I sorely need help driving two MOSFETs

I'm not surprised if you've had confusing advice, perhaps some people have not understood the circuit.

Neither do I I'm afraid. It can't turn on / off coupling unless you physically could move the windings (perhaps with a motor).

By 'dump to ground' do you mean place a load across the transformer? (If not, please clarify - 'dump to ground' is not an

electrical engineering term I've ever heard of).

If you mean to place a load, then since your output is rectified (I notice you have a bridge rectifier there) you only need a single

MOSFET to place a load (i.e. a very small resistance) across the rectified output. It wouldn't require the dual MOSFET

arrangement (which then brings up more questions about why the dual op-amps and dual MOSFETs.

Hi shabaz, thanks for your reply and you're right, I could be clearer in what I mean.

Essentially this is a circuit known as a Q-switch. Following a large RF signal on the resonant circuit the system experience what is known as "ringing", where despite not being driven the coil still emits RF as there is energy stored in the inductance/capacitance. By "dump to ground" I was just trying to emphasise that the MOSFET side of the circuit doesn't do anything except give the secondary wingings a route to ground. When there is a path to ground then the induced voltage can flow to ground and energy is taken out of the oscillating system.

Does that make more sense?

Hi Nick,

I can't find any information on Q-switching related to resonant circuits. I can find information related to lasers.

Do you have a link to a book/journal explaining it? (Not an internet page - there is a lot of junk out there - I want

something written by an engineer and peer reviewed for a reputable journal).

Anyway, if you want to perform switching to a load at any point then it doesn't need two MOSFETs as mentioned earlier - just

use a single one (N-ch) and you can apply your 3.3V logic signal to the gate through a small resistor.

That's a first approximation, anything more may need hiring an engineer to help you with specifics and run measurements for

you, etc.

I can't find any information on Q-switching related to resonant circuits

If I recall my days in RF, it could be referring to this, but I don't recall it being referred to as Q switching.  (However my recall is sometimes less than 10%  )

Amplifier Classes C to H

The transmitter/amplifier relies on the Q of the resonant circuit to provide the rest of the waveform.

mark

Hi Mark,

That's a good reference to the amplifier classes, very interesting read.

I too don't remember anything RF related being known as Q-switching - but my memory level is low too!

Also the use of op-amps and the reference to 2.5MHz and 'ttl' levels of 3.3V makes it hard to figure out the actual requirements and intent unfortunately : (

Best to ignore the op amp idea - as I said, it wasn't a good one.

If you're willing to read a short paper here is one that I am basing elements of my work on [Dropbox download]

"Suppression of Transient Processes in the Oscillatory Circuit of the NQR Spectrometer" - Rudakov (2000)

The reason there are two MOSFETs is so that both halves of the rectified RF have a route to ground.

Hi Mark, thanks for your response but this isn't the application I was thinking of. Q-switching is a term that gets used in NMR / MRI circles referring to changing the quality factor of a resonant circuit (RF coil). For more detail please see the paper I've linked in a reply to shabaz!

Hi Nick,

I think the figure 1 shows basically what I outlined earlier (i.e. just a single N-ch MOSFET is sufficient).

Be aware that figure 1 is just a pseudo-circuit just intended as an outline to the reader, it is missing bits.

Basically the source end of that MOSFET could be connected to 0V and it may approximately work, but

as mentioned probably needs some more intensive work by an engineer to design your circuit - it won't be

as simplistic as the figure 1 diagram.

Hi shabaz

Thanks again for taking time to look at the paper. I've actually built the circuit in the paper and it doesn't work as well as my two MOSFET setup. But back to the matter in hand, and even if I was only using one N-channel MOSFET; what sort of driver should I be using?

Nick

Often you need to drive a lot of current since the gate looks like a capacitance. So, perhaps consider a gate driver IC (there are off-the-shelf ones, from manufacturers like On Semi, Infineon and Linear Tech most likely). There will be some good literature on them from most of the MOSFET manufacturers out there. It is hard to offer any specific recommendations, without examining the circuit and performing some measurements most likely, and you'll need to examine your MOSFET datasheets.