2

In DC squid operation, the SQUID is usually flux-locked into a position on the flux-voltage curve, and the change in supercurrent required to maintain that lock point is used to determine changes in magnetic flux through the loop. Because the flux-voltage curve is periodic with period equal to the magnetic flux quantum, however, it does not seem to me to be possible to measure absolutely the flux coupled into the SQUID loop - rather, it seems only possible to measure changes in flux from some initial value.

Is there a way to use a DC SQUID to measure magnetic flux absolutely, or are only differential measurements possible?

EDIT in response to one of the comments below by @FraSchelle, a slight amendment to the question: Is it ever possible, using a SQUID with a known geometry, to say that the magnetic field $B=0$? Or is it only possible to say that is has changed by $\Delta B$ from the initial value when you first turned the SQUID on?

KBriggs
  • 439
  • You can count the periods. In practice it would probably be a good idea to have two devices that are being kept 90 degrees out of phase to make sure that one doesn't skip any of the periods. – CuriousOne Feb 29 '16 at 20:53
  • How do you know that you started counting at n=0, though? They all look the same. – KBriggs Feb 29 '16 at 21:46
  • I would suggest interference between devices of different size and rotation in the field until a position with zero flux is found. – CuriousOne Feb 29 '16 at 21:56
  • Well, the flux is defined that way, isn't it ? From the definition $\Phi=\iint B\cdot dS=\oint A\cdot dl$, $\Phi$ flux, $S$-surface, $B$-magnetic induction, $A$-vector potential, $l$-path-element. You may see a gauge transform as a change of the reference. I know it's puzzling because you believe a gauge transform does not affect the integrals hence the flux $\Phi$, but in fact it's the same as the problem of absolute voltage: you can only measure a voltage drop because $E=-\nabla V$ up to a gauge transform. What I meant by that is that, if $B=\nabla\times A$, then $B$ – FraSchelle Mar 01 '16 at 12:48
  • and $B^\prime = B - \nabla\times A$ would give you the same flux, but the flux $\Phi^\prime = =\iint B^{\prime}\cdot dS$ must be seen as a reference (being zero here). I've tried to motivate people about this problem a few years ago, without success: http://physics.stackexchange.com/q/62282/16689 I feel important to realise there is no possible measurement of fields, only work, flux and so on (integrals of them) and hence only relative measure can be done. – FraSchelle Mar 01 '16 at 12:53
  • @FraSchelle Fair enough. I guess I should change the question slightly then. Given a choice of reference point (in your analogy to electric field, given a choice of ground) is it possible, using a SQUID, to ever say for certain that you are at that reference level? – KBriggs Mar 01 '16 at 14:59

0 Answers0