Macrohenry: simple audio amplifier using a opamp and a transformer

[tca]

Hi, let me share with you what I found while surfing the web. Don't ask me how or why! I'm getting addicted on building small wattage amplifiers, go tell?!?

Going to breadboard this tomorrow!



A contribution from our amateur radio DIYers friends!

Cheers.

Ref.: http://www.antiqueradios.com/forums/viewtopic.php?f=12&t=60155

[tca]

Found another version of it!



Ref.: http://theradioboard.com/rb/viewtopic.php?t=3393&sid=4da4a76dd203c040c93000a3b2f7912b

[PRR]

Hmm.

[WaveshapeIllusions]

That looks cool. So it uses the chip to put out decent voltage and runs a transformer to get the current high enough to drive a speaker? And the impedance matching of course. Interesting idea...

[tca]

Hmm.

Why do you say that?

[Kesh]

It's not really push pull as both amps operate the full cycle. More just a way to double the voltage available to the transformer. Let us know how it sounds.

[PRR]

> not really push pull

It is.

It doesn't use all the possible advantages, true. As you say, both "devices" work over the whole cycle. However the devices are not simple devices but composite. Internally inside each opamp Q13 swings one way, Q14 the other way, and they idle nearly-off, for Class B functionality. DC in the transformer is nearly cancelled (although this coud be done other ways).

Hum.

You did note the plus and -minus- 9V (18V total) supply?

The Vout computation does not account for transformer ratio.

In the first version the load on each opamp is 125 ohms, well below the optimum. Distortion and power suffer. A single section driving Black-Blue to ground would make a bit more power. Perhaps not important since the 273-1380 is already hurting below 200Hz, and higher power is just uglier bass.

The second plan seems to be missing some bias at pin 5. Replace the top-left 2.2u cap with a short.

The second plan sure does have enough iron! The primary reflects 800 ohms, or 400 per opamp, which is a bit below peak '072 power but somewhere up the curve of improving THD. Perhaps 0.06 Watts.

One 9V into '386 will make four times the power, no iron, less wiring.

[Gurner]

The second plan seems to be missing some bias at pin 5. Replace the top-left 2.2u cap with a short.

Also, to my eyes, the 2.2uf cap from R4 to ground seems a bit surplus for a bipolar supply arrangement (such a cap is normally in the feedback loop of a single opamp supply to decouple the DC from ground)

[PRR]

> cap from R4 to ground seems a bit surplus for a bipolar supply

Without it, the DC gain is also 2Meh/10K or 200. A bad-case '072 can have 5mV input offset. 5mV*200= 1 Volt output offset! Across the low DC resistance of the OT primary, that may suck ALL the current a '072 is capable of, leaving none for signal.

It's rarely that bad, and there may be some cancellation from the other half. A designer could go either way.

You do see such a cap on nearly all bipolar Power Amps, to kill the DC gain so that input stage DC offset is not amplified to the speaker. It's not that uncommon in other bipolar audio opamp work, though it has been out-of-fashion for a while (servos are cheaper and appear to have no caps in the signal path).

As you say, this cap or some alternate scheme is essential in single-supply work.