Flyback design question

[GonzaloEE]

Hi all,

I've just received this flyback SMPS schematic for a on-field power supply. Please find the link below:

http://files.engineering.com/getfil...897d-4bce-879e-1b2efa7b8b6e&file=Flyback1.gif

I'd appreciate any help finding any design flaws. Creative suggestions or enhancements are of course welcome.

Regards,
GonzaloEE

 

[treez]

Hi

Your converter is not a flyback...yours has primary and secondary conducting together.

It's some sort of forward converter but i think your secondary parallel diode should be on the other side of the series inductor (ie nearer the secondary winding).
I am trying to see how you are de-fluxing the core after switch off.

Could you give your output voltage and Watt rating?

 

[GonzaloEE]

I was told it's 12/24V , 30W. sw. freq. around 60-130Khz.
Sec. windings could be in series or parallel to allow those output voltages. Here they are shown in parallel for 12V mode.

Seems like fine Vout adjustment is done by the TL431 V reference along with R8,R10,R11 and Radj pot. I'm checking that setup in the datasheet.
I'm also missing some free-wheel diode for core de-fluxing. Looks like that snubber network at top of secondary (R6, C6 at top) is the only way back for current when the Schottky goes off. I'm trying to contact the designer.
I'm not a power electronics expert, so I just hired a freelance designer for the job.

Thank you for your guidance!

 

[LionelHutz]

Looks flyback to me. R1, C2, D1 and D2 are the flyback circuit to de-flux the core. Looks like a fairly typical circuit.

D7 appears to be for a secondary battery supply.

 

[logbook]

To tell whether this is a flyback converter or a forward converter it is only necessary to look at the phase relationship between the primary drive and the secondary output. This is determined by the dots on the transformer. In this circuit the primary and secondary are in phase. Therefore it is a forward converter. Whilst Lionel has identified a snubber circuit on the primary, this has nothing to do with the main circuit operation.

As a side comment, this circuit is badly and non-standardly drawn, with ground going across, up and around all over the place. It also has a 4-way interconnect, which was deemed unacceptable over 30 years ago.

 

[itsmoked]

Why go to all the isolation effort then put a 2.2nF across the isolation?

4-way interconnect?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Madcow]

CY4 is for EMI reduction.
I'm guessing that the "Y" is for an
safety agency approved Y-cap.
I've been told it helps cancel the primary
to secondary capacitance of the transformer.

 

[logbook]

Keith,

>4-way interconnect?
The anode end of D5 has a 4 way junction on the circuit. The older drawing convention was to draw little humps to cross over wires. Then they changed it so that you could drive straight across without the humps PROVIDED you never joined 4 wires at point. Joining 4 wires by making a bold blob at that point was considered bad form because you used to get blobs on the old dyeline prints (showing my age here I suppose).

 

[itsmoked]

Ok.

They also often look like swastikas. LOL

Thanks, logbook.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[GonzaloEE]

Thanks to all for your useful help so far.
BTW, do you know any replacements for the TOP246Y switcher, or ferrite Xformer suppliers? Management here is planning a long-term supply contract with our client, so I'm prepared for non-stocked parts through the next months.

Thanks in advance!

 

[GonzaloEE]

Yes, D7, R12 and F2 are intended for batt. charge/backup on power outages. I guess that's OK for a cheap design, though I'd like to replace it for a more efficient approach, like a battery changer IC or something like that.

Any suggestions on this issue are welcome!

 

[LionelHutz]

I looked at it again and it is a 1-transistor forward converter. That snubber circuit is still the reset for the core.

There's probably some type of power management IC initially designed for laptop use that would do the battery charging and backup for you.

 

[GonzaloEE]

Thanks to all for your valuable contributions. This freelance guy was hired for doing the sch/PCB layout, and the design looks different from what I'm used to (digital circuits with buses tying RAM lines together, etc.)

CY4 is a ceramic HV disc cap, same like those from the input EMI filter. As the design is for a non-US/EU client, you'll find there's no PF correction stage.

I don't like much the idea of having the battery warming-up the board through D7/R12 when mains is off. That's why I'm looking for some non-resistive approach.
Thanks Lionel for that lead on laptop batt. management ICs. Any suggestions are welcomed, of course.

Regards,
Gonzalo