Talk:Power supply rejection ratio

[Untitled]
I would define the whole story differently.. There are multiple issues with it..Something like this...( in order to make it not too complicated .. )

======================================================================================= An electronic circuit is normally characterized  by at least one input signal in(f), one output signal out(f) and a power supply p(f) so all signals are a function of frequency f. The transfer function h(f) = out(f)/in(f) is the intended transfer function and should be independent of p(f). In order to characterize the impact of any changes in the power supply on h(f) it would be possible to characterize the transfer function t(f) = out(f)/p(f). In this case the definition would be similar to the standard way of describing transfer functions as it is normally done as the ratio of output vs input. For the power supply rejection it has been defined the other way round for probably historically reasons.

So the power supply rejection ratio is a frequency dependent function and defined as

PSRR = 20log[p(f)/out(f)]

The definition is independent of the electronic circuit which has been realized. This parameter can be measured for any circuit. In real circuits the PSRR is also depending on other parameters as well like temperature, imperfections of the implemented circuits and other parameters.

=========================================================================================

Is it worth putting in some typical values, or an example graph of PSRR vs freq? Maitchy (talk) 05:50, 9 May 2009 (UTC)

There is an error in the definition
According to Sergio Franco, the definition of PSRR(dB)=20*log(ΔVcc/ΔVios) and the article says that PSRR is the parameter that defines the variation of supply voltage with output voltage. Therefore the article should be changed because is wrong. —Preceding unsigned comment added by Dudu.bassman (talk • contribs) 15:20, 10 June 2010 (UTC)


 * Yep, I changed it. (Although I sometimes see the ratio expressed in terms of output voltage - but I think that is a mistake, as is using the term PSRR for voltage regulators). I also gave examples. Notice the ratio is meant to be a POSITIVE number of decibels whether we're talking op-amp PSRR or voltage regulator RR... a NS datasheet for the LM723 confusingly had the RR stated as DELTA Vout/DELTA Vin yet still had a positive 64dB ratio!) Maitchy (talk) 01:44, 11 November 2010 (UTC)

It seems to me, there is an obvious mistake in the exemple. It is based on the false assumption that PSRR(dB)=10*log(ΔVcc/ΔVios) for voltage, while it is valif for power. The correct expression PSRR(dB)=20*log(ΔVcc/ΔVios) would result in an equivalent noise of 1 mV. IF ONE ELSE AGREE, PLEASE EDIT THE ARTICLE. — Preceding unsigned comment added by Paulo Farina (talk • contribs) 12:55, 12 April 2011 (UTC)

From the book "CMOS analog circuit design" by P.Allen, the PSRR is defined as the product of the ratio of the change in supply voltage to the change in output voltage of the op amp caused by the change in the power supply and the open-loop gain of the op amp. — Preceding unsigned comment added by 175.143.106.201 (talk) 14:18, 27 December 2012 (UTC)

AKA...
Also known as Power Supply Ripple Rejection (PSRR) in the UK. Measured in dB. --Quatermass (talk) 09:23, 24 June 2011 (UTC)


 * True, Power Supply Ripple Rejection is mentioned just now in a newsletter from Linear Technology, USA. The related product in the web: http://www.linear.com/product/LT3042 . If really thinking about the meanings, Power Supply Rejection is not very intuitive, as we don't reject the power supply, just the fluctuations in the power. But I understand it's not the literal interpretation that counts here, just the usage. Redirect alternative? Some-E (talk) 07:04, 3 March 2015 (UTC)