Tuned coils

[Dave J.]

Sunrise, I don't dispute that the Nexus product "works", it's one way to build a metal detector. However when it comes to the technical essay on the website, a lot of it is erroneous and should be ignored. I don't have the time to go through it and discuss it point by point, but when I say a lot of it is erroneous, I do know what I'm talking about and I'm doing a public service by encouraging experimenters not to rely on it.

The whole discussion in this thread favoring resonance smacks of vacuum tube era thinking. It ignores related design issues like noise figure, system bandwidth in synchronously demodulated architectures, phase stability, manufacturing reproducibility, and how modern discriminators actually work.

In the metal detector industry, Fisher abandoned resonant receiver coil circuits in most of its products during the mid-1980's, and some other manufacturers began doing the same at about the same time. Resonant receivers have been obsolete for 20 years now. Some major manufacturers began moving away from resonant transmitter coil circuits in the early to mid 1990's, but resonant and nonresonant transmitter circuit topologies are going to coexist for a long time before the resonant topologies finally disappear.

--Dave J.

[Sun_rise_pt]

The resonance that I refered do not have to do with the sender block, but only with the receiver.
In fact the noise increases when you exit the resonance point, ie: Noise is a random disturbance. But if the circuit is tuned to receive certain frequency it is specially prepared to receive it and reject any others. This will undoubtedly be the best S / N.
But ... (there's always a but ...) Unfortunately ... It is necessary to discriminate responses of metals with a synchronous detector that tells us the phase shift...
Then there comes a commitment to consider the sensitivity of the circuit when it is resonant and the reading of the phase that requires to leave a little of this ideal situation ...

This is just my opinion translated by some years of practical (hands-on) experience.

[golfnut]

If you have Tx and Rx coils with same amount of turns and Cap values, this may be best case.


I think this because phase pulling from ground proximity should be the same on both - so may be ok.

may be reasonably easy to try with donor HW like .

IGSL

[jlsilicon]

Is this (and the previous nexus article) to suggest that a particularly wound Coil can only support 1 (and only 1) Best central resonant Frequency ?
- no matter the Voltage supplied or Parallel Capacitor or series Resistor ?

... How about a Coil with multiple Winding-# Connection Leads - for Multiple Frequencies ?

[Dave J.]

Is this (and the previous nexus article) to suggest that a particularly wound Coil can only support 1 (and only 1) Best central resonant Frequency ?
- no matter the Voltage supplied or Parallel Capacitor or series Resistor ?

... How about a Coil with multiple Winding-# Connection Leads - for Multiple Frequencies ?

1. Nope. There are well known examples to illustrate this. The FTP-Fisher "Fratbros" DD searchcoils are used on mainstream products running at 7.7 kHz and also 19 kHz; and in the lab for experimental projects we sometimes run them down to about 3 kHz. There are limitations of course, you'd never use those searchcoils in a 1 kHz or a 100 kHz product.

2. In an induction balance, that'd be a nightmare to actually do and make work. Things like that are commonplace in RF work where induction balance is not part of the picture.

[Mark-VA]

Wow, learned something new again.