Talk:Electronic filter

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

See also[edit]

Talk:Linear filter.

"Electrons of atoms?" Huh?[edit]

"For even higher frequencies and greater precision, the electrons of atoms must be used." What does this mean? --Mintie 30 June 2005 00:47 (UTC) Answered via edit. --Mintie 30 June 2005 06:38 (UTC)

Page redirects from "Audio crossover", but...[edit]

Page redirects from "Audio crossover", but there is no mention of audio-specific features and/or requirements for a filter used in audio crossovers. --Rohitbd

It might be better to aim that at Loudspeaker (although that might end up circular!). Or we could write an article, ehh? :-)
Atlant 6 July 2005 15:08 (UTC)
I think it should be included in the filters article as a dedicated section - because crossovers are filters after all... Rohitbd
On second thoughts, audio crossovers may be treated in a separated article as there is a great deal of information specific to audio crossovers...This article can link to both loudspeakers and filters and vice-versa. Rohitbd
Definitely. - Omegatron July 6, 2005 17:41 (UTC)
With conviction and enthusiasm. --Smack (talk) 7 July 2005 04:18 (UTC)
Created the audio crossover article. Please have a look and make changes as necessary. Thank you. -- Rohitbd
Thanks! Atlant 7 July 2005 11:48 (UTC)
My pleasure :-) Rohitbd

Article Rewrite[edit]

The subject of electronic filters is so vast that this article is going to need a heck of a lot of work before it looks anything like decent.It will eventually need splitting up no doubt but headings can be added now.

I'll do abit =- but could anyone else help ( even if your not too sure of your facts as the errors will get corrrected). Al

filter in general[edit]

filter (signal processing) redirects here, but I think it should be an article on its own. Digital filters aren't really "electronic". — Omegatron 01:07, 6 December 2005 (UTC)[reply]

Not only electronic filters at the moment[edit]

The page mentions some other filters as well, like the atomic filter, which is definitely not an electronic filter. Perhaps Filter (signal processing) should not be a redirect to this page?

I want to link to an article about filters in general (see the last paragraph of Spectral splatter) but there's only this page. The reference to filters in Spectral splatter applies both to Audio filters and Electronic filters, and any other generic signal processing filter.

I think Filter (signal processing) should exist. Romanski 19:42, 14 June 2006 (UTC)[reply]

I took the liberty of editing spectral splatter to link to linear filter -- is that closer to what you wanted? Should filter (signal processing) also link to linear filter ? --75.19.73.101 (talk) 08:17, 15 December 2007 (UTC)[reply]
Today I find the stub analogue filter. Should we move information about filters that are technically not "electronic" from "electronic filter" to "analogue filter"? --75.19.73.101 (talk) 20:03, 16 December 2007 (UTC)[reply]

Single pole types etc[edit]

Single and multipole are implied by the structure to be types of passive filters. There is no discussion what a pole is (which may not belong here. Perhaps a mention that it is a measure,in one sense, of how complex the filter is?). It seems to imply that (R)LC circuits are single pole but the simplest ones are two pole.

 3ch Colonel hack 03:47, 17 June 2006 (UTC)[reply]
  • I would also very much like to know what a pole/zero is in relation to filters. There's Pole (complex analysis) and Zero (complex analysis), but they don't seem to mention filters. I can't find out anywhere else on the Internet either (or indeed in the recommended text). --holizz (talk) 17:35, 20 January 2008 (UTC)[reply]
yeah, complex analysis is exactly right. Poles and zeroes are important for amplifier stability as well as filters. Look at Nyquist stability criterion, that article pretty much explains what poles and zeroes are all about in terms of electronics. SpinningSpark 22:45, 19 April 2008 (UTC)[reply]
I dare say Colonel hack is not around any more, but his comment is wrong, RC, RL, are single pole, LC and RLC circuits are all single two-pole and, T and Pi filters are 23-pole. That is not so clear from the headings. In any case, it is a confusing classification. Number of filtering elements is what I think the author was trying to say. The LC circuit is a single filtering element in the context of band pass filter.SpinningSpark 22:51, 19 April 2008 (UTC)[reply]

PI and T filters ...[edit]

I've noticed we have all the fancy names of the filters, but not the common names. Almost not a mention anywhere of a pi type filter unless i've missed something ... -- Wirelain 19:33, 1 November 2006 (UTC)[reply]

Time for you to be bold and corect this oversight!
Atlant 20:20, 1 November 2006 (UTC)[reply]

Yes, please explain why they have these names. My guess is that the horizontal wiggle in the Pi symbol, is somehow related to the horizontal inductor in the LP, whereas the two serial capacitors + one vertical inductor in the HP filter, look like a Times Roman "T". Jahibadkaret (talk) 08:28, 17 April 2019 (UTC)[reply]

proposed merger with brick-wall filter[edit]

