Class D amp using LM319 comparator

Thats It! time for the real world test! The picture below shows breadboard with some parts added. LM319 on the left versus Tl494 on the right for some wave form comparison.

Breadboard layout for testing on the left LM319 right TL494

Lm319 vs tl494

The blue trace is Tl494 while the yellow trace is Lm319

There is no speed difference between the two chips in terms of T rise and T fall therefore lm319 is as reliable as Tl494 in PWM audio generator. But when I try to increase the carrier signal from 235khz to 440khz Lm319 can still deliver respectable waveforms in fact I crank it up to 955khz and still produced good but slightly rounded corner pulses.

Real world probing of voltages across C2 in yellow trace, and across C4 in blue.

Real world comparison of voltages across C2 in yellow and C4 in bluelinear voltage across C4 appearing on pin 9 of U1

The image above was taken by probing the voltage across C4. Even a well trained eye may thought this is generated by a costly precision high speed comparator. It looks very linear to me.

In this breadboard scenario, output pin 12 of U2 produce ~11.2 volts peak at 51% duty factor (235khz) as seen from my digital scope. Then filtered by 25k resistor R10 and 10uf capacitor C5 to smooth it out into vicinity of ~5.6VDC. Another 25k resistor R5 were added so that the audio frequency and carrier frequency will not interact to each other. This is due to capacitor C5 that has a capacitive reactance of equal to short circuit to these two signals.

Input to output of U1

The blue trace is the input at pin 9 of U1 then the yellow trace is the output at pin 7.

Leave a Reply

Your email address will not be published.