Log In   |   Sign up

New User Registration

Article / Abstract Submission
Register here
Press Release Submission
Register here
coolingZONE Supplier
Register here

Existing User

            Forgot your password
December 2005
library  >  Application Notes  >  Tony Kordyban

Everything You Know is Wrong -- PART XV


answers to those doggone thermal design questions
by tony kordyban


dear mr. know-it-all,


i think that i have the basic idea of a thermocouple. you make a circuit with wires of two different metals, heat one connection and cool the other, and you get a voltage. or you can run it backwards by pumping electricity through such a circuit, so one end gets cold and the other gets hot. but i have never been able to get a common sense, intuitive explanation of this effect from anybody. can you help me to get a better feel why heat and electricity are related in this way?


mark time from sector r


dear mark,


thermocouples sure do seem simple, don't they? twist together a couple of wires and you have an electronic temperature sensor. they are simple to make, and simple to use. but why they work is far from simple or intuitive.


i have taken thermocouples for granted for a long time, so i refreshed my memory about the physics behind them by re-reading manual on the use of thermocouples in temperature measurement (astm stp 470b). now i wish i hadn't. my intuition about them vanished in a cloud of integrals and greek letters.


first there is the seebeck effect. thomas seebeck quantified the voltage generated by a temperature difference in a circuit made of different metals.


then jean peltier figured out that if he imposed a current on such a circuit, he could induce a temperature difference. but these effects are not exactly two sides of the same coin.


william thomson (lord kelvin, the guy who invented absolute zero) deduced the relationship between the seebeck and peltier effects, but only if he neglected resistive heating in the wires.


it took the invention of the idea of entropy (and what concept is less intuitive than entropy?) before a complete physical explanation could be developed by lars onsager. he managed to tie together a bunch of complicated processes mathematically. look at what you've got in a thermocouple: two metals with different physical properties, a temperature gradient and a voltage gradient across them. you have the seebeck voltage, the peltier current, the kelvin temperature rise; plus there is heat conduction in the metals, and irreversible heat generation due to current passing through the resistance of the wire. that is too much stuff going on for me to keep straight in my head, much less get a good gut feeling about.


onsager did give me one intuitive idea though. he began his explanation of the thermocouple by saying something a little like this, "with electricity or heat, what we've got here is stuff flowing through a pipe. you have a driving force, and something flowing through a restricted path. for electricity voltage is the driving force, and current is the stuff moving down the wire. for heat transfer a temperature gradient is the driving force, and heat energy is the stuff flowing. it's only natural to think that when you have two flows trying to go down the same pipe, they are going to interfere with each other." perhaps i paraphrased a bit.


if you have to think about thermocouples intuitively, imagine a racetrack running the kentucky derby and the indianapolis 500 simultaneously. neither race is likely to turn out the same way it would have alone.


dear mr. kordyban,


the instruction sheet that came with my handheld thermocouple meter gives some rules of thumb for temperature measurement. one says, "use proper gage wire and carefully route the wire away from the object to be measured to prevent the thermocouple wire from acting as a heat sink."


what the heck are they talking about? can a thermocouple be a heat sink? if so, why don't we just stick wires on our hot components instead of those clunky aluminum extrusions?


betty jo from the old sayme place


dear betty jo,


i don't know why engineers bother going to college for four or five years. i'll admit that most of what i use these days i got from sales tutorials and donut-baited short courses.so i'm not surprised that everything you know about thermocouples comes from a meter instruction sheet.


in this case your meter instructions are right, if not very detailed. a thermocouple wire can act as a heat sink, leading to measurement error. the detail they left out (probably to save room to repeat the content in spanish, french, german, japanese, and basque) was that the wire acts as a heat sink for the bead (i will use the term bead interchangeably with junction) of the thermocouple itself, and not necessarily for the object you are trying to measure.


what's the difference, you ask?


the thermocouple, at best, gives you the temperature of the junction. what you want to measure, in the example i'm going to talk about, is the surface temperature of some solid object, such as the plastic package of an electronic component. to say that the thermocouple is measuring the solid surface temperature, you have to make sure that the junction and the surface are at the same temperature, and also, that the surface is still at the same temperature it would if the thermocouple were not there.


i'll give you an example of the obviously wrong way to do it. you make a thermocouple from a pair of 16 gage wires, welding them together to make a bead about the size of a penny. this would be ok for measuring an engine block, but let's say you want to measure a sot-23 package (a tiny component much smaller than a penny.) you have now attached a large piece of metal to the top surface of a component, with most of its surface area sticking up into the cooling air stream. obviously, this wire, whose size is significant compared to the object being measured, changes the effective surface area of the component just as much as any ordinary heat sink would. the wire will cool the component. the wire might also interfere with the normal air flow pattern. the reading you get will not represent the sot-23 temperature when the thermocouple is not there.


you can minimize the heat sinky-ness of your thermocouple by making your wire as poor a conductor of heat as possible:

