0-15V 0-120A adjustable PSU attempt with average current control. - Page 2 - 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 4th August 2005, 12:55 PM   #11
tekman is offline tekman  Germany
diyAudio Member
 
Join Date: Oct 2004
Speeding up the compensation as the only measure you take might be more problematic. The complete system wil tend to more instability after a certain point.

Since it is a wide range adjustable SMPS (0-15V), I think a feedback with just one timeconstant will not fit for all the needs. If you use a nested feedback (one fast, one slow time constant), it might be worth to see if things get better during a load change.



hth,
Andreas
  Reply With Quote
Old 4th August 2005, 09:57 PM   #12
Danko is offline Danko  Hungary
diyAudio Member
 
Join Date: Apr 2004
Location: Hungary
Eva, where do you get the synchron-signal, to which mosfet turn ON or OFF?
__________________
Best regards, Danko
  Reply With Quote
Old 5th August 2005, 03:17 AM   #13
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
Quote:
Originally posted by tekman
Speeding up the compensation as the only measure you take might be more problematic. The complete system wil tend to more instability after a certain point.

Since it is a wide range adjustable SMPS (0-15V), I think a feedback with just one timeconstant will not fit for all the needs. If you use a nested feedback (one fast, one slow time constant), it might be worth to see if things get better during a load change.

hth,
Andreas
This is an average-current controlled system. It has a current amplifier that forces inductor average current to track the output of the voltage amplifier up to 10Khz. The optimum gain of this current amplifier is inversely proportional to the output voltage, but adjusting it for 16V output does not mean that dynamics are going to be worse for lower output voltages, it just means that dynamics could be better if maximum output voltage was lower.

Anyway, I'm playing with the voltage amplifier and its compensation is not output-voltage dependent, it only depends on output capacitors. It should have a zero at the resonant frequency of the output filter (approx. 300Hz for that design) and a pole above (Fsw/2*pi) in order to filter switching ripple. The magnitude of the dip during load transients depends on the gain it is allowed to have betweem its zero and its pole. Too much gain causes oscillation between 5Khz and 10Khz due to a lack of phase margin. Too low gain causes a huge dip in output voltage when load is increased.

Today I found out that my previous LM358 voltage op-amp was not up to the job of having more than 30dB of gain at 10Khz with little phase shift, so I tried a better one and it allowed me to double the gain to 36dB without any trace of instability. As a bonus, the dip in output voltage during a full load transient was reduced to approx 100..150mV. I didn't use a better op-amp from the beggining because I always like to get the best possible performance from low-cost components, I hate exotic stuff.

Also, the base drive circuit is somewhat flawed since it provides a negative base current for turn-off that is proportional to the duty cycle. In other words, the circuit won't probaly operate reliably with 120A output at low output voltages, so I'm going to add some degree of foldback curreng limiting below 8V output. Anyway, this PSU is intended for automotive electronics applications so it will be adjusted for 12V or 14.4V most of the time.





Quote:
Originally posted by Danko
Eva, where do you get the synchron-signal, to which mosfet turn ON or OFF?

Both push-pull banks are turned on by default, except when estimated inductor current is low (body diodes of MOSFETs are used as plain diodes for low currents). Each positive-going signal from the SG3525 turns off the right bank of MOSFETS just 500ns before the primary full-bridge is switched in that directon. Then, that bank is turned on again when Vds has fallen almost to 0. Independent zero-detection logic for each transformer, together with inhibit logic and gate-drive buffering, is included in the synchronous rectification daugher-board.
  Reply With Quote
Old 1st September 2005, 09:13 PM   #14
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
Oops!!

Click the image to open in full size.

