On Wednesday, 2 September 2015 04:06:13 UTC+10, piglet  wrote:
> On 01/09/2015 13:25, Bill Sloman wrote:
> >
> > The Baxandall class-D oscillator produces sine waves. They aren't perfect sine waves, but a couple of percent of third order harmonic and progressively less  higher order harmonics beat the pants off a Royer inverter.
> >
> Yes, I think it is a tricky topology to get pure sine waves from. When 
> voltage fed you have to ensure a small non-conducting dead time when 
> both transistors are off. When current fed (by inductor) that dead time 
> would be lethal and instead an overlapping conduction time is necessary 
> but that places both transistors as a dead short across the resonant 
> tank. Is that why you are trying a current source instead of inductor?

The Baxandall circuit doesn't bother about non-conducting dead time - it predates MOS-FET swtiches, and doesn't care much about both BJT transistors being slightly on - but not saturated - at cross-over.

Life gets more complicated with MOS-FET swtiches.

http://sophia-elektronica.com/Baxandall_parallel-resonant_Class-D_oscillator5.htm

describes as circuit that's guaranteed to start-up (note the worst-case Vgs difference between the switching transistor models) and not draw too much current when it's working. It's not something I'd build in real life, but it works in LTSpice. The .asc file is on the web-site.

The current mirror delivers a tolerably pure half-sine wave current into the centre tap, which the inductor doesn't. There's nothing more to it than that. 

-- 
Bill Sloman, Sydney

On 01/09/2015 13:25, Bill Sloman wrote:
>
> The Baxandall class-D oscillator produces sine waves. They aren't perfect sine waves, but a couple of percent of third order harmonic and progressively less  higher order harmonics beat the pants off a Royer inverter.
>
Yes, I think it is a tricky topology to get pure sine waves from. When 
voltage fed you have to ensure a small non-conducting dead time when 
both transistors are off. When current fed (by inductor) that dead time 
would be lethal and instead an overlapping conduction time is necessary 
but that places both transistors as a dead short across the resonant 
tank. Is that why you are trying a current source instead of inductor?

piglet

On Tuesday, 1 September 2015 18:20:01 UTC+10, piglet  wrote:
> On 01/09/2015 05:09, Bill Sloman wrote:
> >>> It has been claimed that Jim Williams got the circuit from England, without > >the Baxandall label.
> >>
> >> Got some proof?
> >
> > Someone from the UK e-mailed me the story about giving the Baxandall circuit to a Linear Technology rep to answer a question that Jim Williams had had the reps put around. They weren't prepared to make a big revelation out of the story, which was old - Linear Technology's AN45 was published in 1991 and the Baxandall style cold-cathode driver is figure 36 amongst a bunch of other circuits, so it goes back 25 years, and Jim Williams is dead.
> >
> >>> I do have a low distortion variant of the class-D oscillator, with a current mirror rather than the feed inductor, but that's not the low distortion oscillator I'm working on at the moment.
> 
> The Royer and Jensen circuits go back to 1954 and 1957 respectively and 
> so predate Baxandall. Both were still known in the late 70s textbooks. 
> Are you saying that Jim Williams got the idea from Baxandall but 
> credited Royer?

The difference between Royer and Baxandall is the inductor in series with the centre-tap.

Royer doesn't have it, and produces square waves (usually with spikes at the switching transitions).

The Baxandall class-D oscillator produces sine waves. They aren't perfect sine waves, but a couple of percent of third order harmonic and progressively less  higher order harmonics beat the pants off a Royer inverter.

http://www.butlerwinding.com/electronic-transformer-inverter-trans/

makes the Jensen inverter a variant of the Royer approach. 

Royer's paper, which I did see once, back in 1970, made a lot of fuss about using saturating transformer cores and driving them into saturation, which isn't actually necessary nor a good idea. Letting the drive transistors come out of saturation works just as well.

As far as I can tell, Jim Williams was given a copy of a schematic of a Baxandall inverter, but no reference to Baxandall's 1959 paper, and cited Royer's paper as the closest thing he knew about to prior art. It was a bit silly.

-- 
Bill Sloman, Sydney

On 01/09/2015 05:09, Bill Sloman wrote:
>>> It has been claimed that Jim Williams got the circuit from England, without > >the Baxandall label.
>>
>> Got some proof?
>
> Someone from the UK e-mailed me the story about giving the Baxandall circuit to a Linear Technology rep to answer a question that Jim Williams had had the reps put around. They weren't prepared to make a big revelation out of the story, which was old - Linear Technology's AN45 was published in 1991 and the Baxandall style cold-cathode driver is figure 36 amongst a bunch of other circuits, so it goes back 25 years, and Jim Williams is dead.
>
>>> I do have a low distortion variant of the class-D oscillator, with a current mirror rather than the feed inductor, but that's not the low distortion oscillator I'm working on at the moment.

