[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: SSTC Singing Arc Design - Help needed CLASS D IMPLEMENTATION



Original poster: "Jan Wagner by way of Terry Fritz <twftesla-at-qwest-dot-net>" <jwagner-at-cc.hut.fi>


> >>>>Okay, i think you convinced me to scrap the TL494.
(for PWM and output voltage control purposes it works just fine, but
for audio... nah...)

> Let me see if I have this correct.  Please correct me if I am wrong or if I
> am missing something.
>
> 1.  I should use a 555 or 556 timer IC to generate a sawtooth (which is
> present at the capacitor terminal of 555) at
> the RESONANT FREQUENCY of my secondary coil.  So for me, I need a sawtooth
> waveform at 200kHz.

Precisely! :)

But NE566 or LM566, not a 555.
This IC gives out a square wave and a quite linear sawtooth. The sawtooth
is centered around 5V or so, and has 2.4V peak peak which falls nicely
into typical audio line-out voltage range.
 www.national-dot-com  should have the datasheet

You can look at a schematic at
  http://www.hut.fi/~jwagner/tesla/SSTC/schem/audio-pwm.gif
(please note this is for a stereo audio system - the schematic is just to
give you a general idea)


With the R22 trimmer you can tune the chip. Note that the absolute maximum
voltage between MOD (datasheet says "VC") and Vcc (datasheet V+) pins is
3V, so for a 12V supply it would be 9V to ground. Less and it won't run.

With 12V supply and MOD at 9V, R8 5.6k and C1 1nF it should run somewhere
around 300kHz. Mine runs at ~335kHz.
With MOD at 10..11V, probably your required 200kHz.

The MOD (VC) pin is part of the frequency control.


> 2.  I then feed my audio signal into a comparator with the sawtooth signal
> and I should get a PWM waveform out at the frequency set by my sawtooth
> waveform.

Yup. Audio signal being shifted with R3 (or R6) to the DC level of the
sawtooth. This is around 5 or 6V. It's best you check on a dual channel
scope.


> QUESTION>>>>>>>>>>>  Is there any particular way to bias these signals???
> For example if my sawtooth waveform varies from 0 to 5V, should my audio
> signal be centered within that range or doesn't this matter.

If it were 0V to 5V, then yes, audio centered around 2.5V. This way you
get a 50% duty square wave output at no audio signal. Half of the square
wave is below 2.5V, the other above 2.5V, => 50% duty.


> Also, how
> should one set the amplitude (threshold) for the sawtooth wave (or is this
> just adjusted to get the best audio quality during running)

It doesn't matter much.
And with the LM566 you can't adjust it anyway. ;o)

Main point is that the comparator should have around 5mV or smaller input
threshold. 5mV would give a 400 "steps" resolution for a audio signal of
range -1V to 1V. ( (1V+1V)/5mV=400 ).
And a fast response time of <100ns.


> 3.  I then can use the comparator output (say using a pull-up resistor on
> the output at 5V) to get a PWM pulse from 0 to 5V which can then be fed into
> my MOSFET 4421/4422 drivers.

Yes.

Ah - good you mention the pull-up resistor!!!
LM319 is open-collector. LM311 maybe too. So you need a, say, 4.7k
resistor from the output pin to Vcc, otherwise no output.


If you don't like the output doing "weird" PWM - there will always be one
diagonal of your SSTC fullbridge conducting at any time, which could be a
bit scary - then you can also do simple frequency modulation instead of
pulse width modulation.

This isn't something I've tested yet with LM566, but in theory it works
like this: leave out comparator stuff, and just use a small cap of say
10nF to couple in your audio signal to pot R22 <=> the MOD pin. Then
ignore the sawtooth output, and just use the constant 50% duty square
wave output of the LM566, and let this drive the TC4421/TC4422 pair.

Should be extremely simple - all you need is LM566 and TC4421/22. No
op-amps or comparators.

The audio may have to be damped down to <0.1V peak-peak, so that the
frequency deviation isn't too gross. Additional 10k trimmer, audio in at
top pin, other to GND, center wiper to the 10nF coupling cap, other end
of 10nF to 566 MOD pin, done. :-)


Ok then - good luck!!

cheers,

 - Jan

--
*************************************************
 high voltage at http://www.hut.fi/~jwagner/tesla
 Jan OH2GHR