Talk:Kelvin bridge

Inconsistency
The labeling of the diagram is not consistent with the discussion. Leonard G. (talk) 20:09, 11 May 2008 (UTC)

Where is Rx? -- &#x2611; Sam uelWantman 03:00, 22 December 2008 (UTC)

The diagram in the article doesn't match cited one
The diagram on http://www.allaboutcircuits.com/vol_1/chpt_8/10.html implies a true four-terminal resistance measurement (Kelvin method) but the diagram in this article seems to show a (less accurate) three-terminal method. The diagram at allaboutcircuits isn't very illuminating either. The right-hand bridge appears to be floating with no voltage source applied at the top and bottom vertices. All very confusing... Woz2 (talk) 15:25, 4 June 2012 (UTC)


 * I don't understand your issues.
 * Both articles show the KB trying to eliminate the wire resistances. The all about circuit's ("AAC") explanation does a better job, but its diagram is in error by showing tapped connections on Ra and Rx rather than connections at the top and bottom of both resistors. All about circuit indicates an Rwire, but does not explicitly draw it in the diagram. Similarly, the WP article just calls it out as R.
 * Both articles use 4 terminal sensing. Low impedance drive uses one path; the high Z measurement is taken along another path. One might even say they are using 6 terminal sensing due to the resistive divider bridging R (or Rwire in AAC) adding two more sampling wires.
 * Everything is floating -- neither article shows a ground. There's an implicit null meter in the WP article measuring Uwy (which is labeled U instead of V because schematic is borrowed from another WP); there's an explicit null meter in the AAC's article (it is drawn as a circle, but it is a meter and not a voltage source).
 * I don't understand the problem with the diagram not matching that in the cited reference. The diagram need only be consistent within the article.
 * Glrx (talk) 03:40, 8 June 2012 (UTC)

The information in either or both article may or may not be correct, and both contain useful clues, but neither were very helpful to someone like me (who claims intelligence and motivation, but who is seeing the concept for first time) come to an understanding. The article contains one or more red herrings. For example:
 * "Resistance R should be as low as possible (much lower than the measured value) and for that reason is usually made as a short thick rod of solid copper." (Emphasis mine). As I presently understand the KB, that sentence is bogus. The resistance R is unwanted parasitic contact resistance that the bridge tries to engineer around. It is not a component engineered into the bridge as the article seems to imply.
 * The diagram in the article doesn't explicitly show the four-terminal nature of the measurement like the AAC diagram does. It lacks the parasitic resistances that touch the voltage source.

The floaty nature of the AAC diagram was an artifact of the particular line routing and I now agree the topology is correct at AAC, but the way it is drawn confused me. I like the tapped connections on the AAC diagram because they helped me visualize the parasitics that the KB tries to compensate for. Woz2 (talk) 11:44, 8 June 2012 (UTC)


 * The resistance R is designed to be as low as possible. The designers make it as low as possible to minimize the last term.
 * The diagram in the article doesn't show the resistance in the high level drive, but it does show a four-wire path.
 * I don't like the AAC diagram because they add confusion. There's an issue about who "owns" the distributed resistance in the connection, but a tapped resistor doesn't point that out well. Compare large connection dots used in article's illustration to delinate measured resistance. See also image at Shunt (electrical) where contact resistance is minimized. One could argue that the Kelvin terminals are taps on the resistor, but the zigs are a large part of the schematic symbol.
 * AAC does the better job explaining rationale of the bridge.
 * Glrx (talk) 17:54, 10 June 2012 (UTC)
 * To me the diagram is confusing because I don't know what the dotted lines mean, why R is darkened, or where the power source is. The explanation and diagram for the Carey Foster bridge, by contrast, is crystal clear. — Preceding unsigned comment added by 84.227.254.143 (talk • contribs) 18:54, 30 March 2014

Disputed content
I came to this article from somewhere else. I know my bridges including this one. The diagram at the top of the article is not the correct form of Kelvin bridge as used for measurements (indeed, I don't believe that Lord Kelvin himself would recognise it). It seems to be some form of hybrid between Kelvin's first bridge (which was flawed) and the final form (It's not even strictly that because the original version only had two resistors in the actual bridge part itself). The diagram as shown does not show four terminal connections to either the resitor under test (Rx) or the standard resistor (R2) which is the fundamental objective of the bridge. Further the balance equation in the article states that the last part of the equation becomes negligible if the ratio of R3 to R4 is the same as R'3 to R'4 - and for the bridge as illustrated, that is true (but only if the parasitic resistance R is very small).

In the balance equation for the proper Kelvin bridge, the last part contains a term where the ratio of R3 to R4 is subtracted from R'3 to R'4. If those ratios are equal, the whole term becomes zero (not negligible) leaving the balance equation for a regular wheatstone bridge. The part about making R as low as possible is significant for the bridge as illustrated. But for a true Kelvin bridge provided the ratio of R3 to R4 is the same as R'3 to R'4, R is in the part of the balance equation that becomes zero and thus makes no contribution to the measurement (though it is desireable to keep it low so that the current through the test and standard resistors is not too adversly affected, but even if it were, the accuracy of the measurement is not affected provided the detector can resolve the null adequately). DieSwartzPunkt (talk) 17:27, 4 April 2014 (UTC)


 * I got so fed looking at it, I decided to do something about it. I have been WP:BOLD and rewritten the article.  The only thing I am not happy with is the circuit diagram of the bridge, but this is the best I can get it.  DieSwartzPunkt (talk) 14:06, 5 April 2014 (UTC)

Who is it named after?
Which Kelvin / Thompson should it be attributed to? BP OMowe (talk) 13:36, 28 February 2024 (UTC)