The Royer and Jensen circuits go back to 1954 and 1957 respectively and 
so predate Baxandall. Both were still known in the late 70s textbooks. 
Are you saying that Jim Williams got the idea from Baxandall but 
credited Royer?

piglet

On Tuesday, 1 September 2015 06:54:06 UTC+10, John Fields  wrote:
> On Sun, 19 Jul 2015 01:26:33 -0700 (PDT), Bill Sloman
> <bill.sloman@gmail.com> wrote: 
> >On Sunday, July 19, 2015 at 1:49:42 AM UTC+2, M Philbrook wrote:
> >> In article <c228b7df-ec64-469a-ae5f-f4ae29cfe028@googlegroups.com>, 
> >> bill.sloman@gmail.com says...
> >> > 
> >> > On Saturday, July 18, 2015 at 5:50:11 PM UTC+2, John Larkin wrote:
> >> > > On Sat, 18 Jul 2015 09:39:34 -0400, bitrex
> >> > > <bitrex@de.lete.earthlink.net> wrote:
> >> > > 
> >> > > >
> >> > > >For a hobby project, I'd like to be able to generate +48 volts or so 
> >> > > >from +5 volt USB.  Current requirements are small, maybe 15 mA or so? 
> >> > > >I'd like uh, low noise (I don't have an exact spec so this is just  
> >> > > >kind of existential at this point), so maybe a charge pump would be  
> >> > > >in order rather than a boost switcher?
> >> > > >
> >> > > >Does anyone make a charge pump IC that I could feed an external clock 
> >> > > >to, with outputs suitable for running something like a Dickson pump? 
> >> > > >Maybe that would be too many stages to go from +5 to +48...
> >> > > 
> >> > > Two of these
> >> > > 
> >> > > http://www.digikey.com/product-detail/en/PDS1-S5-S24-M-TR/102-2973-2-ND/4006941
> >> > > 
> >> > > with the outputs in series would work. They are fairly quiet, and you
> >> > > could add a little filtering to help. At light loads, their output
> >> > > tends to be a few per cent high. But then, 5 volts from USB may not
> >> > > really be 5 volts.
> >> > > 
> >> > > For really quiet, consider some sort of sinewave drive step-up
> >> > > transformer.
> >> > 
> >> > A Baxandall Class-D oscillator - one step-up transformer and one inductor - would work well. MOS-FET transistors do a bit better than bipolar transistors, but Baxandall's paper rather pre-dates them.
> >> > 
> >> > http://sophia-elektronica.com/0344_001_Baxandal.pdf
> >> > 
> >> > http://sophia-elektronica.com/Baxandall_parallel-resonant_Class-D_oscillator1.htm
> >> 
> >> Oh God, You just couldn't keep that failed OSC out of the subject!
> >
> >The 1959 Baxandall class_D oscillator isn't "failed". To quote from my web-site 
> 
>"The circuit is probably best known from
> Jim Williams' series of application notes for Linear Technology, on
> high frequency inverters for driving cold cathode back-lights used in
> laptop computers (application notes AN45, AN49, AN51, AN55, AN61,
> AN65)." These apparently were Linear Technology's most popular
> application notes for quite a while.

> So what's your point?

The point was that what I was putting forward wasn't my low distortion oscillator, but rather Peter Baxandall's Class-D oscillator, which has been a success story in a coupe of applications.

> >It has been claimed that Jim Williams got the circuit from England, without > >the Baxandall label.
> 
> Got some proof?

Someone from the UK e-mailed me the story about giving the Baxandall circuit to a Linear Technology rep to answer a question that Jim Williams had had the reps put around. They weren't prepared to make a big revelation out of the story, which was old - Linear Technology's AN45 was published in 1991 and the Baxandall style cold-cathode driver is figure 36 amongst a bunch of other circuits, so it goes back 25 years, and Jim Williams is dead.
 
> >I do have a low distortion variant of the class-D oscillator, with a current mirror rather than the feed inductor, but that's not the low distortion oscillator I'm working on at the moment.
>
> So why bring it up?

Completeness.
 
> >You are distinctly public-spirited in regularly reminding us that you are an idiot, but you really don't have to do it quite as often as you do.
> 
> Nor do you, but your proclivity toward attaining that goal seems to be
> well founded.

Your famous difficulties with comprehending complex sentences may influence that point of view.

