Talk:Thermoelectric cooling

Part Number Identification removal
I've removed the part number identification because, while well intentioned OwenVersteeg, it is far from universal and inclusion here is misleading. The only reference is a bunch of ebay sellers - this represents one anonymous brand and is far from universal. There is no written ISO or industry standard to support this, and numerous counter-examples (including the picture at the top of the article): Marlow part numbers: http://www.marlow.com/products/thermoelectric-modules/single-stage.html Laird (Melcor) part numbers: http://lairdtech.thomasnet.com/viewitems/thermoelectric-modules-2/-series-peltier-solid-state-thermoelectric-coolers http://www.lairdtech.com/Products/Thermal-Management-Solutions/Thermoelectric-Modules/#.U_j1VEspVR1 Morcheeba (talk) 20:14, 23 August 2014 (UTC)


 * Well, Marlow Peltier element part numbers aren't even internally consistent (they used three different formats on the page you linked) so that doesn't count; additionally, the Laird models linked don't really have "model numbers" - they have a series identifier and a comma-separated list of properties. Secondly, the picture at the top of the page does use the format described in the image - look again. (It features a 6.0 tacked onto the end, but the diagram still clearly would indicate the specs of the element.) Thirdly, it's not just "one anonymous brand" that uses the standard; the vast majority of the Peltier elements I have seen "in the wild" have been labelled thusly and there are many different brands that use this identification method or one very similar (for example, one common modification is for the elements to have more dashes.) In fact, the vast majority of Peltier elements use the numbering scheme described. There are 274,000 Google results for the Peltier element ID TEC1-12706, with many relevant YouTube videos, blog articles, and Amazon results. There are 8,360 for the Laird item no. CP08,127,05,L1,W4.5 (first one listed) with most being spam or duplicates of the spec sheet, and there are (wait for it) 173 Google results for the Marlow part # NL1010T-01AC (also first one listed) - half of which are spam and the other half duplicates of the specifications PDF.


 * In short, Peltier element IDs conform to this specification the vast majority of the time. The three arguments you gave were, respectively: flat out wrong (picture @ top of page does match diagram), not valid (Marlow part numbers aren't even internally consistent), and obscure/mostly irrelevant (Laird is fairly uncommon - 8k vs 274k results isn't even close, and those 8k are 95% duplicates/spam). As a result, I'm reverting your edit to remove the diagram. OwenVersteeg (talk) 02:39, 23 September 2014 (UTC)


 * And yet despite what image caption claims, not all TECs are labeled according to this supposed universal labeling scheme--128.138.158.128 (talk) 18:18, 9 January 2015 (UTC)


 * "Well, Marlow Peltier element part numbers aren't even internally consistent" -- exactly my point. There is no universal numbering system. Please reference where I can get a hold of this international standard. Or, if it's a defacto standard, can you point me to a list of companies that use it? Claiming "the vast majority" is a stretch; I've provided two major players that do not use it. The only one that uses the system listed seems to be HB/Hebei International Trading ... http://www.hebeiltd.com.cn/peltier.datasheet/TEC1-12706.pdf ... Can you list at least three more original manufacturers using a similar system so that we show? That would bring the number to 4 of 6 -- 75% --- still far away from "vast majority".  I understand that Hebei sells a lot to europe and may the only manufacturer you encounter in your everyday usage, but I haven't seen one in the united states and see other brands.
 * On your other points: "8k vs 274k results isn't even close, and those 8k are 95% duplicates/spam" ... what do you think are the duplicate/spam numbers for the 274k results? You do not make a strong argument here comparing dissimilar things.
 * Laird had revenue of USD$132,000,000 in their thermal division in 2014 ... http://www.laird-plc.com/laird/ar2014/ ... That does not sound obscure and irrelevant.
 * ... But my main point is that this is not a universal system and I have provided counter examples. You set a very high bar with "vast majority" and need to provide far more OEM manufacturers who do this, not just search results. 128.138.158.128 agrees with me. What should we do? Morcheeba (talk) 16:47, 25 April 2015 (UTC)
 * ... for those interested by the non universal numbering system, it is there: https://en.wikipedia.org/w/index.php?title=Thermoelectric_cooling&diff=prev&oldid=622265092 — Preceding unsigned comment added by 92.169.18.219 (talk) 14:08, 8 May 2016 (UTC)

SLT article
The SLT article on the peltier effect air conditioner was a bit short on details. I'm puzzled as to why this is considered a new or patent worthy invention, as peltiers have been discussed as airconditioning and refrigeration devices for years. The problem has always been the poor performance(10% of carnot?) of thermo-electric effect devices compared with compression cycle or absorbtion cycle coolers (40% of carnot). Is there any other more detailed information? njh 12:12, 5 August 2005 (UTC)

