Hi everyone.

The data sheet of MC34063 doesn't mention anything about voltage/current mode.
The operation seems voltage based, with current limiting feature. But it does not use the usual ramp to compare freedback singal against. Just a simple comparator.
http://www.ti.com/lit/ds/symlink/mc34063a.pdf

Is it a voltage mode controller anyway? Do all voltage mode loop compensation apply to these type of controllers?

Thank you!

 

hi,
The way I read the d/s its a controlled Voltage output with a maximum short circuit current limiter.
E

 

Is it a voltage mode controller anyway?

Yes.
Look in section 9.2 of the data sheet.
It likely can also be used in the current-mode if you take the feedback from a grounded shunt resistor in series with the load.

Do all voltage mode loop compensation apply to these type of controllers?

No.
It is a hysteretic type converter and those don't need loop compensation for stability.

 

I'm guessing the oscillator generates a sawtooth which ramps up until the limit current is reached, then drops back down. Voltage feedback to the comparator turns provides PWM control.

 

I'm guessing the oscillator generates a sawtooth which ramps up until the limit current is reached, then drops back down.

That's true for standard converters, but a hysteretic converter does not generate an internal sawtooth.
The MC3x063A circuitry is as shown in its data sheet with the oscillator generating a pulse to turn on the output latch through its input gating.
This causes the output voltage to start rising, and the comparator then turns the output off when the desired output voltage is reached.
The voltage will now drop due to the output load, and the comparator then changes state to allow the output to be turned back on by the next oscillator pulse.
It's the output ripple voltage that causes the comparator to change states.

This brings up one possible disadvantage of a hysteretic converter which is, since it requires an output ripple voltage to operate, its output ripple may be larger than a standard converter that uses the ramp to generate the PWM duty-cycle.

But an advantage of the converter is that no loop compensation is required so it's unconditionally stable, and without the delay from that compensation, it responds more rapidly to line/load changes (within one clock period).

 

It is a hysteretic type converter and those don't need loop compensation for stability.

Strange they don't state it clearly in the datasheet.
So, a good old TL494 is also a hysteretic converter?
http://www.ti.com/lit/ds/symlink/tl494.pdf

An application note for the MC3x063A shows a flyback converter which uses a simple type 1 controller in the feedback loop. Obviously not enough for a regular voltage mode controller. But it definitely is a compensation.
"[...]consideration must be given to the proper selection of the feedback loop elements in order to insure circuit stability."
https://www.onsemi.com/pub/Collateral/AN920-D.PDF
page 38

 

I stand corrected on the oscillator operation and its control. The App Note in the post #6 link sets it out nicely.

 

So, a good old TL494 is also a hysteretic converter?

The oscillator generates a ramp, so it is not.

An application note for the MC3x063A shows a flyback converter which uses a simple type 1 controller in the feedback loop. Obviously not enough for a regular voltage mode controller. But it definitely is a compensation.

A flyback converter introduces additional points of instability, so apparently requires some compensation that a simple buck-converter does not.

 

It's the output ripple voltage that causes the comparator to change states.

Thank you very much for clarification.
I had seen some hysteretic controller block diagrams, but they shown no oscillator whatsoever. Just the switch turned on when feedback voltage falls below certain level, then turned off after a fixed time. So, duty cycle is fixed, switching frequency changes.
The IC we're talking about doesn't seem to follow that scheme. Maybe the oscillator is just there to turn the switch off and do the current limiting? The datasheet doesn't explain it.

I stand corrected on the oscillator operation and its control. The App Note in the post #6 link sets it out nicely.

Sorry, I'm not sure if I understand. What do you mean?

 

What do you mean?

I mean the oscillator isn't a ramp type; its a set-reset latch as set out in the App Note.

 

I mean the oscillator isn't a ramp type; its a set-reset latch as set out in the App Note.

Sorry, I didn't get it at first. You're absolutely right.

crutschow, forget about my last post. The datasheet doesn't describe the operation, but the app note does. When I found it, I just read the compensation, calculations and practical circuit sections, and skipped the introduction. I'd like to read (and understand) twice the speed I'm really able to Wouldn't you?