And why post something now? You are reacting to a post I made on the 19th July 2015. You are slow, but six weeks is a long time.

-- 
Bill Sloman, Sydney

On Sun, 19 Jul 2015 01:26:33 -0700 (PDT), Bill Sloman
<bill.sloman@gmail.com> wrote:

>On Sunday, July 19, 2015 at 1:49:42 AM UTC+2, M Philbrook wrote:
>> In article <c228b7df-ec64-469a-ae5f-f4ae29cfe028@googlegroups.com>, 
>> bill.sloman@gmail.com says...
>> > 
>> > On Saturday, July 18, 2015 at 5:50:11 PM UTC+2, John Larkin wrote:
>> > > On Sat, 18 Jul 2015 09:39:34 -0400, bitrex
>> > > <bitrex@de.lete.earthlink.net> wrote:
>> > > 
>> > > >
>> > > >For a hobby project, I'd like to be able to generate +48 volts or so 
>> > > >from +5 volt USB.  Current requirements are small, maybe 15 mA or so? 
>> > > >I'd like uh, low noise (I don't have an exact spec so this is just kind 
>> > > >of existential at this point), so maybe a charge pump would be in order 
>> > > >rather than a boost switcher?
>> > > >
>> > > >Does anyone make a charge pump IC that I could feed an external clock 
>> > > >to, with outputs suitable for running something like a Dickson pump? 
>> > > >Maybe that would be too many stages to go from +5 to +48...
>> > > 
>> > > Two of these
>> > > 
>> > > http://www.digikey.com/product-detail/en/PDS1-S5-S24-M-TR/102-2973-2-ND/4006941
>> > > 
>> > > with the outputs in series would work. They are fairly quiet, and you
>> > > could add a little filtering to help. At light loads, their output
>> > > tends to be a few per cent high. But then, 5 volts from USB may not
>> > > really be 5 volts.
>> > > 
>> > > For really quiet, consider some sort of sinewave drive step-up
>> > > transformer.
>> > 
>> > A Baxandall Class-D oscillator - one step-up transformer and one inductor - would work well. MOS-FET transistors do a bit better than bipolar transistors, but Baxandall's paper rather pre-dates them.
>> > 
>> > http://sophia-elektronica.com/0344_001_Baxandal.pdf
>> > 
>> > http://sophia-elektronica.com/Baxandall_parallel-resonant_Class-D_oscillator1.htm
>> 
>> Oh God, You just couldn't keep that failed OSC out of the subject!
>
>The 1959 Baxandall class_D oscillator isn't "failed". To quote from my web-site 

---
As if it wasn't self-serving: "The circuit is probably best known from
Jim Williams' series of application notes for Linear Technology, on
high frequency inverters for driving cold cathode back-lights used in
laptop computers (application notes AN45, AN49, AN51, AN55, AN61,
AN65)." These apparently were Linear Technology's most popular
application notes for quite a while.

---
So what's your point?
---

>It has been claimed that Jim Williams got the circuit from England, without the Baxandall label.

---
Got some proof?
---

>I do have a low distortion variant of the class-D oscillator, with a current mirror rather than the feed inductor, but that's not the low distortion oscillator I'm working on at the moment.

---
So why bring it up?
---

>You are distinctly public-spirited in regularly reminding us that you are an idiot, but you really don't have to do it quite as often as you do.

---
Nor do you, but your proclivity toward attaining that goal seems to be
well founded.

John Fields
 

On 7/24/2015 9:26 AM, Phil Hobbs wrote:
> On 7/24/2015 9:18 AM, Klaus Kragelund wrote:
>> On Friday, July 24, 2015 at 2:04:50 AM UTC+2, Phil Hobbs wrote:
>>> On 7/23/2015 7:22 PM, Klaus Kragelund wrote:
>>>>
>>>> It's well past midnight for me, sorry excuse, but a cap
>>>> multiplier inside a DC loop in a buck or boost , what the heck is
>>>> that?
>>>>
>>>> Cheers
>>>>
>>>> Klaus
>>>>
>>>
>>> Capacitance multipliers are amazingly good at cleaning up SMPSes,
>>> but they're a bit squishy on the DC regulation.
>>>
>>> If you close the DC loop of an LM2594 (or something like that)
>>> around the emitter of the cap multiplier, and the AC loop around
>>> from the switcher's output, you get the best of all worlds: decent
>>> efficiency and very very low noise.
>>>
>>> It's pretty simple: normal buck + cap multiplier, FB pin driven by
>>> a resistor from the cap multiplier output (to keep the regulation
>>> reasonable) and a capacitor from the FB pin to the SMPS output (to
>>> keep the loop stable).
>>>
>>>
>> Ok, but for the efficiency, you loose some power in the capacitance
>> multiplier. Like those made for class A amplifiers, right?
>>
> Don't know about class A amplifiers, at least if you mean audio.  A cap
> multiplier is just an emitter follower hung on an N-pole RC lowpass.
>
> For what I do, the tradeoff is generally between losing a bit of
> efficiency on a switcher, or having to use a shunt-regulated cap
> multiplier as a linear regulator, because switcher noise is intolerable.
>
> As I often repeat, "there's SMPS quiet and then there's instrument quiet."
>
> Cheers

