I'm working on my first opamp preamplifier design, and I'm wondering if there's a real advantage to using a +/-15v power supply instead of something lower like +/-9v. Most of the designs I've come across seem to use the +/-15v "standard", but it seems to me that any reasonable power amp will be clipping long before those rails are reached. If a power amp only needs a couple of volts to provide full power, what is the advantage to using such high rails?
In Small Signal Audio Design, for instance, Doug Self advocates using a +/-17v supply, because it's the highest that most opamps will tolerate. I suppose that might be because he favors putting the volume control in the last stage, in order to avoid re-amplifying a previously attenuated signal. In that case, it makes sense to want maximum headroom in prior stages because any boost is applied to a source signal of unknown level. Would that still be a concern, though, if the volume control were placed up front, prior to any amplification stages? Are there other reasons to use higher rails in a preamp?
In Small Signal Audio Design, for instance, Doug Self advocates using a +/-17v supply, because it's the highest that most opamps will tolerate. I suppose that might be because he favors putting the volume control in the last stage, in order to avoid re-amplifying a previously attenuated signal. In that case, it makes sense to want maximum headroom in prior stages because any boost is applied to a source signal of unknown level. Would that still be a concern, though, if the volume control were placed up front, prior to any amplification stages? Are there other reasons to use higher rails in a preamp?
That was the standard for older op-amps. The new op-amps do just fine with lower voltages. I can't speak for Self, but it appears he is keeping the signal voltage as high as possible (thus needing higher supply voltages) because that reduces noise.
As long as you know your maximum signal voltage and set the rails a few volts above that, and verify that the manufacturer has stated that the voltage falls within what they consider tolerable, you'll be fine at a lower supply voltage.
Case and point: A lot of circuits use a 9V battery for each rail and perform adequately.
As long as you know your maximum signal voltage and set the rails a few volts above that, and verify that the manufacturer has stated that the voltage falls within what they consider tolerable, you'll be fine at a lower supply voltage.
Case and point: A lot of circuits use a 9V battery for each rail and perform adequately.
You would think so (and I thought so), but.... take a look at this,
http://www.diyaudio.com/forums/everything-else/249479-purely-subjective-test-do-you-prefer.html
Have you repeated that test with many op-amps? It could be device dependent.
I've had very good luck with OPA2134 at just about any voltage on the datasheet. Besides, +/- 5V is probably cutting it close for anything more than 500mv, even though most datasheets claim within 2V of supplies is okay.
I haven't repeated it with many but I found the TL072 was similar whereas the LM4562 seems excellent. Strange thing is that the data sheet for the TLE2072 specifically states its fully specified at -/+5volts. I would have thought that a 5.6 pk/pk volt swing on -/+5 volt rails would be fine. There wasn't even any heavy loading either.
I suspect that even -/+ 6 or 7 volts would have made a big difference tbh. Its a very strange result indeed.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.