Baxandall class-D oscillator squegging and the VBIC model

if you

ipolar

e

able

urrent

ctually

iefly

real

he Gummel-

ess.

that

etter -

to do

even

. e

I know. I've done it. It doesn't have to take long to be tedious.

Any particular "dual-winding" inductor? It's not a name I'm familiar with. As far as I'm concerned, any inductor carrying two or more separate but magnetically linked windings is either a transformer or a common mode choke.

-- Bill Sloman, Nijmegen

you

bipolar

current

actually

Gummel-

Things like this:

People call these dual-winding inductors, probably because they are characterized to carry a lot of DC current. Of course they work nicely as transformers.

I used their big brothers in a power supply that we discussed here a while back.

ftp://jjlarkin.lmi.net/ESM_power.pdf

John

g

nst

Fairly early on it was appreciated that complicating the feed inductor by adding an extra LC section to block the second harmonic component reduced the harmonic content of the sinusoidal output without much damage the efficiency. It's mentioned on my web-site

scillator2.htm

According to Tony Williams it was demonstrated in the early 1960s, but tuning the resonant trap to match actual oscillator frequency was too expensive to make it a practical option.

-- Bill Sloman, Nijmegen

ife if you

d bipolar

the

stable

e current

d actually

briefly

In real

t the Gummel-

ddness.

rs that

a better -

le to do

eg even

ged.

aste

h

It's the first time I've heard of it being called a dual-winding inductor. There one in my Peltier thermostat paper

Sloman A.W., Buggs P., Molloy J., and Stewart D. =93A microcontroller- based driver to stabilise the temperature of an optical stage to 1mK in the range 4C to 38C, using a Peltier heat pump and a thermistor sensor=94 Measurement Science and Technology, 7 1653-64 (1996)

I called it a balun on page 1663 and in Figure 8. The - English - supplier called it a common mode choke.

Lots of people do.

T1 and T2?

-- Bill Sloman, Nijmegen

if you

bipolar

stable

current

actually

briefly

real

Gummel-

oddness.

that

better -

to do

even

I used them as transformers so I called them transformers. PADS lets you do that.

The other reason they are called "inductors" is that they have a specified inductance and tolerance. Transformers are seldom specified for inductance, and seldom specified for their DC current capacity.

It really doesn't matter what you call them. What you call them shouldn't constrain what you are willing to use them for. You could use these as baluns, common-mode chokes, inductors, power transformers, signal transformers, flyback transformers, resistors, or RTDs. Nice little parts.

John

"John Larkin" wrote in message news: snipped-for-privacy@4ax.com...

I hi-potted a $1.60 common mode choke at 15kV (peak, AC) between windings. Leakage is big of course, but still usable for a few things.

Planar transformers of course you can use however the hell you want; just make sure you have enough clearance around the routed holes to stand off the voltage you need. Slower to design in, but as cheap, and lower leakage.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

e:

l life if you

and bipolar

for the

o a stable

the current

and actually

ors briefly

). In real

but the Gummel-

r oddness.

eters that

ad a better -

able to do

squeg even

ost

he

uegged.

a waste

p.

ough

er

ne"

les

if

is

a .

Unless you want other people to know what you are talking about.

Obviously.

They are a bit bulky for use as resistors and RTDs. The bulk might be useful if you wanted to dissipate a significant amount of power, but it strikes me as an expensive way of getting it.

-- Bill Sloman, Nijmegen

life if you

bipolar

the

stable

current

actually

briefly

In real

the Gummel-

oddness.

parameters that

better -

able to do

squeg even

squegged.

waste

We follow proper conventions for reference designators. A resistor is R1, not RN1 or RV1 or POT1. A transistor is Q, not T or TR or TRN or any of that nonsense. Since thess parts look like a transformer [1] on the schematic and act like a transformer, I called them T on the schematic. When we use these as inductors, we put the windings in series or in parallel and call them L.

[1] except that they have two cores. That's because the part was created as two inductors, and we plopped them alongside one another to make it look like a transformer.

If you allow yourself to think, more opportunities are available. Add "gain control device" to the list. Maybe even "delay line."

John

rote:

real life if you

eed and bipolar

el for the

n to a stable

es.

hat the current

-up and actually

istors briefly

les). In real

ly, but the Gummel-

ular oddness.

rameters that

o had a better -

be able to do

ly squeg even

almost

s the

squegged.

of a waste

t up.

0 .

enough

ather

plane"

holes

rts if

ers is

ar

or a

9...

ely

- a

With two coils you could use one as a "saturable reactor"

I've never seen it done. and would have thought - as the wikipedia article does - that the approach is close to obsolete.

You can build up delay lines out of all sorts of RLC components. I hadn't realised that common mode chokes had anything special to offer as the inductive components in such a synthesis.

does use two-coil inductors, but controls the coupling between the two coils - his example circuit (figure 7) relies on having 50% coupling, which a common mode choke wouldn't offer.

-- Bill Sloman, Nijmegen

life if you

and bipolar

for the

a stable

the current

and actually

transistors briefly

roles). In real

but the Gummel-

oddness.

parameters that

a better -

able to do

squeg even

almost

squegged.

waste

enough

plane"

holes

if

is

Probably not practical with this series of parts. They typically hit a thermal/current limit before magnetic saturation.

It never hurts to understand what parts can do.

Since it's probably bifalar wound, maybe its windings could be used as a twisted-pair transmission line, a delay line all on their own.

John

in real life if you

t feed and bipolar

model for the

down to a stable

ycles.

s that the current

art-up and actually

ansistors briefly

roles). In real

oddly, but the Gummel-

ticular oddness.

parameters that

who had a better -

ell be able to do

ually squeg even

ld almost

was the

nly squegged.

ng of a waste

e it up.

t 30

lad.

tic enough

I rather

nd plane"

the holes

parts if

ormers is

ne.

s.

.

iliar

e

er or a

D/19...

are

nicely

ler-

mK

re a

s

ed

or

Some are bifilar wound, some not. The length of the winding - and thus the delay - isn't something the manufacturer specifies all that tightly on the data sheet, and might change without notice if they changed core materials.

Probably not the ideal building block.

-- Bill Sloman, Nijmegen