0
$\begingroup$

My question is about the variation of conductivity in a volume of material and its effect on measured current. A volume is comprised of two metals joined symmetrically in a cuboid shape as in the image below. These metals have different conductivities: $\sigma_{1}$ and $\sigma_{2}$. Does $A_{1} = A_{2}$ or $A_{1} \neq A_{2}$?

enter image description here

From what I’ve read so far it seems to me that the average local charge density and drift velocity vary inverse with each other to keep the current density constant throughout the entire sample such that $A_{1} = A_{2}$. However, a friend suggested that diagram 1 is topologically the same as diagram 2 and therefore measured currents $A_{1}$ and $A_{2}$ would be different, where $R_{1}$ and $R_{2}$ are some function of the conductivity of each material and geometry (where $R_{1} = f \left( \sigma_{1}, \sigma_{2} \right)$, $R_{2} = g \left( \sigma_{1}, \sigma_{2} \right)$). I am confused what approach is correct: that $A_{1} = A_{2}$ or that $A_{1} \neq A_{2}$? I'm also wondering the best way to calculate the current measured at $A_{1}$ and $A_{2}$. enter image description here

Apologies in advance, this is a relatively new topic to me, so my understanding is lacking. If you have any suggestions on books or literature to read or look at that would be really appreciated. Many thanks for your help!

$\endgroup$
6
  • $\begingroup$ No need to apologize. Would you mind telling us where this came from? It's a very poor method to measure conductivities and in practice more or less useless. While the two circuit diagrams are equivalent (the voltage source was simply moved between the two current loops), it misrepresents what would actually happen in such a setup once we include actual wire and contact resistance. A much better method goes back to Thomson and Kelvin (and there are improvements over this, too): en.wikipedia.org/wiki/Four-terminal_sensing $\endgroup$ Commented Apr 24, 2023 at 15:39
  • $\begingroup$ Clearly in the limit that one conductivity or the other goes to zero the ammeters will not read the same... $\endgroup$
    – Jon Custer
    Commented Apr 24, 2023 at 15:42
  • $\begingroup$ @FlatterMann Thanks for your help! We are considering the possibilities with mapping conductivity in a long metal components as part of our research. At the moment its just trying to understand the nature of what ideas could be viable. When you say in practice it would be useless, I can imagine that could be due to being unable to measure sufficient resolution or high possibility for errors, or do you mean in resolving what the current measurements actually mean in terms of conductivity, or something else? Thank you for the reading suggestions as well-I'll look into them! $\endgroup$
    – jackw2556
    Commented Apr 25, 2023 at 9:24
  • $\begingroup$ @JonCuster Thanks Jon! That was my initial thought as well. I think I managed to confuse myself when looking into it further. $\endgroup$
    – jackw2556
    Commented Apr 25, 2023 at 9:26
  • $\begingroup$ Conductivity in metals is hard to measure reliably because the total resistance of the sample is usually much smaller than that of the wiring and the contacts between the test sample and the probes. There is a lot of literature (both in physics and engineering) that describes experimental techniques that removes errors that are due to both wiring and contact resistance (like the four terminal method). You will still need a precise Voltmeter with mV or higher resolution if you want to make precise measurements on bulk metals and you may have to correct for sample geometry. $\endgroup$ Commented Apr 25, 2023 at 12:58

0