I'd like to have such an AC square wave oscillator putting AC through a load - which can go to 60 VDC and deliver no more than 50 mA with sharp, fast rise times and very "square". I would hope a topology exists.

The biggest request/demand I have is a circuit which would use the least amount of components, especially ICs if there would be any ICs then please a topology with ICs that have the least amount of pins. In other words, I am looking for an oscillator which can be thrown together onto a breadboard rather quickly. I need an actual circuit, not an output from a function generator.

Frequency: 3.9 Hz Duty: 50% Load: Resistive, no smaller than 1 kΩ

Squareness, sharp and fast rise times are just desires. I'm sure people won't suggest an oscillator which is slow enough to turn a square into a triangle wave.

  • 1
    \$\begingroup\$ Frequency, duty cycle, and quantification of "sharp, fast" will be needed. Also some idea of the input characteristic of the load. \$\endgroup\$ – awjlogan May 19 '18 at 7:30
  • 1
    \$\begingroup\$ Just added specifics you mentioned. \$\endgroup\$ – user33915 May 19 '18 at 7:33
  • \$\begingroup\$ In that case, a 555 timer driving a MOSFET should do the trick. \$\endgroup\$ – awjlogan May 19 '18 at 7:35
  • \$\begingroup\$ Thats not going to make AC into load. \$\endgroup\$ – user33915 May 19 '18 at 7:36
  • 1
    \$\begingroup\$ You can build one with a relay and a capacitor. Doesn't get much simpler than that. \$\endgroup\$ – Brian Drummond May 19 '18 at 9:16

The simplest is a blocking oscillator, using one transformer, a transistor, and a power source (battery?) with a couple of passive components. The frequency isn't easy to adjust, the transformer has to be customized for the power input and desired output voltage, but it IS simple. typical blocking oscillator If your output doesn't share a negative or positive rail with the input power source, a three-winding transformer is used.

While it is simple (in the sense of component count) the design is subtle and depends on magnetic nonlinearity in the transformer. It isn't flexible, or inexpensive, in a one-off (because of the customized transformer). If you want symmetric square waves, a two-transistor variant is a possible improvement.

A high-voltage-capable op amp with three resistors and capacitor is a more common solution, but 50 mA and 60V are higher power than most such amplifiers offer.

  • \$\begingroup\$ This is probably a rather bad answer. The thing is, it's really hard to get a transformer which works at 3.9 Hz. \$\endgroup\$ – WhatRoughBeast Jan 23 '19 at 21:45

Suggest using a Colpitt's or Hartley oscillator fed into a CMOS inverter.

  • \$\begingroup\$ Can you explain how the OP will get 60 V AC (that's +60 V to -60 V) at 50 mA from a CMOS inverter? \$\endgroup\$ – Transistor May 20 '18 at 15:42
  • \$\begingroup\$ I don't think the 50ma constraint was on the supply. OP? \$\endgroup\$ – Toby Sanchez May 21 '18 at 3:45

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.