My need is to build (or purchase, if existent) a nanopulser, but not an ordinary one: it has to have some specific characteristics.

The premise is:

  1. I have a device which acts as a variable capacitor as the environmental conditions change; its working region is in the microFarad's range (approx. 4uF...300uF)
  2. this device has to resonate at different harmonics and, once found the best one (the user decide which one, by observing the device's behaviour in response to the environmental conditions) that resonant point has to be be maintained "fixed", accordingly to the harmonic chosen, following automatically the minimal variations in the device's capacitance - if there where any - so that the resonance takes always place (like a radio tuner, with a PLL or a similar method).

The nanopulser circuit:

  • has to produce DC sharp pulses in the amplitude range of +3V...+200V (user selectable)
  • the single pulse should have a duration no longer than 100ns
  • the frequency has to be variable, in a range between 35KHz...100MHz (user selectable)
  • the pulses will drive a MOSFET, or an IGBT, giving the opportunity to output an amperage of 20A maximum (user selectable) to the LCR system which comprises the device itself coupled with a fixed inductor
  • the frequency selected has to be maintained unchanged even if the load varies in its capacitance, inductance or resistance, through an autotuning system (PLL + VCO or similar method)

The device is a sort of oxygen/hydrogen separator (not HHO) which splits different kind of water mixtures (mainly dirty ones), plus it cleans itself, thanks to the different nanopulses issued: different polluttants mixed in the processed water require different frequencies to be separated and the device's plates require different frequencies for the periodically self-cleaning process.

A similar device can be watched here: https://www.pinterest.it/pin/542331980106296417/

The device is presently operative (in its prototype stage) and it works almost as expected but it is VERY HARD to me to achieve:

  • a stable resonant frequency in the regions over 200KHz (due to the continuously capacitance changes when the different polluttants reach the cell)
  • a so narrow nanopulses (I've achieved 800ns at minimum)
  • an adequate power output with such pulses (I'm unable to surpass 1A total output)

My present circuit is this one: [![NanopulserCircuit][1]][1]

Why nanopulses? After a series of researches and case studies, it has been established that working with resonance allows to achieve:

  • a power consumption reduction
  • a better separation of pollutants from the water
  • a more precise splitting in Oxygen/Hydrogen

The final goal of this research of mine is to create a filter for sanitation which extracts from dirty, black waters (sanitation, drainage, sewer) oxygen - to be released directly in open air - and hydrogen - to be further processed by a dedicated management system.


Is there a circuit (or a series of different, matchable circuits) that perform the function of nanoimpulse generation - with autotune, which is the essential part needed - as per the specifications above?

Between the nanopulser and the LC circuit (my device) there should be a circuit able to "sense" the LC's variations. How can I automatically adjust the pulser's output accordingly with the resonant point desired?

  • 1
    \$\begingroup\$ What the heck is this thing called a nanopulse? Is it some name you have invented? \$\endgroup\$ – Andy aka Aug 17 '18 at 10:12
  • 2
    \$\begingroup\$ So, as I'm really not understanding all that much about the device you want to control with it, can you confirm that you want: A pulse generator, with a pulse repetition rate of up to 100 MHz, but a pulse length of 100 ns? That's not pulses anymore, that's just a constant "on". \$\endgroup\$ – Marcus Müller Aug 17 '18 at 10:55
  • 3
    \$\begingroup\$ @Devesh but you literally can't have a pulse that is 100 ns long (or even longer) repeat with 100 MHz. Draw the signal. It's a constant. So either your 100ns is too long or 100 MHz is too high. Which one is it? \$\endgroup\$ – Marcus Müller Aug 17 '18 at 11:09
  • 3
    \$\begingroup\$ @Devesh no, it's not obvious unless you state something like "the duty cycle has to be limited to 10%, and a pulse never longer than 100ns", and correct me, you're not saying that. \$\endgroup\$ – Marcus Müller Aug 17 '18 at 11:49
  • 2
    \$\begingroup\$ Also, bad news: your MOSFET driver has a rise and fall time of 4ns; that leaves you with a best-case 2ns of "proper pulse amplitude" in a 10ns pulse. Your reality will be far worse. I think your question would really benefit if you added a list to the end that summarizes all constraints for the pulse generator you have, and puts everything into numbers as far as possible! \$\endgroup\$ – Marcus Müller Aug 17 '18 at 12:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.