More than 60% efficiency 5W travel charger

Bob J. skrev:

Well - thats kind of impossible to answer without the actual circuit diagram. Post a picture of this and we might be able to help you

Also what are the conditions. Input voltage - output load?

To increase the efficiency: Use schottky diodes on the output, reduce core losses of transformer, reduce copper losses of the windings, use a better output cap to reduce ESR losses, use better RFI filter with less loss, reduce aux. circuit consumption,...............................bla bla bla

Regards

Klaus

Am I seeing a PNP pass transistor in the output?? What's that doing there when you have a switching supply that can be controlled?

Tim

-- Deep Fryer: a very philosophical monk. Website:

This is a battery charger supply with MCU controlled constant voltage, constant current mode. The most direct route to efficiency improvement is through a modern LDO in CV mode with tight preregulation by the PWM regulator, with low loss current monitoring and limiting operation. See Linear Technology for the applicable class of regulation feature availability.

Why not dump the linear regulator altogether and do as Tim suggests, controlling and using the switcher directly to charge the battery? That saves cost and cuts dissipation too.

Cheers, James Arthur

Because in addition to making the control loops a bit hairy in some cases, it can dramatically increase noise in the battery voltage.

The circuit is an off-line battery charger that uses an MCU together with linear components to execute some kind of multistage battery charger algorithm such as CC->CV->CV,trickle w/cutout at Imin; it very well could be for LiPo. The so-called efficiency probably degrades somewhat during the low output current CV trickle phase due to the regulator circuit overhead, and is not a constant.

The standard off-the-wall cell-phone chargers I have are straight switchers driving right into the battery (either NiMH or LiIon, depending on the charger). The LiIon chargers are constant 4.2V voltage supplies current-limited to about 400mA, typically current-mode flyback switchers. Easy.

True, there'll be some ripple voltage across the battery...if the full 400mA appeared across the 300 millohms of a typical LiIon battery, this would yield as much as 120mV of ripple, but if you really cared you could add an output filter, yes?

Best, James Arthur

But ordinary LiPo charging doesn't require *that* much sophistication: don't charge if the temp is too low or too high, and precondition low-voltage cells @

I don't see a whole lot of off-line switchers that charge Lithium directly like that, the noise must be too much for precision voltage regulation.

That is buck switch.

I've traced a unit or two ... I'll see if I can dig up a schematic to post when I get home after the weekend. They're kind of cute.

Best, James Arthur

Best, James Arthur

No, it is a current limiting circuit, somewhere around 300 to 500 mA.