Power Supply Industry News Knowledge is Power ™

Original Message
Topic SEPIC or ZETA DC-DC synchronous controller IC 

I am in need of an extremely low-ripple DC-DC converter that can output 12V at 1A with a 9V to 24V input. I would like to use the 'Zeta' topology in synchronous form. See http://www.linear.com/pc/downloadDocument.do?id=4416 for a non-synchronous version.

I use the LTC3780 buck-boost four-switch regulator, but discovered that it alternates between buck and boost in the transition region where Vin is close to Vout, causing unacceptable low-frequency ripple. I need a converter that stays locked to an external clock, without spurious frequency content.

I decided the Zeta is preferable, although very similar, to the SEPIC topology because the output is through the inductor, rather than a diode or MOSFET, reducing ripple. I simulated such a design using the LTC3845 with a two-stage output filter. This is a buck regulator synchronous controller, but the MOSFETs work similarly, at least at first glance. At first it looked very promising and efficient. Then I realized the high side MOSFET switching the primary side of the 1:1 transformer (inductor with two windings on a common toroid) would experience -12V at the source lead during the off period, a condition that the LTC3845 gate driver is not rated for.

Does anybody know how I can resolve this issue of switching a MOSFET with an inductor in the source that transitions from +24V to -12V? I think I need some sort of isolated gate driver that is referenced to the MOSFET source lead?
Date 7/30/2007 12:43:17 PM 
Author Lifespeed 
Pageviews 2738 


Response topic SEPIC or ZETA DC-DC synchronous controller IC  
Dear Lifespeed.
Try to use one BUCK CONTROLLER not a BUCK regulattor. There the high side MOSFET might be the topology which accept N-type or, alternativelly, other which accept P-Type. Using the N-type as a example: the source may go negative relate to gate the VGS parameter: in the IRFL4315 this parameter is limited to +/-30V, in spite of this mosfet may stand up 150V VDS. Converselly, if you use a P-type controller, where the Mosfet' drain is tied to the output inductor, the negative voltage will be blocked by the VDS of the p-type mosfet, which you may select a very high voltage part.
Think about it.
Author J.C.S. 
Date 3/22/2016 9:58:56 AM 
Response topic SEPIC or ZETA DC-DC synchronous controller IC  
Hello Lifespeed,

I have done a SEPIC design in the past.
9-30V input, 12V 3A output, efficiency at 9V 83% and 87% at 15V
ripple and noise below 50mVpp without additional filters which
seem to me a reasonable values.
PCB layout has been a very critical concern.
If you are interrested in sharing further information,
contact me.

Author htuerk 
Date 8/2/2007 2:40:39 PM 


Powered by:
SMPS Power Supplies 

Power Supply Industry News, online since 2002.
Copyright © SMPS Power Supplies™. All rights reserved.