Self oscillaing Buck LED driver has too many components

[treez]

Self oscillating Buck LED driver has too many components

Hello,

I have here a simulation (LTspice) of a self oscillating Buck LED driver. It just regulates the LED current by switching ON/OFF its FET when the inductor current hits the requisite peak and trough levels. One comparator acts on the peak of the inductor current, the other, on the trough. (there is a resistive inductor current sensor)

The glue logic that I have is a bit messy, (basically it just allows one comparator to disable the other comparator when it trips, to allow proper operation).

Do you know how I can use say NAND latchs in order to tidy up this circuit?

Ltspice simulation and schematic attached..

 

[BradtheRad]

I have thought my hysteresis-driven op amp buck converter ought to be a million-dollar idea. Theoretically, that is. Because so far it only works in theory.

 

[mtwieg]

If you just want to build a functional simulation, I would try to replace as much as possible with behavior sources and voltage controlled switches. You should be able to get rid of basically all the comparators and gate drivers that way.

 

[treez]

yes I see what you mean, though say with the comparator which trips on the peak of the inductor current....
it must turn the fet off when the inductor sense goes above the peak threshold, but must not act on the trip
when it is falling back down through it. This is a little tricky.

 

[mtwieg]

Voltage controlled switches in LTspice are neat because they have hysteresis voltage built in, so you can use those to implement a comparator with hysteresis. Or just replace your power FET with one.

 

[emontllo]

Note that there are thousands of hysteric step-down converter control IC's for LEDs. Anyway if you want to design it by yourself and want to simplify the circuit I would suggest using an operational amplifier wired as a Schmitt trigger comparator. The input of the comparator should come from the high side current sense (here again, you can design a differential amplifier circuit with good common mode rejection or use a high side current sensor. Regarding MOSFET driver, you could even use the opamp as a gate driver if it has sufficient peak output current (say 500 mA for the MOSFET you have chosen).

A final comment: which is your LED set current? If you chose ~ 1 A, recirculation Schottky diode will be undersized. Note that with 20 V input, MOSFET duty cycle will be low.
2Ω current sense resistor losses will also be high.

 

[treez]

thanks, LED current is to be 100mA

 

[BradtheRad]

Here is a hysteresis-driven buck converter. A simple op amp (or comparator) controls a transistor switch.

The sense voltage is tapped at a strategic place. It senses partly load voltage, partly inductor current.

The schematic is untested. If you make a million dollars with it, please remember where you saw the idea.

 

[D.A.(Tony)Stewart]

Hysteretic controlled current sources can have sub harmonic instability if you ignore the phase margin.

Read this to understand why.