Zobel Network Simulations

[Hari Iyer]

Over the past couple of months I was studying the effect of Zobel network on filter phase and how it can alter imaging and transparency in the sound stage. Below are some simulations and findings,
1. Rz=4.2 ohms and Cz=20mfd
The cross-over frequency is 1925Hz (first order) as can be seen by the filter phase. The filter phase perfectly bisect the low-pass and high-pass roll-off to give precise imaging and transparency. Also the impedance phase is showing resistive loading after the cross-over frequency.

2. Rz=4 ohms and Cz=22mfd
Altering the zobel resistance and capacitance value gave the below response,

As you can notice even by slight modification of the zobel resistance and capacitance gave me a huge shift in the filter phase which is now around 1816Hz and does not bisect the low-pass/high-pass FR roll-off. Also the impedance phase gets slightly inductive before the cross-over value which could result in loss of imaging and transparency.

3. Rz=4.5 ohms and Cz=15mfd

In this case too there is a change in the filter phase now to 2234 Hz and it does not bisect the low-pass/ highpass FR roll-off. Again this will cause imaging and transparency issues and the low-frequency will be very loose.

Conclusion: Its quite important that the values for the zobel resistance and capacitance are calculated accurately and they are also in-sync with the high-pass filter response for getting the correct imaging and transparency. Even modification of the values by few fractions can play havoc with the overall sound stage.

 

[keith_correa]

This imaging, transparency, sound stage and bass issues - are they so pronounced that you actually heard these issues?

 

[Hari Iyer]

Yes, with the phase shift the most significant impact is the shifting of the low frequency center imaging moving off-centre and the vocal moving above the horizontal plane by about 2 to 3 feet. The sound stage also does not seem to be involving with the phase shift.

 

[keith_correa]

the vocal moving above the horizontal plane by about 2 to 3 feet

Some would call that a feature

 

[Hari Iyer]

Some would call that a feature

yeah, it would give you a feeling that you are seated below the stage and the orcheastra is playing at a higher level-

 

[Fantastic]

Has this actually been tested practically by a group ?
Amazing behaviour with very small changes. Are you postulating the changes or you have actually tried it and you have heard the difference ?
Why does the bass get 'loose' ?
Cheers.

 

[Hari Iyer]

Has this actually been tested practically by a group ?
Amazing behaviour with very small changes. Are you postulating the changes or you have actually tried it and youI'm myave heard the difference ?
Why does the bass get 'loose' ?
Cheers.

Not tested by a group but only by me. The changes were not done immediately but over a period of time but have been documented. These details are from my own experiences with my TL FS speakers build.
I think the bass looses punch due to the filter phase shifting. The speakers are placed at the same spot before and after modifications of crossover but the results were amazingly different. In my experience if there is a loss of low frequency centre imaging in almost all tracks then probably three could be filter phase issues. IMO no amount of tampering with speaker placement or using room treatment is going to be of much help. Instead getting the filter phase correctly could be a easy and cheaper option.

 

[Nikhil]

Hari,

Can you please explain what is being graphed by the 5 curves you have shown?
I tried zooming in but could not make out anything.

Best Regards.

 

[Hari Iyer]

Hari,

Can you please explain what is being graphed by the 5 curves you have shown?
I tried zooming in but could not make out anything.

Best Regards.

@ Nikhil,
If you notice the first graph the low pass roll off (green) and high pass roll off (red) bisect the filter phase (yellow) at around 1924Hz which is the cross-over frequency. Also the lower impedance phase (dark green) offers a resistive load from the crossover point (1924Hz onwards) o 0 deg phase)

If you notice the other two graphs the filter phase is not bisecting the low pass and high pass roll off but is a little offset. In the second graph it's around 1816Hz and in the 3rd graph it's around 2230Hz.

This means your FR rolling at 1924Hz but the filter phase does not and is not in sync with the amplitude response causing anomalies as explained earlier. If the phase is not accurate you can hear them as non-linearity in the overall speaker system response. It's precisely this reason that a phase control is given in all subwoofer amplifier to get them in sync with your main speakers.

The darker yellow line is the speaker system FR response and the lower red line the speaker system impedance curve.