Prev: 24Hz to 60Hz PLL?
Next: Protecting 3x16 bits with 16 bits ? (crc8 + 8 bit hamming code seems perfect)
From: MooseFET on 13 Feb 2010 11:01
On Feb 9, 5:06 pm, D from BC <myrealaddr...(a)comic.com> wrote:
> My offline smps design has 3 parts making the most heat.
> A power diode, a mosfet and a bridge rectifier.
>
> I'm interested in any cookbook designs that replaces a bridge rectifier.
> Any pointers?


Imagine the no input ripple version of the transformer coupled SEPIC
circuit.
ie: there are 3 windings the third being the isolated output.
Now take the power MOSFET and replace it with a pair in series source
to source.

This circuit can be connected directly to the unrectified maines. It
makes
an isolated squarewavish wave form. This can then be run into a
sychronous
rectifier to make a DC output.

All this involves no diode drops from the mains to the DC output.

From: D from BC on 13 Feb 2010 13:47
In article <aabc57bc-4faf-4067-8ed2-df32f921f214
@b9g2000pri.googlegroups.com>, kensmith(a)rahul.net says...
>
> On Feb 9, 5:06�pm, D from BC <myrealaddr...(a)comic.com> wrote:
> > My offline smps design has 3 parts making the most heat.
> > A power diode, a mosfet and a bridge rectifier.
> >
> > I'm interested in any cookbook designs that replaces a bridge rectifier.
> > Any pointers?
>
>
> Imagine the no input ripple version of the transformer coupled SEPIC
> circuit.
> ie: there are 3 windings the third being the isolated output.
> Now take the power MOSFET and replace it with a pair in series source
> to source.
>
> This circuit can be connected directly to the unrectified maines. It
> makes
> an isolated squarewavish wave form. This can then be run into a
> sychronous
> rectifier to make a DC output.
>
> All this involves no diode drops from the mains to the DC output.


Neato :)


--
D from BC
British Columbia
From: Joerg on 13 Feb 2010 18:51
D from BC wrote:
> In article <aabc57bc-4faf-4067-8ed2-df32f921f214
> @b9g2000pri.googlegroups.com>, kensmith(a)rahul.net says...
>> On Feb 9, 5:06 pm, D from BC <myrealaddr...(a)comic.com> wrote:
>>> My offline smps design has 3 parts making the most heat.
>>> A power diode, a mosfet and a bridge rectifier.
>>>
>>> I'm interested in any cookbook designs that replaces a bridge rectifier.
>>> Any pointers?
>>
>> Imagine the no input ripple version of the transformer coupled SEPIC
>> circuit.
>> ie: there are 3 windings the third being the isolated output.
>> Now take the power MOSFET and replace it with a pair in series source
>> to source.
>>
>> This circuit can be connected directly to the unrectified maines. It
>> makes
>> an isolated squarewavish wave form. This can then be run into a
>> sychronous
>> rectifier to make a DC output.
>>
>> All this involves no diode drops from the mains to the DC output.
>
>
> Neato :)
>

Except then you still have to replace the Schottky diode in the SEPIC
with a synchronously controlled FET circuit ;-)

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.
From: Joe G (Home) on 13 Feb 2010 23:13

"MooseFET" <kensmith(a)rahul.net> wrote in message
news:aabc57bc-4faf-4067-8ed2-df32f921f214(a)b9g2000pri.googlegroups.com...
On Feb 9, 5:06 pm, D from BC <myrealaddr...(a)comic.com> wrote:
> My offline smps design has 3 parts making the most heat.
> A power diode, a mosfet and a bridge rectifier.
>
> I'm interested in any cookbook designs that replaces a bridge rectifier.
> Any pointers?


Imagine the no input ripple version of the transformer coupled SEPIC
circuit.
ie: there are 3 windings the third being the isolated output.
Now take the power MOSFET and replace it with a pair in series source
to source.

This circuit can be connected directly to the unrectified maines. It
makes
an isolated squarewavish wave form. This can then be run into a
sychronous
rectifier to make a DC output.

All this involves no diode drops from the mains to the DC output.

Why not model it in a spice program (eg like the free LTSpice) to prove your
theory. Then build it.

Joe\


From: MooseFET on 14 Feb 2010 10:18
On Feb 13, 3:51 pm, Joerg <inva...(a)invalid.invalid> wrote:
> D from BC wrote:
> > In article <aabc57bc-4faf-4067-8ed2-df32f921f214
> > @b9g2000pri.googlegroups.com>, kensm...(a)rahul.net says...
> >> On Feb 9, 5:06 pm, D from BC <myrealaddr...(a)comic.com> wrote:
> >>> My offline smps design has 3 parts making the most heat.
> >>> A power diode, a mosfet and a bridge rectifier.
>
> >>> I'm interested in any cookbook designs that replaces a bridge rectifier.
> >>> Any pointers?
>
> >> Imagine the no input ripple version of the transformer coupled SEPIC
> >> circuit.
> >> ie: there are 3 windings the third being the isolated output.
> >> Now take the power MOSFET and replace it with a pair in series source
> >> to source.
>
> >> This circuit can be connected directly to the unrectified maines.  It
> >> makes
> >> an isolated squarewavish wave form.  This can then be run into a
> >> sychronous
> >> rectifier to make a DC output.
>
> >> All this involves no diode drops from the mains to the DC output.
>
> > Neato :)
>
> Except then you still have to replace the Schottky diode in the SEPIC
> with a synchronously controlled FET circuit ;-)

Yes, it requires some good timing control on the fets and a pair of
perhaps 600V MOSFETs. This is two less of the costly parts than the
MOSFET bridge idea.

It can be a PFC replacement for the transformer, bridge and capacitor
style power supply. The SEPIC doesn't put all that high of demands on
the core. It only has to hold the energy of one cycle plus a little
to make the current continuous to keep the ripple at zero.
 |  Next  |  Last
Pages: 1 2 3 4 5 6 7
Prev: 24Hz to 60Hz PLL?
Next: Protecting 3x16 bits with 16 bits ? (crc8 + 8 bit hamming code seems perfect)