Hi,
Maybe my question has been answered before, but:
Is the sequence of the filter elements in a passive bandpass filter important ?
Many bandpass schematics show a HP followed by a LP filter.
But to me it seems better to have a LP followed by a HP filter.
(I use a kind of 2. order filters for both LP and HP in the Bandpass filter)
It also seems to sound better in my current DIY project.
The Zin impedans also seem to be higher this way!
Any comments or explanations?
Cheers
Flemming
Denmark
Maybe my question has been answered before, but:
Is the sequence of the filter elements in a passive bandpass filter important ?
Many bandpass schematics show a HP followed by a LP filter.
But to me it seems better to have a LP followed by a HP filter.
(I use a kind of 2. order filters for both LP and HP in the Bandpass filter)
It also seems to sound better in my current DIY project.
The Zin impedans also seem to be higher this way!
Any comments or explanations?
Cheers
Flemming
Denmark
Last edited:
With some filters such as the first order bandpass filter, there is no actual difference in swapping the locations of the low and high pass elements.
With some other filters the argument could be made that having the high pass first reduces the current that flows through the low pass sections increasing the levels required to saturate inductors, etc.
Regardless, the filters interact and cannot be simply swapped. They can be redesigned to be the other way around, but just swapping them may give unexpected side effects.
With some other filters the argument could be made that having the high pass first reduces the current that flows through the low pass sections increasing the levels required to saturate inductors, etc.
Regardless, the filters interact and cannot be simply swapped. They can be redesigned to be the other way around, but just swapping them may give unexpected side effects.
The primary issue is the movement of inductor coil resistance. When you start moving that around, some small changes will take place due to the way it effects other x-over components and their effects on the system.
I would be hard pressed to believe that an average human ear could detect any change without a side-by-side instant swap of the bandpass components. The difference would be pretty minor in most cases.
I would be hard pressed to believe that an average human ear could detect any change without a side-by-side instant swap of the bandpass components. The difference would be pretty minor in most cases.
Last edited:
If you are not talking about a crossover or high-current filter, you should consider inserting a buffer amplifier (op amp) before and/or after each stage, or just use active opamp filters. Download the free filterpro version from ti.com. Your results can be dramatically improved, without much cost and with very little complexity.
For passive line-level or low-current filters, you can do whatever works. Theoretically, the sequence would not matter. But in practice you have to worry about the impedance-loading effects and any other interactions between the stages. If you use unity-gain buffers, you get close to the theoretical responses.
There are very good chapters on filters in this free book:
http://www.analog.com/library/analogdialogue/archives/39-05/op_amp_applications_handbook.html
For passive line-level or low-current filters, you can do whatever works. Theoretically, the sequence would not matter. But in practice you have to worry about the impedance-loading effects and any other interactions between the stages. If you use unity-gain buffers, you get close to the theoretical responses.
There are very good chapters on filters in this free book:
http://www.analog.com/library/analogdialogue/archives/39-05/op_amp_applications_handbook.html
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.