PFC Switched-Mode Power Supply - Page 3 - diyAudio
Go Back   Home > Forums > Amplifiers > Power Supplies

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 28th March 2005, 06:14 PM   #21
diyAudio Member
 
N-Channel's Avatar
 
Join Date: Mar 2003
Location: Sol System
Send a message via AIM to N-Channel
Default Flyback -v- Input Rectifiers

The Input Rectifiers are ultrafast types, because of the hi-frequency pulses being drawn through them. Let me clarify this- the diodes are conducting high-frequency pulses, modulated at 120Hz (the rectified haversive frequency here in the States, 100Hz overseas).

The flyback diode need not be an ultrafast type, but I chose the flyback to be ultrafast because the input units already were. Perhaps I will change this in the future
  Reply With Quote
Old 1st April 2005, 10:59 PM   #22
Alme is offline Alme  Ukraine
diyAudio Member
 
Join Date: Jan 2005
Location: in Korea, presently
Steve: I do not doubt that flyback diode switches at PFC frequency
I just wanted to point at diode current waveforms for different operation mode of PFC.
In fixed-frequency PFC like my favorite L4981A this diode conducts current more similar to square form with triangle superimposed; that is at turn-off moment it has to cut quite significant current very quickly (because at this moment FET switch is beginning to conduct, passing the baton of inductor current, so diode must not short it to outpur capacitor!)
On the contrary, critical-conduction PFC draws pure triangle current waveform thru inductor, that is requirement to diode off-state recovery time is not that strict.
You can see different diode production types, called like "fast", "ultrafast", "hyperfast" etc. on IR, OnSemi and the like manufacturers' websites.
So I just meant that MC34262-based circuit can be satisfied with mere "fast" diode (200-250ns recovery) but for example L4981A-based one definitely needs "ultrafast" (35-50ns recovery).
So what do you supply with your PFC?
  Reply With Quote
Old 2nd April 2005, 08:41 PM   #23
diyAudio Member
 
N-Channel's Avatar
 
Join Date: Mar 2003
Location: Sol System
Send a message via AIM to N-Channel
Quote:
Originally posted by Alme
So what do you supply with your PFC?
I'm not sure I understand this question, but here goes:

If you're asking what kind of power do I supply it with, here in the 'States, most American (and Canadian) are wired for 120/240V, 60Hz. We have 240 by virtueof center-tap gorunding, giving two out-of-phase 120V legs. I do have the opportunity to explore inputs at 240V, though I haven't done this, yet.

If you're asking how I am running the control, it's definitely critical conduction mode. I realize that this is NOT fixed-frequency, ans I would like to explore a fixed-frequency type of PFC, as this would greatly facilitate synchronizing the PFC to the PWM, eliminating beat frequencies between the two converter stages.

If I haven't answered your question, please clarify it. Thanks Alme.

Steve
  Reply With Quote
Old 7th April 2005, 02:09 PM   #24
Alme is offline Alme  Ukraine
diyAudio Member
 
Join Date: Jan 2005
Location: in Korea, presently
Well, Steve, my question was about the purpose of your PFC
I suppose it wasn't created just for itself, or did you just try it to learn?

By the way, synchronizing modules is a really great thing. Yesterday during test of my current switching amlifier, I synch'd secondary oscillator (LM393-based) with primary converter SG3525 and immediately got residual noise improvement by 12dB! This is pretty easy done with 20-cent optocoupler.
  Reply With Quote
Old 10th April 2005, 04:51 AM   #25
diyAudio Member
 
N-Channel's Avatar
 
Join Date: Mar 2003
Location: Sol System
Send a message via AIM to N-Channel
Default PCF Purpose

Alme,

OK, I've got it now. The purpose of this PFC was to pass my senior project class for my undergraduate degree, now, going on more than a few years ago!

I had hoped to power a DC-DC PWM Section, either MC33025 or SG3525-based, but with the fuse blowing right away, I decided againse it for now. BTW, how big a fuse is recommended for the input? This is PFC rated for 180W. (The boost coil is made by Coilcraft specifically for operation with the '33262.) I believe I was running with a 4-5 A Slow-blow, and it still blew. Until I get a decent-sized isolation transformer for safety reasons, I am holding off any further high-voltage or AC Mains-powered ckts.

That's pretty sweet on the synchronizing thing! 12dB drop in noise, it's a beautiful thing!

Steve
  Reply With Quote
Old 28th April 2005, 08:30 AM   #26
diyAudio Member
 
John Hope's Avatar
 
Join Date: Feb 2005
Location: nr Oxford UK
Default PFC preconverter

Hi N-channel

Attached is the schematic of the PFC preconverter I use in my PowerDAC2 digital amp since 2003. It's rated at 760W from universal input mains and is really nothing novel i.t.o. PFC technology. The control chip used is the L4981 chip from ST which really is a fine device - it's basic, but robust and does exactly what it says on the tin, so I've not been motivated to move to the newer chips that promote slightly more efficient switching at the expense of higher component count.

