Questions, Single Switch Low - voltage Flyback SMPS ... / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Questions, Single Switch Low - voltage Flyback SMPS ...

1 2 next

Move Thread

LAN_403

Patrick

Fri Feb 13 2015, 07:31AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Ive got to build a small-ish SMPS for my drone as i cant draw power through the balance leads anymore, the loads and voltage drops are too much that way. Power should be 50w max, 12 ,5 and -12 volts from 10 - 14 Vdc input.

Issues i need to address.
1) RFI EMI
2) single switch operation
3) Flux walking to saturation

So, looking and the TI/Unitrode pages and Uzzor2k's work im worrying about saturation. Im trying to remember if reset windings remove power passively or with a diode. Id like to avoid a rest winding if possible or keep it simple to add.

When i wind the transformers i guess ill use copper foil between the primary and secondary, then again around the core externally?

1423812708 2431 FT0 Uc3842 Flyback Converter

Uzzor2k's work.

Can i just keep the core from saturation by leaving the duty cycle below 50% with a tiny load on the secondary continuously?
that pic from here:

ive never used a single switch SMPS that wasnt a iggy coil.

EDIT:

Does this diode/winding put the excess volt/seconds back into the power rails? it seems like it.

Shrad

Fri Feb 13 2015, 08:16AM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

can't you fit a micro ITX power supply which will be fed 12V, like such ones?

Patrick

Fri Feb 13 2015, 08:44AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

will they work down to 10 volts input?

Sulaiman

Fri Feb 13 2015, 10:50AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

Patrick,
an smpsu ic will not start with a supply voltage less than it's 'under voltage lockout' start voltage
for the UC3842 that's 16V. (+/-1V)

I'd go for something like a UCx825 (x = 1,2 or 3, depending on temp range)
uvlo start voltage is just acceptable at 9.2V (+/- 0.4V)
and the operating frequency can be much higher than a UC3842,
allowing much smaller/lighter magnetics.
A push-pull forward converter would probably be your best topology for multiple supply voltage outputs.

Shrad

Fri Feb 13 2015, 11:49AM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

Patrick wrote ...

will they work down to 10 volts input?

that's to be checked with them, but such supplies must surely exist... this is however the closest type of readily built supplies you can find which matches your requirements close enough

the price should be approximately the same if you were to create a prototype yourself, though, so it might be worth trying

Patrick

Fri Feb 13 2015, 04:02PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

well id find one with a lower ULVO, but it was that desaturation winding that worries me. but your suggesting a push-pull, two transistor, two winding transformer. i geuss if im winding a desat and primary, theres no advantage in the transformer in a single switch mode.

Sulaiman

Fri Feb 13 2015, 05:04PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

you will need at least four windings (primary, +12, +5,-12) for flyback
adding a bifilar winding to the primary is not a big difference
AND
using one smps ic for all three outputs will probably be lower power consumption
(22mA not counting gate drive) than three separate converters.
AND
overall weight will be lower,
especially if you use the higher frequency capability of the UC1825
(smaller magnetics, lower value capacitors etc.)
There are probably newer/better smpsu ic for your application,
I'm not up to date with developments.

hen918

Fri Feb 13 2015, 07:40PM

Registered Member #11591 Joined: Wed Mar 20 2013, 08:20PM
Location: UK
Posts: 556

Sulaiman wrote ...

...
adding a bifilar winding to the primary is not a big difference
...

Sorry Sulaiman, I don't like to correct people, however you mean a centre-tapped primary. Bifilar are two windings joined back to back, so the overall inductance is close to 0. (One is wound the opposite way) A bifilar winding would just act like a dead short.

Sulaiman

Fri Feb 13 2015, 10:04PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

no, I meant bifilar.
it is electrically connected as a centre-tapped winding but wound bifilar.
just an easy way to ensure both primaries have the same number of turns etc.

however I do make mistakes sometimes and we should all help each other to avoid mistakes
so when polite (like you) no one should take offence.

Patrick

Sat Feb 14 2015, 08:45AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

ok sulaiman ill change to a push-pull topology. but what out radio frequency? i really got to suppress RF since Im flying...
i wont use electrical isolation on the feed back loop, so i hope that path doesn't leak radio waves.

I may need a 8 volt winding too. but ill try these first.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob