gyrators

[sinclairuser]

has anyone in the metal detecting design world ever seriously considered using gyrators in md design?
i know there are problems with gyrators and they cannot totaly replace transformers and inductors/capacitors but it seems to me they have been relatively ignored in real world applications and i wonder if its not time to look again at the "fifth linear component" after all we have come a long way in production methods since 1948! aly

[deemon]

Never used it in my life As for me , this is exotic but absolutely useless device ... if I need inductance , I take good wire-wound inductance . But in IC design - it has some sense , of course ...

[Dave J.]

Most op amps are gyrators when connected in the unity gain voltage follower configuration. Same with most linear regulators, since they're basically just a special type of op amp. In either case, capacitive loading in certain ranges without adequate series resistance can cause them to oscillate. These are active devices and unless you're designing an oscillator ya gotta make sure the Q doesn't exceed infinity.

--Dave J.

[Dave J.]

single op amp high Q bandpass filter

However, the fact that an op amp has an internal integrator can be put to use to make a single op amp high Q high gain biquad bandpass filter. (The second integrator and bandwidth control resistor are in the feedback network.) Never seen this in print, just figured it out on my own about 20 years ago, built it, and it worked.

It does have peculiarities. First, the input has to be an RC differentiator to the + input of the op amp in order to set DC gain to zero. Second, the op amp has to be of a type with GBWP that's stable with temperature (many are) because GBWP is where the integration constant shows up in a data sheet. Third, the op amp frequency compensation has to be a single dominant pole, no tricky little phase straighteners thrown in there to stretch the GBWP. Forth, unlike other op amp bandpass filters, this one naturally wants to run at high gain.

I believe that this is a comparatively low noise topology (unlike for instance a conventional single op amp multiple feedback type BP filter), but haven't actually done a formal noise analysis on it. The "secret sauce" is that the conventional topology uses the op amp integrator against you so most of the gain of the op amp can't be utilized. Using the Miller stage in the op amp itself as one of the integrators, it's working for you instead of against you and all of the gain of the op amp becomes useful. This topology naturally works at high gain using positive feedback, rather than doing like a conventional single op amp BP filter which keeps gain under control by using lots of negative feedback to cancel most of the incoming signal.

--Dave J.

[sinclairuser]

The "secret sauce" is that the conventional topology uses the op amp integrator against you so most of the gain of the op amp can't be utilized. Using the Miller stage in the op amp itself as one of the integrators, it's working for you instead of against you and all of the gain of the op amp becomes useful. This topology naturally works at high gain using positive feedback, rather than doing like a conventional single op amp BP filter which keeps gain under control by using lots of negative feedback to cancel most of the incoming signal.

--Dave J.

exactly! it seems to me even if you go back to the begginings of metal detectors there has always been something missing with constant reworkings of the same old designs trying to overcome the relative shortcomings of the various designs
the latest approach is digital control or indeed digital designs i personally d'ont like this analogue is more than adequate for our needs as long as its the right design!
my own view is for my requirements a two board design a "hybrid" but finding two boards with similar working frequencies and coil specs ideally so the same coil can be used is currently where im at
which is why i asked the solid state question while trolling the net in my quest i came across a piece by a designer(now sadly no longer with us) where he seemed allmost annoyed that gyrators had been what he called"ignored by moronic designers brainwashed by the dogma of component manufacturers" and this made me wonder aly

[Dave J.]

My "secret sauce" comment wasn't about gyrators, it was about the single op amp biquad bandpass filter.

Thanks for not naming that other "designer", he's quite wrong about the moronic brainwashed etc. stuff. The reason metal detector engineers rarely if ever use gyrators, is because gyrators don't represent an optimum solution to any of the problems we're trying to solve. On the contrary, we fight the problems of circuits that are gyrators when what we really wanted was resistive impedance.

--Dave J.

[ivconic]

Last time i met gyrators was about 25 years ago on some Mesa Boogie amp!
Even than it was "relic" and exotic!

[Tepco]

What you met is not genuine Mesa Boggie amp, it used classic LC circuit in its equalizer stage, in later numerous copies\modifications this stage is replaced with gyrators, to mimic relatively high inductances of original design, with significant loss of performance, increased noise and reduced overload level.

(I think putting here original Mesa schematic and modifications will be heavily OT)

Gyrators are in general good to mimic some exotic value components, not easily achievable by normal means, but some performance loss is normal due to circuit inherent design.

However, “anti-gyrators”, circuits behaving like ones when it is not needed or intended to do, are common problem in analog design.

[ivconic]

It was long time ago, i am not sure if i remember right; but i think it was Mark III or V. Something like that...
There are free schematics on the net. I saw somewhere recently.

[ivconic]

Curiosity; i founded it! It was Mark III, this one:

[ivconic]

Tepco you are right; Mark III is having classic LC circuits (according to schematic)!
But i remember clear that than were gyrators instead!?
And it supposed to be original MB. Not copy!
Is there any chance that Mark III had further revisions than?

[sinclairuser]

thanks guys like i said i only wondered
coming into md's from servicing radios and hacking xbox 360 md's are a whole new challenge got the bug now though i got my scope back of my mate and cleared out all my 360 stuff cleared my bench and when i get back off my hols i'm gonna knock up some of these kits we will see from there! aly
p.s. allways had a mate who sorted my md's out and i did his shortwaves but he died in april so its forced me to fix my own then i found this site now i want to build them from scratch!