Talk:Switched capacitor

Hi, the article states that the SC behaves like a LOSSLESS resistor. This is not correct. The capacitor itself is lossless but the equivalent loss that a conventional resistor would incur is realized in the switching devices. Someone else had previously mentioned that but left the article unchanged. So at this point I am removing the word "lossless". One could also explain how the SC incurs losses but I believe this would be beyond the scope of this article — Preceding unsigned comment added by BrunoHaider (talk • contribs) 00:24, 4 February 2014 (UTC)

Hi everyone, I added a short description that shows that Req = 1/(f*Cs). This is true of integrators; I don't know whether it also applies to other cases.

--Biscay 18:28, 15 December 2006 (UTC)

Improving this article
This article seems to focus on the mathematics of switched capacitors. I think it should include more on the application(s) - eg. the Switched-capacitor filter which currently redirects here.


 * There should just be more information. The mathematics part is a required part. If someone could make a picture of the timing of the 2 switches and the current flowing, that would help alot understanding this. Aphexer (talk) 11:19, 14 January 2009 (UTC)


 * Agreed, for someone who knows nothing about switched capacitors to see what's happening to the charge around the opamp without a picture is pretty tough... I guess that means I should work on it  Guerberj (talk) 16:06, 9 March 2009 (UTC)

Adding to WikiProject Electronics as a stub. --RobBrisbane 03:27, 24 June 2007 (UTC)

Diagram of MDAC does not contain a feedback capacitor. Is the switch in the feedback path tied to the wrong node? —Preceding unsigned comment added by 128.12.239.181 (talk) 19:09, 14 January 2010 (UTC)

The MDAC schematic is definitely wrong, the feedback should go to the input side of C1. Also, this isn't a particularly good MDAC - no auto-zeroing and the opamp output is left floating during phi1. Really, an MDAC is an article unto itself, though it can be mentioned here. —Preceding unsigned comment added by 129.34.20.23 (talk) 19:38, 15 June 2010 (UTC)

I was intrigued by the idea of a "lossless resistor"! I believe that the device is not lossless, energy is lost when the voltage on the capacitor suddenly changes from one value to another when it is switched. You can analyse this by assuming the switch has a small resistance, calculating the current transient and the resulting energy lost, then taking the limit as the resistance goes to zero. — Preceding unsigned comment added by 84.92.196.225 (talk) 16:15, 26 March 2013 (UTC)


 * Certainly correct. And the source is often modeled to have a series resistance while the switching devices are, by their nature, nonlinear. An opportunity, if one desired, to bring out the fact that switching capacitance into and out of a circuit generally leads to large transient currents, with consequent design issues: (1) current carrying capacity of the switching devices including their ability to dissipate of heat, (2) in the case of the now-rare use of mechanical commutation, management of arcing, and (3) unwanted coupling via induction and other means with nearby circuitry. 129.176.151.19 (talk) 12:53, 24 April 2022 (UTC)

Parasitic Insensitive Integrator Error
It appears the equation should really be Vfb(T) = Vfb(T-t) + C1/C2(Vin(T-t)). Perhaps I should update it. Guerberj (talk) 00:46, 18 June 2009 (UTC)

Should kT/C voltage noise be added to the "Difference with real resistor"
After this article says "SC circuits can have substantially lower Johnson–Nyquist noise" I'm thinking kT/C should be added. Plenty of search results for "kT/C switched capacitor" can be found. For instance, "Simulation and analysis of noise in switched capacitor amplifier circuits" says "the basic technique is to use large transistors and capacitance values such that kT/C noise dominates" and https://www.seas.ucla.edu/brweb/teaching/AIC_Ch12.pdf says "The problem of kT/C noise limits the performance in many high-precision applications. In order to achieve a low noise, the sampling capacitor must be sufficiently large, thus loading other circuits and degrading the speed." Em3rgent0rdr (talk) 04:55, 29 June 2024 (UTC)