I considered trying a merge, and having brick-wall filter REDIRECT into a new sub heading under electronic filter#Classification by transfer function but neither seemed as useful as providing a short explanation before linking. My motivation for brick-wall filter was finding the term used in other articles and struggling to understand it. I didnt call it a stub as I didnt think it could usefully be extended. It was more of a REDIRECT (with explanation). Rod57 (talk) 18:40, 25 October 2008 (UTC)[reply]

I would dissaprove of putting it in here. This article is about (or should be about) filters that can be built with electronic components. Notwithstanding that there is already a large amount of stuff here that also should be in some other article (WP:other stuff exists) I am against because this is not a real filter. Rather it is a theoretical artefact that cannot actually be achieved in practice, just approximated to more closely with more complex filters. Filter design might be a better place, but you should note (and probably any mention of this in an article should note) that this is more or less engineer's slang and not (at least as far as I know) a mathematically formally defined term. In particular, the term says nothing about the phase response of the filter which can be all-important in signal processing.
Looking at the history, your problem seems to be that someone is trying to speedy delete your article rather than that you truly want to merge it anywhere. The grounds for speedy is A1 which is "no context". You will save the article form this just by adding a couple of introductory sentences to explain the context. It could do with some better quality refs as well. SpinningSpark 20:49, 25 October 2008 (UTC)[reply]

Remove "analog filter" and "digital filter" pages[edit]

The analog filter and digital filter pages give the impression that whether a filter is "analog" or "digital" has everything to do with the domain over which the signal is defined. In other words, the "analog" and "digital" filter pages read more like "continuous-time" and "discrete-time" filter pages. Despite its other weaknesses, electronic filter does a better job explaining the distinctions between all of these different adjectives. In particular, the first paragraph of electronic filter admits that there can be discrete-time analog signals; that's a major improvement over the analog filter and digital filter pages. Perhaps those pages can be merged into this page? Or perhaps separate continuous-time filter and discrete-time filter pages can be made? —TedPavlic (talk/contrib/@) 00:06, 4 July 2009 (UTC)[reply]

Add some clarifying detail and those titles as redirects; the name digital filter is pretty conventional, so should probably stay as is. Though there is no analog filter article now – it's a peculiar redirect, which might be worth sorting out. Dicklyon (talk) 06:54, 27 August 2009 (UTC)[reply]

Polynomials[edit]

I don't understand a single thing of what you are saying, I'd just like to know how you merge the theoretical polynomials with the practical. This sounds stupid, I know, but still I'd like to see some more than theoretical assemblance. —Preceding unsigned comment added by Knoppson (talkcontribs) 01:22, 27 August 2009 (UTC)[reply]
There's got to be some more clearifying way to make use of the different polynomials. Can't somebody make an example? This is exciting!Knoppson (talk) 03:00, 27 August 2009 (UTC)[reply]
K, I did some work on Bessel filter to try to give you a better clue what the polynomials are about; look at the pictures and the equations, follow the links, etc., and ask again if this isn't doing it for you. And I'm unclear about why you added "sources" there; were those sources used in writing the article? Are they accessible? I found a good ref online and added that as a proper citation. They way anyone can look, verify the info, and learn more about it. Dicklyon (talk) 06:53, 27 August 2009 (UTC)[reply]

transient response[edit]

Almost all the discussions I have seen of filter designs have talked about filter response in the steady state. I believe from experience that in some circumstances it is necessary to include considerations of the filter transient response in the design. For instance, I had to create a 50 kHz bandpass filter to isolate a signal that had also been chopped at a low frequency, 30 Hz. My first bandpass design used a combination of low-pass and high-pass Butterworth filter sections. Despite a theoretically great low frequency attenuation, the low frequency chopping signal appeared clearly in the filter output. I had much greater success when the bandpass filter was redesigned using Bessel filter sections, despite the theoretically less low frequency attenuation. I attribute this result to the differences in the transient response of the filter types. While this may be an esoteric special case, it was a real problem. I have always wished since to see a discussion of filter design that could have provided useful guidance. --AJim (talk) 02:33, 8 September 2009 (UTC)[reply]

ALL filter theory and most design assumes a perfectly linear response to any stimuli. In practise if your signal exceeds the linear range of your components you will have distortion. In your case, assuming the same components and signal strength, the Bessel filter may benefit you from having a linear phase passband. 178.15.145.204 (talk) 09:48, 22 March 2012 (UTC)[reply]

ideal elements[edit]

