diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Parts (http://www.diyaudio.com/forums/parts/)
-   -   Question about Zobel/Series notch filter (http://www.diyaudio.com/forums/parts/112592-question-about-zobel-series-notch-filter.html)

silentwf 26th November 2007 02:35 PM

Question about Zobel/Series notch filter
 
Hi everyone, new here (first post too XD)

I'm not sure if this post is in the correct spot, but since it has something to do with crossovers and electronic parts...i hope this place is the correct spot.

ok...to main point

My main question is: are Zobel and Series notch filters necessary in passive crossovers?

My woofer driver is a MG18SK09 4ohm one by Vifa (product specs: http://www.madisound.com/catalog/pro...oducts_id=1125) and then there's this big mountain of impedance at around 20~100Hz. If I installed a series notch filter into the crossover, will I remove this impedance? If it does, does my bass response become better? Or does it have no effect?

I read another source of information from lalena.com[1] and it seems pretty confusing (at least to me). The last graph of that page seems to say that the series notch filter actually pushes the entire SPL line to the right, and that would mean decrease in bass. So I'm really....LOST! Can any one clear things up for me?

[1]http://www.lalena.com/Audio/Calculator/SeriesNotch/Help.aspx

consort_ee_um 26th November 2007 03:48 PM

Welcome
When you look at an Impedance graph it is non linear but this does not necessarily mean you need to compensate. This article explains some of the issues:
http://www.audiodiycentral.com/at_jeffxovertips.shtml

AndrewT 26th November 2007 04:03 PM

Hi,
the resonant peak at the low frequency end compensates for the higher sensitivity at these frequencies.
If you deliberately suppress the peak then the driver will play quieter than if left alone because the amp sees the high impedance around the peak in parallel with the correcting network and the amplifier output voltage falls because it's does not have a perfect voltage source capability.

The ideal passive filter can only achieve the correct response shape when connected to zero source impedance and to a resistive load. The rising response of the driver causes the filter to behave in a far from ideal manner.
Applying a correction to the rising impedance allows the filter design to be easier, but that does not necessarily make the combination of filter and driver sound better. The filter can take account of the reactive load and be "voiced" to sound nice by selecting non ideal filter components that compensate for the drivers anomalies.

All these compensations have the overall efect of reducing the sensitivity of the speaker and add extra loading on to the amplifier.
If you can get good sound from the crossover without applying these extra corrections, then you gain twice, the amplifier sees an easier load and the speaker needs less power to sound just as loud. But this style of designing needs a lot of experience or luck.

dangus 27th November 2007 01:58 AM

I did some playing around with a "textbook" second order crossover, with and without a "rule of thumb" zobel. The measured electrical response was very much better with a guesstimated Zobel than without. The rule of thumb was to use a resistor the same impedance as the woofer, and a capacitor with the same value in microfarads as the the woofer nominal diameter in inches. (one inch = 2.54 cm).

silentwf 27th November 2007 11:31 AM

Hi
Thanks for the response. So according to the three replies, a series notch filter is not necessary (as in, it wont help with any of my acoustics)? Although dangus mentioned about better electric response, but does that also mean better acoustic response?

My friend and I think that if it doesn't affect the acoustic output much, then we dont think we should toy around with it.

AndrewT 28th November 2007 04:04 PM

Hi,
a series notch filter may be required to suppress a resonant peak either within the passband of one of the drivers or outside the passband but still with significant signal causing some acoustic anomaly.
If needed, then it will sound odd without it.


All times are GMT. The time now is 04:44 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2