From: "Bill Sloman"
Subject: Re: linear power supply noise questions
Date: Tue, 12 Nov 2002 23:52:52 +0100
Organization: Planet Internet
NNTP-Posting-Date: 12 Nov 2002 22:52:47 GMT
X-Newsreader: Microsoft Outlook Express 6.00.2800.1106
"Walter Harley" wrote in message
> This weekend I was fooling around with a simple noninverting audio opamp
> circuit. The feedback divider was AC-coupled to ground, rather than
> DC-coupled, to avoid amplifying the DC offset of the opamp along with the
> audio. However, in order to avoid using a nonpolar electrolytic, I
> to couple it to the negative supply, rather than to ground per se,
> that since the AC source impedance of the negative supply should be very
> it wouldn't make any difference.
> Bad plan! There was somewhere around 0.5mV rms of broadband noise (not
> sitting on the supply rail, and that of course coupled right through to
> output. In fact, the way I discovered it was by pulling the opamp from
> circuit and observing that the noise persisted, unless I turned off the
> power switch. I changed the circuit to couple to ground instead of the
> negative supply, and got a 30dB improvement in noise at the output.
> I have three questions:
> First, I'm using an LM337T as the negative regulator. It's bypassed at
> input with a 0.1uF ceramic, and at output with a small tantalum (don't
> remember value offhand). Am I right that the next thing I should do to
> reduce the supply noise is bypass the ADJ pin, presumably with a ceramic
> polypropylene cap? Or would I be better off putting a poly or ceramic cap
> across the output (in parallel with the tantalum), or should I do both?
> Second, is this problem typical, and is that why these circuits usually
> a cap to ground instead of a supply rail? It seems like it would be
> engineering to use a cheaper, more widely available part in place of the
> nonpolar cap, but not if it means increased noise.
> Third, my Hitachi V212 oscilloscope's max calibrated sensitivity is 5mV
> division; with a x10 probe, that's 50mV per division, which means I can
> barely see the 0.5mV supply noise (instead, I discovered it with an Audio
> Precision analyzer). Do people generally use scopes with higher
> sensitivity, or outboard preamplifiers, or is a scope just not the right
> tool, or should I get my eyes checked?
The data sheet . of the LM337
includes an RMS voltage noise spec on page 2 - typically 0.003% - which is
30uV per volt of output. Most regulators include a similar sort of crude
noise specification, at a similar sort of level - it is pretty much the
noise on the band-gap reference.
If you have to get rid of it, you need a passive filter after the output.
For you application, a simple RC might be good enough, but this is
effectively just another way of making a "non-polar" electrolytic capacitor
out of two normal electrolytics of twice the capacity.
Bill Sloman, Nijmegen