A reference was just added to a single "hybrid" (combined LC) element filter. I have not read the reference. Whether this reference is kept or not is not my concern. My point is that this "muddies the waters" for newcomers, and that this is inevitable. I think, as a consequence, that the article should go a step further and introduce the distinction between theoretical circuit elements, as used in the article, and actual components. The article should explain that the "pure" theoretical elements discussed here can only be approximated in the real world; a complete model of a real-world element always contains all 3 theoretical types (R,L, and C). I think there must already be an article or articles showing elaborated models of real elements. The article could define or link to the ideas of the "inductance" and "capacitance" of a real resistor, for instance. The article could then explain how these "stray" or "parasitic" elements appear in models of real filters and how they affect the behavior of a filter, and the circumstances under which they can be ignored, and when the matter a great deal. It would help newcomers to understand what is meant by a description such as a "non-inductive" resistor, or a "low equivalent series resistance" capacitor, and when having such components might matter in a filter. The article could also explain the distinction between "lumped" and "distributed" elements. AJim (talk) 19:14, 15 November 2010 (UTC)[reply]


Silly question[edit]

Just wondering if anyone with relevant background could enlighten me on an off-topic. My question is : is it possible to approach the problem of wind turbulence, with regards to wind mills through "turbulence filtration" using the same approach as with electronic or audio filtration ? —Preceding unsigned comment added by 193.157.194.209 (talk) 14:27, 22 December 2010 (UTC)[reply]

How does this relate to power supplies!?[edit]

I do think the output filters in power supplies are important, however its not even talked about here! How are they designed (eg, types, frequency and current considerations like inductor core heat dissipation)? Charlieb000 (talk) 01:34, 21 June 2013 (UTC)[reply]

Boucherot[edit]

The page on Boucherot filters links to here but the topology (R and C in series to gnd) is not described on this page.

https://en.wikipedia.org/wiki/Boucherot_cell

Kenif (talk) 04:56, 16 March 2014 (UTC)[reply]

Your characerisation of the Boucherot cell is incorrect. The RC element is only half the cell, the LR element is the other half. SpinningSpark 00:30, 17 March 2014 (UTC)[reply]
The Boucherot cell is the R and C; this cell is half the circuit. So, yes, it would not make a lot of sense to have it analyzed in isolation in this article. Probably linking here is OK though. Dicklyon (talk) 02:44, 17 March 2014 (UTC)[reply]

Layman's Terms[edit]

As a layman, I come across words that I am unfamiliar with on these articles, such as "Attenuation", that are not linked to their corresponding article. Would it be better to use the technical term and link it to the corresponding article or rephrase the sentence and use simpler words for ignorant, but enthusiastic newbies like myself? — Preceding unsigned comment added by 75.142.183.45 (talkcontribs) 26 jul 2015 01:37‎ (UTC)

I have wikilinked attenuation in the article. Feel free to add links to other appropriate pages. Technical terms should always be explained, either directly, with a link, or with a glossary. I think attenuation is a simple enough term, it is only one word. SpinningSpark 01:32, 8 August 2015 (UTC)[reply]

Distributed element filters added back in[edit]

@Dicklyon:, the navigation established at Filter (signal processing)#Technologies clearly points lumped circuit filters to electronic filter and distributed element filter to its own separate article. Delay line filters should also have their own article and these are pointed to (the as yet unwritten) transversal filter. A lot of material has previously been moved off this page into separate articles. Do we really want to start expanding out this page again to other topics? In my view, material on other methods of filter construction should be added to the top level filter page if they don't already have their own page. SpinningSpark 16:58, 8 September 2019 (UTC)[reply]

I'm OK if we make this article be about lumped-element filters, but that's not what the title suggests. Maybe there's a better descriptor for that class? I mean, the problem is that as i started reading it, it seemed artificially limited to lumped circuits, but didn't really clearly say so or point to where the rest of the class could be read about. It read, "Electronic filters are a type of signal processing filter in the form of electrical circuits consisting of discrete (lumped) electronic components. Such filters ..." in which the second sentence would apply just as well to the broader class, so the "consisting of discrete (lumped) electronic components" seemed out of place. If you want to revert that and fix some other way, that's great. Dicklyon (talk) 22:03, 8 September 2019 (UTC)[reply]
Do you have a suggestion for a title? And where should the current title redirect? I don't want to rename such a long standing title without at least some cursory discussion. Perhaps a better idea is just to explicitly clarify the scope in the lead or a hatnote. SpinningSpark 17:57, 9 September 2019 (UTC)[reply]
By the way, there was already a hatnote clearly defining the scope before you started making changes. I'll add a link to DEFs in the hatnote to make it completley explicit. SpinningSpark 15:25, 11 September 2019 (UTC)[reply]
OK, thanks. Sorry I missed that. Dicklyon (talk) 16:45, 11 September 2019 (UTC)[reply]