Forums

smps with a fixed duty cycle?

Started by Unknown March 15, 2016
The classic linear power supply (transformer, bridge rectifier, cap) relies on AC at a fixed frequency of 60 Hz (50 Hz outside the USA?), so that's basically a 50% duty cycle, right?

And if higher frequency means less iron is required for the transformer...

Would a simple SMPS, operating at several hundred kHz, with a fixed 50% duty cycle still outperform the linear power supply?

Thanks,

Michael

On Tuesday, March 15, 2016 at 2:37:32 PM UTC-7, mrda...@gmail.com wrote:
> The classic linear power supply (transformer, bridge rectifier, cap) relies on AC at a fixed frequency of 60 Hz (50 Hz outside the USA?), so that's basically a 50% duty cycle, right? > > And if higher frequency means less iron is required for the transformer... > > Would a simple SMPS, operating at several hundred kHz, with a fixed 50% duty cycle still outperform the linear power supply? > > Thanks, > > Michael
... or would it just be a miniature space heater? :) Michael
On Tue, 15 Mar 2016 14:36:35 -0700 (PDT), mrdarrett@gmail.com wrote:

>The classic linear power supply (transformer, bridge rectifier, cap) relies on AC at a fixed frequency of 60 Hz (50 Hz outside the USA?), so that's basically a 50% duty cycle, right? >
I don't know what duty cycle means in that context. Most such supplies have rectifier conduction spikes that are fairly narrow.
>And if higher frequency means less iron is required for the transformer...
Yes. Airplane power systems are 400 Hz to save weight.
> >Would a simple SMPS, operating at several hundred kHz, with a fixed 50% duty cycle still outperform the linear power supply? > >Thanks, > >Michael
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
<mrdarrett@gmail.com> wrote in message 
news:49d4383e-22ee-4806-ba80-5e22ecc8de2b@googlegroups.com...
> The classic linear power supply (transformer, bridge rectifier, cap) > relies on AC at a fixed frequency of 60 Hz (50 Hz outside the USA?), so > that's basically a 50% duty cycle, right? > > And if higher frequency means less iron is required for the transformer... > > Would a simple SMPS, operating at several hundred kHz, with a fixed 50% > duty cycle still outperform the linear power supply? > >
You are trying to compair two totally different things. The linear supply could care less about the AC frequency in one sense. As long as the transformer is rated for the frequency and the capacitors are sized for the frequency , the pulsating DC has nothing to do with the regulation. For the linears the higher the frequency within limits the less iron can be used and the smaller the filter capacitors.
On Tuesday, March 15, 2016 at 3:20:00 PM UTC-7, John Larkin wrote:
> On Tue, 15 Mar 2016 14:36:35 -0700 (PDT), mrdarrett@gmail.com wrote: > > >The classic linear power supply (transformer, bridge rectifier, cap) relies on AC at a fixed frequency of 60 Hz (50 Hz outside the USA?), so that's basically a 50% duty cycle, right? > > > > I don't know what duty cycle means in that context. Most such supplies > have rectifier conduction spikes that are fairly narrow.
The point I was trying to make is, no feedback was used to make an attempt to modify the primary waveform shape in any way for the 50/60Hz transformers. If someone were to design an SMPS with no feedback loop whatsoever (which normally changes the duty cycle), running at a constant 50% duty cycle, what would be the ramifications? Thanks, Michael
mrda...@gmail.com wrote:

> > > The point I was trying to make is, no feedback was used to make an > attempt to modify the primary waveform shape in any way for the > 50/60Hz transformers. >
** Some linear PSUs have a triac in the primary circuit that allows the DC output voltage to be regulated.
> > If someone were to design an SMPS with no feedback loop whatsoever > (which normally changes the duty cycle), running at a constant 50% > duty cycle, what would be the ramifications? >
** Such PSUs are in fact common, where regulation is not essential and the aim is to reduce the weight and cost of having a large iron transformer. Generally known as a "square wave inverter" they can be found in many high powered audio amplifiers. The output DC still contains ripple at twice the supply frequency but requires little filtering as a rectified square wave is pure DC. The switching is mostly done with IGBTs these days and can have outputs up to 10kW. .... Phil
On Tuesday, March 15, 2016 at 2:37:32 PM UTC-7, mrda...@gmail.com wrote:
> The classic linear power supply (transformer, bridge rectifier, cap) relies on AC at a fixed frequency of 60 Hz (50 Hz outside the USA?), so that's basically a 50% duty cycle, right?
There's no 50% duty cycle. The positive half-wave and negative half-wave both contribute. There's a 'conduction angle' that can be calculated, and there are series-choke options that will change the conduction angle, if that matters to you. But that doesn't just depend on power supply design, it also depends on load.
On 2016-03-15, mrdarrett@gmail.com <mrdarrett@gmail.com> wrote:
> On Tuesday, March 15, 2016 at 3:20:00 PM UTC-7, John Larkin wrote: >> On Tue, 15 Mar 2016 14:36:35 -0700 (PDT), mrdarrett@gmail.com wrote: >> >> >The classic linear power supply (transformer, bridge rectifier, cap) relies on AC at a fixed frequency of 60 Hz (50 Hz outside the USA?), so that's basically a 50% duty cycle, right? >> > >> >> I don't know what duty cycle means in that context. Most such supplies >> have rectifier conduction spikes that are fairly narrow. > > > The point I was trying to make is, no feedback was used to make an attempt to modify the primary waveform shape in any way for the 50/60Hz transformers. > > If someone were to design an SMPS with no feedback loop whatsoever > (which normally changes the duty cycle), running at a constant 50% > duty cycle, what would be the ramifications? >
Depends on the type. if it's a forward converter it will behave senibly. -- \_(&atilde;&fnof;&bdquo;)_
On Tue, 15 Mar 2016 15:37:53 -0700 (PDT), mrdarrett@gmail.com wrote:

>On Tuesday, March 15, 2016 at 3:20:00 PM UTC-7, John Larkin wrote: >> On Tue, 15 Mar 2016 14:36:35 -0700 (PDT), mrdarrett@gmail.com wrote: >> >> >The classic linear power supply (transformer, bridge rectifier, cap) relies on AC at a fixed frequency of 60 Hz (50 Hz outside the USA?), so that's basically a 50% duty cycle, right? >> > >> >> I don't know what duty cycle means in that context. Most such supplies >> have rectifier conduction spikes that are fairly narrow. > > >The point I was trying to make is, no feedback was used to make an attempt to modify the primary waveform shape in any way for the 50/60Hz transformers. > >If someone were to design an SMPS with no feedback loop whatsoever (which normally changes the duty cycle), running at a constant 50% duty cycle, what would be the ramifications? >
Cheap potted SIP DC/DC converters are usually unregulated forward converters running about 50%. With a fixed DC input, the output is usually stable enough to run opamps and logic and stuff. https://dl.dropboxusercontent.com/u/53724080/Circuits/Power/Cui_DCDC/VASD1_sch.JPG https://dl.dropboxusercontent.com/u/53724080/Circuits/Power/Cui_DCDC/VASD1_top.JPG One of my products makes isolated 24 volts into +-10, with more regulators downstream. https://dl.dropboxusercontent.com/u/53724080/Circuits/ESM/ESM_power_B.pdf https://dl.dropboxusercontent.com/u/53724080/Circuits/ESM/ESM_rev_B.jpg -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com