Frequency Response Tailoring

Started by momo, January 26, 2021, 11:42:34 AM

Previous topic - Next topic

momo

Hello everyone, it's been a while since I posted.
I hope y'all are doing great in these difficult times..

I am looking to tailor the frequency response on a full spectrum speaker ( not a guitar speaker).
I am looking for some simple circuit, SMD if possible to cut off all frequencies below 500Hz.
I would pass a full freq signal thru this circuit and have it tailored as I explained.
Anyone know of an existing circuit for this?

thanks much !
"Alas to those who die with their song still in them."

GGBB

I assume you want to do this at the speaker, and the speaker is passive. In other words, connect the "circuit" between the amp output and the speaker. If not, please says so.

You need to build a high-pass filter. The simplest kind is a choke/inductor in parallel with the speaker. Kind of the opposite of a capacitor which provides a low-pass filter when connected the same way. Most passive two or more way speakers have these inside them.

There are many different ways to build a passive speaker-level filter. Back in the day when I was into car audio, car audio guys would do this for custom subwoofer and satellite installations with a single power amp. They probably still do, but I think active crossovers and bi-amping are more commonplace.

Anyway - this is the world of speaker crossover design. Searching on that subject should give you some results.
  • SUPPORTER

PRR

> a choke/inductor in parallel with the speaker.

That would short-out the amplifier for any bass sounds.

A series capacitor is first-order bass cut.

A series capacitor and shunt inductor is second-order bass cut.

As you say, these are "crossover" designs.
  • SUPPORTER

momo

#3
Thank you gentlemen for the responses and yes it would be a passive circuit at the speaker level.
I would want this thing to be as small as possible using SMC components...
I did a search here for a SMC high pass filter circuit and could not find.
Anyhow, if you have any suggestions.
thanks a bunch.
PRR, nice to see you are still here!
You helped me years ago on a Deacy style amp, worked great and I am still using it!
:icon_mrgreen:
"Alas to those who die with their song still in them."

iainpunk

its nor rally feasible to make a passive crossover with SMD components, the currents are way to high!!!!
you can look in to buying a passive crossover on the internet tho, they aren't too expensive.

cheers, Iain
friendly reminder: all holes are positive and have negative weight, despite not being there.

cheers

Rob Strand

#5
QuoteThank you gentlemen for the responses and yes it would be a passive circuit at the speaker level.
I would want this thing to be as small as possible using SMC components...
I did a search here for a SMC high pass filter circuit and could not find.

You want to remove "all the frequencies below 500Hz" but in practice this boils down to a trade-off
between how quickly the response rolls off below 500Hz and how complicate the crossover becomes.

Here's the response for first order and second order crossovers,  you can continue with third order and above.
For circuits made with passive components (inductors, capacitors, resistors) the order is generally between
1st order and 4th order.   Second order is probably the most common.


This site is a reasonable introduction,
https://www.calsci.com/audio/X-Overs.html

You would like to use SMD parts.   The size of the inductors (and capacitors for that matter) depends
on the signal levels.    A 400W system will need a very large inductor to handle the current.  A 400mW
system can use significantly smaller inductors.    A common HiFi speaker will use air-cored or ferrite rod
inductors which are large and definitely not SMD.    You can see the scale of the crossover parts,



For an 8 ohm speaker the caps tend to be large, say 22uF to 50uF for a 500Hz crossover.   The SMD parts in this size,
and you might need to parallel some, are ceramics.   Ceramics this size are very non-linear at voltages over 5V (that means distortion).

A final problem is the speaker impedance does not look like a resistor.    This affects the crossover response.    The way this is handled is with an impedance compensation network.   A low-pass filter will use a Zobel and/or a resonant network.  A high-pass filter, which is what you want, will often use a resonant network.

https://calsci.com/audio/X-Overs3b.html

You want to avoid these if you can because the parts end-up being expensive and it complicates the crossover.   You can often get away without them.   However, you need to tweak the values of the crossover parts to compensate for the speaker impedance ie. the part value formulas no longer work.    A small 3" speaker might have a resonance around 200Hz and this will have an effect on a crossover at 500Hz.
Send:     . .- .-. - .... / - --- / --. --- .-. -
According to the water analogy of electricity, transistor leakage is caused by holes.

GGBB

Quote from: iainpunk on January 26, 2021, 05:03:28 PM
its nor rally feasible to make a passive crossover with SMD components, the currents are way to high!!!!
you can look in to buying a passive crossover on the internet tho, they aren't too expensive.

I agree - easier to buy something pre-made. For example https://www.parts-express.com/Dayton-Audio-500-HPF-8-High-Pass-Speaker-Crossover-500-Hz-12-dB-Octave-260-1758.


  • SUPPORTER

momo

thank you very much for your research, this helps much..
"Alas to those who die with their song still in them."