For a parallel RLC notch filter, all components are in line with the driver. I have noticed improvement of sound quality with better quality R, L, and C components for this case. For instance, I replaced a NP cap with a better quality polypropylene cap. Also I replaced a 20 ga coil with a 16 ga coil. I don't know if the sound quality would improve further with more expensive cap and coil.
For a series RLC notch, where the components are shunted across the driver, that part of the signal bypasses the driver and doesn't go through it.
For a series RLC notch, where the components are shunted across the driver, that part of the signal bypasses the driver and doesn't go through it.
Last edited by a moderator:
I try to use Jantzen zcaps and air-cores for both first-order XO and parallel notch filter (really the same thing) because they sounded much better than the locally made parts. Blue/red/silver zcaps (cocktail) to complement/tune the driver: silver frictionless, red tackier, blue neutral. I also use OCC interconnect wires. Fullrange driver speakers require fewer parts but demand better quality and tuning -- the SQ bar can be very high.
@Cal Weldon No, I'm sorry, he's not correct. At least not with the circuit in his attachment as a reference. In fact, it's even labeled as "flattens impedance"....because that's all it does.
With a typical power amplifier, it wouldn't change the voltage or current supplied to the driver at all.
Most power amplifiers are close to voltage sources. If that's the case, that RLC circuit will do nothing at all to alter the drive to the driver.
Now, if a power amplifier had significant output resistance and/or another component was introduced in series (maybe an inductor), then we'd have a different situation.
'Better brush up on your basics Cal.
Dave.
With a typical power amplifier, it wouldn't change the voltage or current supplied to the driver at all.
Most power amplifiers are close to voltage sources. If that's the case, that RLC circuit will do nothing at all to alter the drive to the driver.
Now, if a power amplifier had significant output resistance and/or another component was introduced in series (maybe an inductor), then we'd have a different situation.
'Better brush up on your basics Cal.
Dave.
Last edited:
@Cal Weldon Ah, I see what you mean now. I didn't notice that previously.
The labels being incorrect in his pictorial caught me. The series should be the parallel and the parallel should be the series. At least that is the traditional definition of those topologies.
Regardless, his last statement....."For a series RLC notch, where the components are shunted across the driver, that part of the signal bypasses the driver and doesn't go through it." is still incorrect.
Dave.
The labels being incorrect in his pictorial caught me. The series should be the parallel and the parallel should be the series. At least that is the traditional definition of those topologies.
Regardless, his last statement....."For a series RLC notch, where the components are shunted across the driver, that part of the signal bypasses the driver and doesn't go through it." is still incorrect.
Dave.
Last edited:
@Cal Weldon No, that is not the traditional definition of those filters. A series notch filter has the components located in series with the driver. A parallel notch filter has the components located in shunt with the driver.
It is not uncommon for these filters to be mislabeled. If you Google you will find depictions in both ways. We will have to agree to disagree on this one.
I'm sorry, but you are not reading his last statement literally. I explained the rationale already.
There is no "signal bypassing the driver and not going through it" with that configuration.
Dave.
It is not uncommon for these filters to be mislabeled. If you Google you will find depictions in both ways. We will have to agree to disagree on this one.
I'm sorry, but you are not reading his last statement literally. I explained the rationale already.
There is no "signal bypassing the driver and not going through it" with that configuration.
Dave.
For an example, the Lars Risbo App Note that caused all of the DIYaudio discussion recently does correctly label/identify the applied notch filters........regards "series" and "parallel."
https://purifi-audio.com/blog/app-notes-2/low-distortion-filter-for-ptt6-5x04-naa-11
Dave.
https://purifi-audio.com/blog/app-notes-2/low-distortion-filter-for-ptt6-5x04-naa-11
Dave.
For what it's worth, the terms series notch and parallel notch are inherently confusing.
Most of the references I've looked at refer to the configuration where components are in parallel - and then all of that in series with the driver - is referred to as a "parallel notch". For example, the Loudspeaker Design Cookbook uses this convention.
That is why I attached the picture with labels on post #4. (this was borrowed from another thread on this forum)
@pinholer Agreed. every component in a parallel or series network will have some effect. Usually, a parallel notch can make the most noticeable effect on frequency response. A series notch is used for impedance compensation as @Davey notes, but often has negligible (or at least not noticeable) effects on the frequency response.
Most of the references I've looked at refer to the configuration where components are in parallel - and then all of that in series with the driver - is referred to as a "parallel notch". For example, the Loudspeaker Design Cookbook uses this convention.
That is why I attached the picture with labels on post #4. (this was borrowed from another thread on this forum)
@pinholer Agreed. every component in a parallel or series network will have some effect. Usually, a parallel notch can make the most noticeable effect on frequency response. A series notch is used for impedance compensation as @Davey notes, but often has negligible (or at least not noticeable) effects on the frequency response.
It has been for the 50 plus years I've been playing this game. The only time I hear it any other way is from persons confusing the terms, I can't say I've seen it wrong in print before. Even persons like our old friend Sreten got it wrong.
Last edited:
@Cal Weldon Well, you're not paying close enough attention then. I've seen it used both ways myriad times.
And, I just gave you an example of a respected speaker designer who used the terminology correctly. (or incorrectly, in your opinion.)
The comment from @Dave R though is not incorrect. The terms can be confusing.
Dave.
And, I just gave you an example of a respected speaker designer who used the terminology correctly. (or incorrectly, in your opinion.)
The comment from @Dave R though is not incorrect. The terms can be confusing.
Dave.
Last edited: