Difficulties with SEPIC converter - 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 7th August 2008, 05:00 AM   #11
diyAudio Member
 
Join Date: Sep 2007
Hi all,

thanks for your opinions, I really appreciate them! Currently my spare time is limited, so it will take a while before I'm able to look into this matter.

It should be evident by now that I'm merely a beginner with the practical part of switching power supplies, but I did not implement current-mode control for several reasons: First, it was easier this way. Second, I was under the impression that current-mode control was useful mostly because of the current limit capability and faster transient response. I was'nt aware of it's effects on the control loop stability.

I haven't posted any pictures of the converter, but the control IC and most of the resistors and caps related to it are placed on a small separate board which is attached vertically to the main board. Very much in a way that many off-line converters are build. I did it just for the reason that it would be more easy to change the control circuitry if needed. Now the urgent need has risen, so I only need to define a stable control loop, not to design a whole new board.
__________________
-Juha
  Reply With Quote
Old 9th August 2008, 08:11 AM   #12
diyAudio Member
 
Join Date: Sep 2007
Some progress made:

The following files show the Pspice & SwitcherCAD models I used to tackle this compensation obstacle. What I have recently done is:
-I made spice model from the UC3842 error amplifier
-Plotted the magnitude and phase responses of the initial circuit (the one with the stability problems)
-Added RC filter to freeze the magnitude response much earlier in the frequency range
-Plotted the responses once again, noting the possible improved performance
-Used the LTSpice model to see if any improvement happened

It seems that this converter should now work a bit better. I read the article Mr. Claude Abraham kindly told about and that prompted me to try the radical decrease in magnitude response. For now, it worked. Haven't tried fixing the physical circuit yet.

Here are the files:
-The pspice model of the initial feedback loop
-The pspice model of the corrected feedback loop
-The initial frequency response plot
-The corrected frequency response plot
-The initial gate voltage waveform
-The corrected gate voltage waveform
-The LTspice model of the converter

p.s. Don't pay too much attention to the error amp model, it may not be accurate enough when it comes to input/output impedances, I tried to make the frequency responses equal with the ones on the datasheet.

If you look at the initial gate voltage waveform, I can tell you that it indeed look exactly like that when observed with a digital scope! I really hope to get this converter operating soon!
__________________
-Juha
  Reply With Quote
Old 9th August 2008, 10:33 AM   #13
diyAudio Member
 
Join Date: Sep 2007
Here are results from a 5A pulse-loading simulation:

5A pulse load response

The converter is loaded with two 5 A pulses, duration 30 ms each. Load rise/fall times are 10 ns. These load pulses are visible on the pdf, but they are not well visible.

The output voltage follows the suden load chage nicely, perhaps more than I expected. The settling time should be enough for my purposes. But still, the converter doesn't like a similar 10 A load. I would blame the controller, since the controller ouput duty cycle reaches approx. 70 % maximum instead of the near 100 % promised in the datasheet.

Could there be something in the compensation loop that reduces the maximum duty cycle?
__________________
-Juha
  Reply With Quote
Old 22nd August 2008, 06:57 PM   #14
diyAudio Member
 
Join Date: Sep 2007
Hello!

Good news at last: I fixed the compensation network as previously described and now the converter works quite well . As Dr. Ridley stated (or someone else, cannot remember), this converter is very slow. This can be verified with a large, sudden change in load. Practically my converter isn't able to handle sudden loads below ~15 ohms (Vout=14.4 V, Vin=12 V). But if I gradually increase loading, I can go down to 10 ohms - but not further.

I'll post here again once I've conducted more precise measurements with that lovely 4-channel digital scope at work Thank you all for help and support!
__________________
-Juha
  Reply With Quote
Old 11th November 2012, 01:26 PM   #15
djQUAN is offline djQUAN  Philippines
diyAudio Member
 
Join Date: May 2004
Location: manila
Send a message via MSN to djQUAN Send a message via Yahoo to djQUAN
upping a 4 year old thread. Here's my story

I'm building a SEPIC based CC/CV mode switchmode supply for various applications so I designed and had boards made. some of my planned applications are battery chargers and LED drivers. (see attachments for PCB layout and schematic)

The board uses the ancient TL494 (as I already had a bunch of those) for PWM control, IRFZ44 as the main switch. Hi-Flux core for the main inductor, all Panasonic FM for the input, output and sepic caps, MAX4172 for the high side current sense (can eliminate that and use a low side current sense resistor if output ground doesn't have to be the same as input ground)

The problem:
First application I have for it is a 15Ah LiFePO4 battery pack charger which needs about CC at 3A and CV at 14.60V.

I'm having a really hard time stabilizing the circuit. The CV mode is fine, transient response works well with a small overshoot and there's a small lag at no load to full load response.

in CC mode, current is maintained but the circuit goes crazy and makes buzzing noises. I've been working on and off on this thing for about two weeks until I stumbled upon this thread and the mentioned app notes. Ended up having to add a snubber (3K + 1uF) across R32 to slow down the response greatly.

Once it was slowed down, circuit behaved perfectly. It's actually fine since the load does not need a fast response since in battery charging, the voltage and current does not change quickly. One minor problem I have is that if the circuit is powered before the battery is connected, once battery makes contact, there is a huge current spike as it takes a bit before the current limiting takes effect. If the battery is connected before power applied to the circuit, there is no problem as the circuit has soft start implemented on the TL494.

I can post pictures of the actual unit if someone's interested.
Attached Images
File Type: png sepic-ckt.png (33.8 KB, 56 views)
File Type: png sepic-top.png (42.4 KB, 50 views)
File Type: png sepic-bottom.png (31.0 KB, 51 views)
  Reply With Quote
Old 11th November 2012, 01:41 PM   #16
djQUAN is offline djQUAN  Philippines
diyAudio Member
 
Join Date: May 2004
Location: manila
Send a message via MSN to djQUAN Send a message via Yahoo to djQUAN
I forgot to add:

The point when the circuit goes crazy is at moderate loads. It occurs well below the point where the circuit goes from discontinuous mode to continuous mode but I tried another coil with more turns and still occurs at approx same current level.

At first, when I wound the inductor using single #18 wire per winding then switching to 20 strands of #31 wire per winding made things much better but the oscillation in current mode persisted but much much less than the single #18 wire. Turn counts were the same. Switching frequency is around 56kHz. I couldn't explain how that happened. It must have something to do with better current flow due to the litz wire at high freq and high current but I don't know much more.

Then I also noticed that the instability occurs at about >2.7A input current. if I increase input voltage or decrease output current limit threshold to maintain 2.7A input current, the circuit is stable without the RC snubber across R32.

The PCB also has options to be used as a boost converter but I don't have immediate application for that so it is not yet tested for that use.
  Reply With Quote
Old 17th November 2012, 12:13 PM   #17
djQUAN is offline djQUAN  Philippines
diyAudio Member
 
Join Date: May 2004
Location: manila
Send a message via MSN to djQUAN Send a message via Yahoo to djQUAN
The converter works pretty well now so I made an acrylic box for it.

output is set at 14.4V and current limits at 3A

I can use it for charging the 15Ah LiFePO4 battery pack or providing a constant 14.4V supply for loads under 3A with a 10V-18V input.

*reposted because I uploaded the wrong size pictures*
Attached Images
File Type: jpg IMG_1373.jpg (120.8 KB, 43 views)
File Type: jpg IMG_1374.jpg (108.6 KB, 37 views)
File Type: jpg IMG_1376.jpg (120.3 KB, 24 views)
File Type: jpg IMG_1384.jpg (107.8 KB, 22 views)
File Type: jpg IMG_1387.JPG (87.8 KB, 23 views)
  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
Difficulties with DC/DC converter 12V to +/-40V, 8 amps vena2a Power Supplies 16 18th June 2013 01:35 PM
citation 12 hum und bias difficulties to adjust leloga Solid State 2 10th March 2007 11:23 AM
PCM2902 difficulties jmar Digital Source 7 13th January 2006 09:07 PM
Overwhelming difficulties with regulators, please help! DIAR Parts 31 4th November 2004 10:10 AM
I've got difficulties to solve this electronic exercise Bricolo Everything Else 13 5th June 2004 02:31 PM


New To Site? Need Help?

All times are GMT. The time now is 12:16 PM.


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