universal smps ....without pfc fron end

[phoenix87]

hi,
I would like to make a universal smps 500W without using a pfc ......
Wat are my options ....
1. Can I use a half bridge and control the duty cycle to keep the o/p voltage constant for a input voltage of 90 to 270.
2. Use a voltage doubler and keep a switch to switch between 90-110 range and 180 - 270V
3 Or use a flyback smps and use feed back to keep o/p voltage constant....

regards
phoenix

[eem2am]

Over that range of vin, even with the mains voltage doubler......you are going to have problems.

How are you going to minimise the mains frequency ripple in those capacitors....and keep the Primary DC Bus from sagging down too much?

-on top of mains frequency ripple you also have switching frequency ripple in those caps.

How are you going to minimise the mains harmonic currents when you have no pfc.............?

even if pfc is not a requirement for your application, mains harmonics will have certain limitations....and your smps is likely to violate them.


I am afraid to tell you that your wide vin range, (even with the doubler ) is likely to make it impractical.

Also, theres no much point doing a half bridge after a voltage doubler link..................because the half-bridge just divides the vin by two.....the half bridge primary only sees vin/2

[phoenix87]

@ee2am so wat am i supposed to do ?????????
without pfc smps cant become unviversal?????????

[eem2am]

.........you are quoting 90-265VAC....

that is very wide range.

and you are quoting 500W.
..you have not said whether that is peak power or average power so i had to assume that you want 500W continuously.

...so in that case.

..absolutely no way can you avoid using PFC.......because if you do not use PFC...then your mains input current harmonics will be far too high....and the 50Hz ripple in your input electrolytic capcitors will be far too much.....

you can try and increase the amount of electrolytic capacitance in order to have enough ripple rating, but the more capacitance you put after the diode bridge ...the more rms ripple you get...............so its like a dog chasing its tail.

im afrain PFC, even if not mandatory for your application...will be essential in order to avoid you having too much ripple current in the electrolytics.

-you have probably seen 1000W smps's by Hypex and Coldamp and abletronik, and Abletek, and Connexelectric with no pfc.....but they are for applications with very low average power and they have special current monitoring and limitation circuitry on them to assure that it never stays at 1000w for long.

so, is it 500w continuously?

[gmarsh]

It's possible. There's been 500W and much larger, non-PFC, universal input supplies built for decades.

Magnetics, transistors, capacitors etc. have to be sized for the full operating range, and the control loop has to be designed to be stable over the entire input voltage range of the supply. It's harder, but it's not impossible by any stretch.

[eem2am]

sorry gmarsh, but i must beg to differ with you.

At 500W of continuous power, you cannot manage without a PFC.....because you will incur too much ripple current in the primary side electrolytic capacitors.

....you can try and solve this by putting ridiculously large numbers of paralleled electrolytic capacitors after the primary rectifier bridge, but then your rms ripple current in the capacitors simply increases...

...and also your mains input harmonic currents get ridiculously large and you will not pass the regulatory bodies approval standards on harmonic currents.

I have tried and failed to do 500W without PFC from the mains.......its the ripple current in the primary electrolytics that show-stops the whole deal.

......and as i mentioned the mains harmonic currents.

If such SMPS's exist , then they are illegal, even by lax audio-style approval bodies, since the mains harmonic currents will be too high.

All smps's , even 30W ones, have restrictions on mains harmonic current levels...even if they dont need to be fully PFC.

If you just do the calcultion yourself , or run a few simulations, you can see that this is the case.

...the only exception is , say, a 500W smps, that on average, over any 15 second interval, draws no more than about 50W.

..if your talking 500W continuously, then no deal.

.....and lets face it, if say Ozzy Osbourne plays the guitar, your amp is going to be on max power continuously.

[sawreyrw]

Originally Posted by eem2am

sorry gmarsh, but i must beg to differ with you.
.

eem2am,

gmarsh said it wasn't real simply, but it could be done. I agree with him. Also, if you do some research on line harmonic limits, I don't think they apply for equipment thet consumes less than 1 KW.

Rick

[eem2am]

http://www.ospmag.com/files/pdf/whit...-and-Input.pdf

The above page 6 tells of the limiting values of mains harmonic currents allowed.......

...unfortunately it does apply below 1KW.........

-I know about this because i designed an 330W , Full bridge SMPS (90-265VAC) without PFC that was above the mains harmonic current limits and i was therefore sacked with immediate effect.

I violated the mains harmonic current limits as you see in the guide.

It was to supply an audio amplifier.

[sawreyrw]

OK, I agree with you. It would be possible to build a 1000 watt supply that meets the harmonic limits without a PFC (by using a huge filter), but that is totally impractical, and not nearly as cost effective as PFC.

[gmarsh]

This one's 180 to 264 (not 90 to 264) but it's 1500W and doesn't use active PFC:

http://www.meanwell.com/search/se-1500/default.htm

That being said, at higher power levels using an active PFC front end does make the DC/DC section of the supply a lot easier to design.