Efficiency can be higher
Somebody with writing talent should modify the section about how efficient a peltier is. The real efficiency is not so cut and dry as it is made out to be in this article. A peltier (TEC) is more efficient as voltage decreases. When you get down to about the 20%-of-maximum range, a typical commercial TEC is more efficient than even traditional phase change often is. A TEC at 20% will typically move ~5.7W of heat for every 1W emitted as heat in the device's own operation. At maximum voltage, that same TEC might move 0.5W of heat for every 1W it emits as heat. —Preceding unsigned comment added by 72.24.104.194 (talk) 15:34, 10 April 2008 (UTC)


 * The efficiency numbers you quote sound highly unlikely. 50% at full power sounds plausible (50% is really low for a cooling apparatus), but if it was 570% at 20% power, that would be 1.14W for 0.2W input power for the same device. Why would anyone put in 1W if the work done would be less then when they would put in just 0.2W? Besides, 570% is extremely high, so why do you think engineers would build elaborate installations, if a whole lot of Peltier elements at 20% of their maximum power would give them a far better overall efficiency? Besides all this, the temperature differential is important when calculating the efficiency; when the temperature differential rises (trying to cool something in a hot environment), the efficiency gets worse. You need to at least specify these conditions when throwing numbers around. 92.64.106.106 (talk) 16:34, 1 September 2008 (UTC)


 * I read in a forum that COP could indeed be quite high, 4 to 10, provided the difference in temperature is low (less than 10K) and the voltage used is also low (10% of max).
 * I looked for data supporting this claim and found this datasheet https://peltiermodules.com/peltier.datasheet/TEC1-12706.pdf
 * You will notice that the resistance of the module keeps more or less the same, around 2 ohms. A 6 amp current will produce 12 W heat. What changes is the amount of heat moved ~(60 / 50 / 40 / 30 / 20 ) depending on the difference in temperature ~(0 / 10 / 20 / 30 / 40 ), resulting in very different coefficient of performance, greater than 5 indeed.
 * So the reason these are not used is somewhere else. For a fridge, you usually need close to 0C in the main and -18C in the freezer, while the cooling unit at the back of the fridge is really hot (50 C ?) : this translate into >50K temperature difference, where Peltier is really bad. For air conditioning, the supporting units (pump/fan to move water/air) also come at a price, and would be the same, so it probably make little sense to have them supporting a Peltier device working at 10% power, or to oversize them so that they keep the difference in temp low enough — Preceding unsigned comment added by 2A01:E0A:1DC:4570:2566:6E5B:6BAD:B78E (talk) 12:28, 25 October 2019 (UTC)

Removed Hydroden power generation
I've deleted the reference to a hydrogen using power generation device because it is not real yet and is not the same thing. Tchannon (talk) 15:04, 4 April 2010 (UTC)

Temperatures
It would be very informative if this article gave some values of temperatures that can be achieved using thermoelectric coolers to cool a system that otherwise would be at room temperature. I imagine that there are all sorts of devices and methods that have a range of temperatures, but even general range of temperatures achievable would be helpful. How does it compare to cooling by liquid nitrogen? This should be in the article.

Longevity or Life of device
as of June 2010, no reference has been made to the lifespan of these devices, do they degrade or lose efficiency over the years, or do they last for ever ? 79.67.5.229 (talk) 09:12, 22 June 2010 (UTC)

Efficiency
"Because heating can be achieved more easily and economically by many other methods, Peltier devices are mostly used for cooling."

I'd dispute that. Certainly a furnace or electric heater is easier, but economical? Certainly in terms of energy input a heat pump, no matter how inefficient, will produce more heat than the energy input alone would suggest. sitharus (talk) 20:54, 3 January 2011 (UTC)


 * The word "economical" means either "less money" or "less energy". I guess here it's "less money". Feel free to reword, for example "...can be achieved more easily and less expensively...". :-) --Steve (talk) 21:52, 3 January 2011 (UTC)
 * I rewrote. :-) --Steve (talk) 22:38, 3 January 2011 (UTC)

Citation regarding the use of peltier systems in CCD sensors
Will a product specification page from a ccd manufacturer saying that their product uses Peltier cooling will count as citation? Check the 5th paragraph of link given below, titled "Chilled" http://www.olympus-sis.com/en/1610_7101.htm

similar info can be found on many manufacturer's pages — Preceding unsigned comment added by Ipcamit (talk • contribs) 12:39, 13 December 2012 (UTC)

"Written like a personal reflection" - is this a necessary tag?
I can't find any evidence of this anywhere in the article, so if nobody contests my decision in the next week I'll remove the tag. It seems highly unnecessary. OwenVersteeg (talk) 04:55, 14 October 2013 (UTC)

The lack of citations has been resolved
I've resolved the lack of citations in all obvious places. I've cited "Some Aspects of Peltier-Cooler Optimization Applied for the Glove Box Air Temperature Control." as well as some other reputable sources.

