1. Home
  2. Electronics Design
  3. Single output half bridge going discontinious at light loads OK?

Single output half bridge going discontinious at light loads OK?

I have a offline (300VDCin) halfbridge design that supplies 24V at 240W max. The PSU goes discontinious at ~ 18W. There is one operational mode when the load drops to 5W and the supply would go discontinuous. Typically I've avoided this but there is no room for a larger output inductor. (isn't that always the case) Controller is a UC3846 current mode with current slope conpensation.

My question is, what changes, if any, would be required to make the supply stable while the output inductor in discontinuous? The load current will be farly stable at 5W then will step up to 24W when a load (mostly resistive) turns on. Then the current draw will vary relatively slowly between 24W and

240W.

Regulation requirement are loose at +/- 10% so if I get some loss of regulation performance while discontinious that is OK.

Reply to
mook johnson
Loading thread data ...

As long as the drive and housekeeping supply circuits maintain integrity, discontinuous operation of a single output supply is benign, as auxiliary outputs are not present to experience excessive droop.

By the way - half bridge and peak cycle-by-cycle current mode control don't tend to co-habit easily. It has to be degenerated to voltage mode to maintain stability at duty cycles approaching the limit. Current mode feed components can be maintained in the control loop to retain ~most of the hf and transient benefits of the control mode, but it will look more like average current mode control under those conditions.

If you need to avoid discontinuity for unstated reasons, you might convert the output to sync rectification, if the output is not intended for parallel apps. Self-driven sync rectification in a current doubler is fairly easy, with some simple precautions over wider input voltage ranges.

RL

Reply to
legg

What if it were a full bridge?

Reply to
mook johnson

What is the question?

There will be no difference re discontinuity effects - the secondary waveforms and filter characteristics are identical for both versions, sync rec or not.

Chaotic tendencies at increased duty are the ~same for peak current control if the primary is capacitively coupled. That seems to be the feature that produces 'theoretical' incompatibility, and forces the need for an over-riding voltage mode influence at the pwm comparator input.

I'm still not sure what your problem is. 'Instability' at light loads for a single output converter is not an issue if the drive and housekeeping circuitry is robust and the output maintains regulation without abnormal ripple. You can often induce whining or hissing in any conventional unpotted converter by straying within minimum loading boundaries.

Some minimal loading or a switched load has been used in the past to avoid hijinks in the zero-load condition. Some regulation circuits that are secondary-powered may naturally load that branch under light external loads, just as a natural matter of course in their basic function.

Zero is often a very hard thing to regulate, particularly if there is no natural option to reverse polarity, in servo-ing to the set-point.

RL

Reply to
legg

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.