  • use metal pairs that have the lowest possible thermal conductivity. iron, for example is 12% as good a conductor as copper, so iron/constantine is a better choice than copper/constantine.
  • use the thinnest wire that you can tolerate. conduction along a wire goes up with the square of the diameter.
  • use wire that is jacketed with thermal insulation in the region close to the bead, where there is the biggest change in temperature. if the surface of the wire is not exposed to the cooling air, it will act less like a heat sink. thick insulation is not such a great idea if the bulk of the wire will disturb the air flow around the component.


these ideas will prevent your thermocouple from changing the overall temperature of the component itself. but they don't totally eliminate "heat sink" measurement error.you still have the possibility that the wire, as thin as you make it, will act as a heat sink for the bead. the overall average temperature of the component surface doesn't change appreciably, but the wire, depending on how it is routed away from the bead, can cause a small "cold spot" right at the bead itself, leading to a measurement that is lower than the surface temperature of the component.


in the next three pictures color represents temperature. figure 1 gives an example of a chunk of plastic, internally heated, sitting in a cold air stream. undisturbed by any curious engineers with probes, the surface temperature is about 73 degrees c. ideally, when you stick a probe on it, you'd like it to read 73 degrees c, too, plus or minus a degree or two. 



figure 1. a solid chunk of plastic has a heat source inside at 100 degrees c.
in an air stream at 1 m/s and 0 degrees c, the surface temperature is about 73 degrees c.


figure 2 shows the same chunk of heated plastic, with a thermocouple attached to the surface. in fact, i've shown the bead of the thermocouple buried into the surface a little bit to minimize convection from the bead. but the wire has been carelessly routed straight up into the air stream. the thermocouple is very small compared to the component, and so the heat conducting up the wire does not change the overall temperature very much. in fact, the surface temperature is still about 73 deg c almost everywhere. but notice that the bead, and a small area around the bead, are quite a lot cooler than the rest of the surface. the thermocouple reads 34 degrees c, instead of 73 degrees c.because plastic is a poor conductor of heat, it can have a fairly large internal temperature gradient even with the small amount of heat sucked away by the cold thermocouple wire. 



fig. 2 - the wire routed perpendicular to the hot surface cools the bead, creating a local "cold spot",
and a pretty gigantic measurement error of the surface temperature.


figure 3 shows how to minimize this effect. route the wire away from the bead along the hot surface, instead of straight up into the air. that way the portion of the wire leading away from the bead will be at the same temperature as the bead. heat won't conduct along the wire if there is no temperature difference. problem solved (or at least reduced)! 



figure 3. running the wire along the hot surface reduces conduction out of the bead,
and helps the bead stay close to the true surface temperature.


there is still a local "cold spot" where the wire eventually parts company with the surface. that is ok, because the cold spot is not at the bead, and so the bead will measure something much closer to the true surface temperature.


to answer your last question, i guess you could use wires instead of extruded aluminum fins to cool your hot parts. isn't that what pin fin heat sinks are? (see chapter 6 in my book for my opinion of pin fins. hint: the title is "when is a heat sink not a heat sink?")







about tony kordyban


tony kordyban has been an engineer in the field of electronics cooling for different telecom and power supply companies (who can keep track when they change names so frequently?) for the last 20 years. maybe that doesn't make him an expert in heat transfer theory, but it has certainly gained him a lot of experience in the ways not to cool electronics.


he does have some book-learnin', with a b.s. in mechanical engineering from the university of detroit and a master’s in mechanical engineering from stanford. in those 20 years tony has come to the conclusion that a lot of the common practices of electronics cooling are full of baloney. he has run into so much nonsense in the field that he has found it easier to just assume "everything you know is wrong" (from the comedy album by firesign theatre), and to question everything against the basic principles of heat transfer theory.



tony has been collecting case studies of the wrong way to cool electronics, using them to educate the cooling masses, applying humor as the sugar to help the medicine go down. these have been published recently by the asme press in a book called, "hot air rises and heat sinks: everything you know about cooling electronics is wrong." it is available at https://www.amazon.com/hot-air-rises-heat-sinks/dp/0791800741. this advice column is an extension of that educational effort. 


Choose category and click GO to search for thermal solutions


Subscribe to Qpedia

a subscription to qpedia monthly thermal magazine from the media partner advanced thermal solutions, inc. (ats)  will give you the most comprehensive and up-to-date source of information about the thermal management of electronics


Submit Article

if you have a technical article, and would like it to be published on coolingzone
please send your article in word format to [email protected] or upload it here

Subscribe to coolingZONE

Submit Press Release

if you have a press release and would like it to be published on coolingzone please upload your pr  here

Member Login

Supplier's Directory

Search coolingZONE's Supplier Directory
become a coolingzone supplier

list your company in the coolingzone supplier directory

suppliers log in

Media Partner, Qpedia


Heat Transfer Calculators