If nobody objects before the end of the week, I'll remove the "Citations needed" flag at the top of the page. OwenVersteeg (talk) 10:37, 15 October 2013 (UTC)

2nd paragraphe:
This technology is far less commonly applied to refrigeration than vapor-compression refrigeration is. The main advantages of a Peltier cooler (compared to a vapor-compression refrigerator) are its lack of moving parts or circulating liquid, near-infinite life and invulnerability to potential leaks

Under "Uses":
TECs have limited life time. Their health strength can be measured by the change of their AC resistance. When a TEC gets "old" or worn out, the AC resistance (ACR) will increase.

Seems to contradict itself. I am no expert in the field, but some correction or clarification would be helpful.


 * I also don't understand this. What causes them to wear out? They have no moving mechanical components and so should be very long lived. Am I missing something? 173.33.95.196 (talk) 02:04, 10 March 2015 (UTC)

Temperature difference
The temperature difference of 1 celsius degree is exactly like the temperature difference of 1 fahrenheit degree and 1 Kelivin degree!

External links modified
Hello fellow Wikipedians,

I have just added archive links to 1 one external link on Thermoelectric cooling. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Added archive https://web.archive.org/20130308061031/http://www.tellurex.com:80/technology/peltier-faq.php to http://www.tellurex.com/technology/peltier-faq.php

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

Cheers. —cyberbot II  Talk to my owner :Online 04:34, 18 October 2015 (UTC)

Exact Physics Needed.
The Peltier effect has not really been covered in the article. The difference in electron densities in the P and N materials is not a sufficiently good explanation. To alter the temperature there must be a change in the mean kinetic energy of electrons. The total density in related only to the total heat capacity.220.245.43.121 (talk) 03:02, 22 November 2015 (UTC)

Thick plate with long/tall electrodes?
As I understand it, there is no actual heat conduction occurring in the metal rods between the junctions. The heating and cooling of the junctions has no involvement of the metal rods to transfer heat. The rods conduct electrons between the junctions and that is all. (Or is this wrong?)

Therefore apparently it should be possible to make a "thick" plate, with insulation or convolutions of the rods between the junctions, to reduce thermal conduction from the hot side to the cold side and which reduces overall efficiency.

Basically, imagine one of these with the plates say 10cm apart, and 10cm long conductors between the junction ends. This should be much more efficient because parasitic thermal losses must travel further and more slowly through the 10cm long conductors.

I don't know if there is an increase in thermal resistance and heating from making the rods longer.

Also it is unclear to me if the rods need to be as thick as the junctions, or could actually be much thinner, to further reduce thermal-conductive losses. The rods are typically several times the cross-section of the wires providing power to the plate. -- Dale Mahalko, Gilman, WI, USA -- DMahalko (talk) 01:35, 3 May 2016 (UTC)


 * The trouble is, the longer the rod, the higher resistance, the bigger Joule effect. While the heat production at one end / heat consumption at the other end will remain just the same. So, no. You want the rod as short as possible
 * rods are stacked, which basically as the same effect as enlarging them
 * Gem fr (talk) 11:13, 27 October 2019 (UTC)

Equation consistancy?
Is the Seebeck constant in the first equation and the peltier constant in the second equation the same constant? If so, they should also likely be represented as the same constant Fort33dople (talk) 14:09, 29 April 2019 (UTC)
 * I don't see any way those could even be dimensionally consistent. Why would you think they might be the same? Dicklyon (talk) 14:43, 29 April 2019 (UTC)

efficiency numbers should be up front
This sentence should be in the first paragraph or so: In refrigeration applications, thermoelectric junctions have about 1/4 the efficiency compared to conventional means (they offer around 10–15% efficiency of the ideal Carnot cycle refrigerator, compared with 40–60% achieved by conventional compression-cycle systems (reverse Rankine systems using compression/expansion).[11]) — Preceding unsigned comment added by Encyclopedant (talk • contribs) 20:38, 3 June 2020 (UTC)

Collaboration to improve article citations
Hi everyone,

I have noticed there is a citation tag on the article. I am going to start cleaning this up and finding some reliable sources. Would anyone be interested in working together on this?

Thanks!

Starlights99 (talk) 16:11, 8 April 2023 (UTC)

Is this true? is this supported?
from the article:  "However, a well-designed Peltier cooler will be a mediocre thermoelectric generator and vice versa, due to different design and packaging requirements.". I have some experience with peltier devices. I do not understand the basis for this claim. I think this should be clarified. 75.164.53.14 (talk) 18:08, 29 September 2023 (UTC)