This was caused by a loose connection in one of the test loads...
That multi-stranded wire just fused it's no joke, the wire actually broke due to overheating.
  Reply With Quote
Old 2nd September 2005, 10:40 PM   #15
rinox is offline rinox  Italy
diyAudio Member
 
Join Date: Oct 2004
Really impressive...... do you want to go in the iperspace or you want to create an hole in the temporal space continuos... please advert us before you turn on your time machine
  Reply With Quote
Old 6th September 2005, 09:57 AM   #16
e96mlo is offline e96mlo  Sweden
diyAudio Member
 
Join Date: Apr 2002
Location: Helsingborg, Sweden
Are you really using bipolar transistors as switches in the PSU? Or have I missed something?

Why aren't you using MOSFETs - especially for the rectification?
  Reply With Quote
Old 6th September 2005, 11:45 AM   #17
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
I am using bipolar transistors for the full bridge switching stage and MOSFETs for the synchronous rectification stage.

I decided to use bipolars because there is no TO-220 MOSFET in the market rated at 400V Vds and capable of switching in excess of 10A with only approx. 1V voltage drop. On the other hand, modern switching bipolar transistors like MJE13009 yield that performance while providing reasonable switching losses.

Also note that each TO-247 MOSFET featuring 0.1 ohm Rds-on costs almost the same as a dozen of bipolar transistors and tnat kind of devices are very hard to drive because such a low Rds-on requires a huge die size with hugue capacitances, particularly the harmful reverse transfer capacitance.

The base drive circuit has gone a bit complex, but I'm happy because I get in excess of 1800W output with less than 40W loses in the bipolar full bridge. I'm trying to reach the 2KW mark in a reliable way.

Anyway, the circuit works fine at high powers, the problems I'm facing now are related to the complexity of average current control when only the primary side current waveform is available and it has a big switching peak at the beggining of each cycle (causing troble at very low duty cycles).
  Reply With Quote
Old 6th September 2005, 02:14 PM   #18
Danko is offline Danko  Hungary
diyAudio Member
 
Join Date: Apr 2004
Location: Hungary
eva, how do you menasure such high output currents? with a current sense transformer?
__________________
Best regards, Danko
  Reply With Quote
Old 6th September 2005, 02:29 PM   #19
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
I use a pair of current transformers, one placed in each side of the full bridge, to get an isolated sample of the exact current waveform flowing in the primary side. I use two transformers because that allows to sense asymmetric currents. The main transformer is wound with a 12:1 turn ratio so 10A in the primary side must produce 120A output.

I also tried current transformers in the secondary side but I didn't like how it worked because transformer saturation was not sensed and compensated at all by the current loop. Also, it was harder to implement due to the hugue turn ratio required and the gauge of the secondary side magnet wires.

I also considered a hall effect sensor placed in the gap of the output inductor, but the cheap ones are too slow and the fast ones are so expensive that they may cost as much as the sum of the rest of the components of the PSU. Furhtermore, hall effect sensors in that arrangement would lose their measurement capability as the output inductor becomes saturated, so I discarded them idea.
  Reply With Quote
Old 7th September 2005, 09:35 AM   #20
mflorin is offline mflorin  Romania
diyAudio Member
 
Join Date: Jun 2005
Location: IASI
Hi Eva

Could you tell us something about your magnetic snubber circuit (L3, L4, D5, D6) ? How dose it works ? Can you recommend a good reference for magnetic circuits used to achieve soft switching?

Thx.
  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
UC3846 current mode control vs. SG3525 with average current mode control mflorin Power Supplies 35 7th June 2014 09:06 AM
Adjustable B+ and Cathode Current linestage Brit01 Tubes / Valves 2 30th July 2009 01:12 PM
forward regulation of push-pull average current mode control? zilog Power Supplies 9 19th January 2009 07:06 AM
will this average current mode controller work? zilog Power Supplies 0 23rd January 2007 10:17 PM
Can some one give me a schematic on a current source, adjustable around 3 A. Progg70 Solid State 42 17th November 2005 11:18 PM


New To Site? Need Help?

All times are GMT. The time now is 11:09 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