Oh, and if you're desperate for that last half a volt, you can use a 
low-sat NPN and source a little current into the RC from the input 
supply.  That'll reduce the dropout voltage.  You do have to keep enough 
headroom for V_CEsat plus the peak-to-peak ripple.

Cheers

Phil Hobbs

-- 
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

On 7/24/2015 9:18 AM, Klaus Kragelund wrote:
> On Friday, July 24, 2015 at 2:04:50 AM UTC+2, Phil Hobbs wrote:
>> On 7/23/2015 7:22 PM, Klaus Kragelund wrote:
>>>
>>> It's well past midnight for me, sorry excuse, but a cap
>>> multiplier inside a DC loop in a buck or boost , what the heck is
>>> that?
>>>
>>> Cheers
>>>
>>> Klaus
>>>
>>
>> Capacitance multipliers are amazingly good at cleaning up SMPSes,
>> but they're a bit squishy on the DC regulation.
>>
>> If you close the DC loop of an LM2594 (or something like that)
>> around the emitter of the cap multiplier, and the AC loop around
>> from the switcher's output, you get the best of all worlds: decent
>> efficiency and very very low noise.
>>
>> It's pretty simple: normal buck + cap multiplier, FB pin driven by
>> a resistor from the cap multiplier output (to keep the regulation
>> reasonable) and a capacitor from the FB pin to the SMPS output (to
>> keep the loop stable).
>>
>>
> Ok, but for the efficiency, you loose some power in the capacitance
> multiplier. Like those made for class A amplifiers, right?
>
Don't know about class A amplifiers, at least if you mean audio.  A cap 
multiplier is just an emitter follower hung on an N-pole RC lowpass.

For what I do, the tradeoff is generally between losing a bit of 
efficiency on a switcher, or having to use a shunt-regulated cap 
multiplier as a linear regulator, because switcher noise is intolerable.

As I often repeat, "there's SMPS quiet and then there's instrument quiet."

Cheers

Phil Hobbs


-- 
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

On Friday, July 24, 2015 at 2:04:50 AM UTC+2, Phil Hobbs wrote:
> On 7/23/2015 7:22 PM, Klaus Kragelund wrote:
> >
> > It's well past midnight for me, sorry excuse, but a cap multiplier
> > inside a DC loop in a buck or boost , what the heck is that?
> >
> > Cheers
> >
> > Klaus
> >
> 
> Capacitance multipliers are amazingly good at cleaning up SMPSes, but 
> they're a bit squishy on the DC regulation.
> 
> If you close the DC loop of an LM2594 (or something like that) around 
> the emitter of the cap multiplier, and the AC loop around from the 
> switcher's output, you get the best of all worlds: decent efficiency and 
> very very low noise.
> 
> It's pretty simple: normal buck + cap multiplier, FB pin driven by a 
> resistor from the cap multiplier output (to keep the regulation 
> reasonable) and a capacitor from the FB pin to the SMPS output (to keep 
> the loop stable).
> 
> 
Ok, but for the efficiency, you loose some power in the capacitance multiplier. Like those made for class A amplifiers, right?

Cheers

Klaus

On 7/23/2015 7:22 PM, Klaus Kragelund wrote:
>
> It's well past midnight for me, sorry excuse, but a cap multiplier
> inside a DC loop in a buck or boost , what the heck is that?
>
> Cheers
>
> Klaus
>

Capacitance multipliers are amazingly good at cleaning up SMPSes, but 
they're a bit squishy on the DC regulation.

If you close the DC loop of an LM2594 (or something like that) around 
the emitter of the cap multiplier, and the AC loop around from the 
switcher's output, you get the best of all worlds: decent efficiency and 
very very low noise.

It's pretty simple: normal buck + cap multiplier, FB pin driven by a 
resistor from the cap multiplier output (to keep the regulation 
reasonable) and a capacitor from the FB pin to the SMPS output (to keep 
the loop stable).


Cheers

Phil Hobbs

-- 
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net