltier area, which reduced the temperature differential across the Peltier, which stopped the fan, which caused the temperature to climb, which melted the solder. Self destruction, thermal runaway, negative TC, bad design, and creative aerodynamics, all at the same time. I destroyed three Peltier devices before I noticed what was happening.

With some shims and guesswork, I managed to make one that worked, which I gave to the ladyfriend. It has survived for about 10 years without further failures. Next project will probably be a Sterling Cycle wood stove fan:

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

Degrees C or F?

That would be the Seebeck effect. It's just the Seebeck coefficient (uV/K) times the temperature difference:

Nothing on the data sheet for generating power: You could calculate the Seebeck coefficient experimentally, but without corroberation from the mfg data sheet, it would be a guess(tm). My guess(tm) would be somewhere around 2 uV/K at room temperature, but that's a wild guess. If you're stuck, bug me and I'll measure what some of my Peltier devices produce.

Measure the hot and cold side temperatures with a thermocouple or thermistor probe (not an IR thermometer). Do the math.

Volts = Seebeck_Coefficent(uV/K) * delta_Temp(K)

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
 Click to see the full signature
Reply to
Jeff Liebermann

yeah, the solder cracks and the unit fails open. There's hundreds of solder joints in those things.

Reply to
Cydrome Leader

looks like you beat me to it. I have had them fail if not properly cooled, and that's on both sides.

Reply to
Cydrome Leader

Have you tried the web? There are like a bazzilion hits. Here's one.

formatting link

OK I didn't read all that...(lots of icky labview :^) so I don't know how they defined efficiency. But ~1% seems like a reasonable number. So how much (heat) power goes through your device, take about 1% of that number. Higher delta T's will give better efficiency.. that's just basic Thermodynamics

George H.

Reply to
George Herold

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.