The L4981 is able to vary the PFC frequency within a small range in order to 'spread' spectral peaks that could otherwise violate EMC norms. Switching frequency varies downwards from 145kHz to (I seem to remember) about 115kHz. Note that the fs is always below 150kHz so that the fundamental is excluded from EMC measurements, yet is high enough to allow a Magnetics Mollypermalloy toroid core to be used for the boost inductor, which also helps keep emissions down. I can post details of this if there's interest.

It might be noticed that at 500uF the design is a liitle 'thin' on storage cap value; I was able to get away with this because the PFC feeds a 385:72V dc:dc block of 2 series'd 2nd generation Vicor converter 'bricks' that operate down to below 300V. (I had these available and didn't have to sell my car and house to buy them) The predecessor of this PFC had more storage capacitance and was used to feed a 750W ZVS full bridge converter.

The 12V supervisory supply was obtained from a separate 5W universal input switchmode supply that was used to power standby circuitry.

Finally, I feel obliged to say this - many of us know this, but not all do:
WARNING: The circuit attached is not isolated from the mains and should not be built or tested without a suitable isolation transformer, or used to power any other circuitry or device that does not provide isolation. The circuit should not be built by persons not familiar with CE or UL safety regs.


Hope this is of interest.


Cheers


John
Attached Files
File Type: pdf pfcsch.pdf (27.8 KB, 684 views)
__________________
At Spes Infracta
__________________
  Reply With Quote
Old 28th April 2005, 08:38 AM   #27
diyAudio Member
 
John Hope's Avatar
 
Join Date: Feb 2005
Location: nr Oxford UK
Default Mains input filter

I forgot to mention that the ac input for the above PFC is obtained through an external Schaffner RFI filter.
__________________
At Spes Infracta
__________________
  Reply With Quote
Old 28th April 2005, 01:06 PM   #28
Alme is offline Alme  Ukraine
diyAudio Member
 
Join Date: Jan 2005
Location: in Korea, presently
John Hope, thanks for sharing experience! It is of much interest to me; I have tried only fixed frequency type of L4981 (A index) before. For higher power application (about over 500W) they suggest to include small inductor (4-8uH) and snubber/clamper in series with FET drain, for any conduction mode PFC, but recently I couldn't find that application note on ST website.
Can you please recall or estimate power loss/efficiency of your PFC? What is the component with highest losses there - FET, or diode, or maybe inductor? What you think is better to use as main switch at such power level, MOSFET or IGBT? My calculation shows that IGBT is going to have lower static (on-state) losses, but I'm not sure about switching losses at such frequency. Supposing if to use the fastest IGBT type currently available
  Reply With Quote
Old 28th April 2005, 02:06 PM   #29
Eva is offline Eva  Spain
diyAudio Member
 
Eva's Avatar
 
Join Date: Oct 2003
Location: Near the sea
Send a message via MSN to Eva
Hard switching at such high frequencies ~150Khz causes most of the losses to concentrate on the switch. In these circumstances MOSFETs are preferable

I'm using IGBTs but my operation philosophy is different since my circuit uses soft switching at 45Khz with controlled dI/dt. This forces to use a bigger inductor but allows for smaller heatsinks, power devices and EMI filter
  Reply With Quote
Old 28th April 2005, 07:02 PM   #30
diyAudio Member
 
N-Channel's Avatar
 
Join Date: Mar 2003
Location: Sol System
Send a message via AIM to N-Channel
Default PFC Schemo

John,

Wow! That's quite a write-up and a good schematic! I, too, am interested in the power losses at the 115-145kHz region, and would include in the power loss question the sensing resistor. What is its value and power loss? I'm glad you followed up with the external EMI filter comment, because that was to be my first question. Also, what is the lowest input AC voltage you have tested it to? 85VAC? What was the highest? 264VAC? This would really be a robust design, if it could maintain max outout power at lowest line (85VAC) with only moderate losses. What is the value of the coil? The part number on it is kinda confusing.

I am a little confused, though- your schematic has the L4981 run variable-frequency, yet Alme uses it in a fixed-frequency mode. Is this possible with the same chip, or did I miss something in the L4981's application notes?

Great work!

Steve
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
switched-mode power suppies Dimitar13 Everything Else 0 19th February 2008 12:50 AM
Switched mode power supply MMicke Car Audio 3 19th February 2005 10:42 AM
filtering switched mode PSU for SI 5066 amp supply jives11 Parts 0 24th January 2005 09:19 PM
Switched mode power supplies Magura Parts 13 3rd January 2004 05:44 PM
Switched Mode vs Linear Power Supply royalmp2001 Solid State 4 12th July 2002 08:04 PM


New To Site? Need Help?

All times are GMT. The time now is 08:33 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2