Low pass filter help

Thread Starter

andyhof

Joined Jun 22, 2011
13
I need some help

I want to create a simple, passive low pass filter before this high frequency driver:

https://www.atlasied.com/d/3228/ATS001637 PD60A_PD60AT Datasheet Rev C.pdf

I want low pass filter to allow 4kHz and down to pass.

Audio amplifier is tube 40w 16ohm:
http://www.captain-foldback.com/Leslie_sub/Leslie_schematics/147.GIF

And it hits a passive dividing network before going to the high frequency driver:
http://lesliexovers.homestead.com/147Xover.html

So the amp is 16ohm out 40w vacuum tube
Woofer is 16 ohm
High frequency driver is 16ohm

I have no idea how to choose resistor type/value as well as capacitor size/value to make an RC lpf. Also open to a coil idea . Size is not an issue, I’ve got plenty of space. I’m a totally newbie to this but can solder like a champ!

Any and all suggestions welcomed.
 

Audioguru

Joined Dec 20, 2007
11,248
It is not a high frequency driver. Its very narrow frequency response cuts lows and cuts highs.
Its horrible old frequency response is fairly flat from 250Hz to 1700Hz much worse than an old telephone.
It is made to be very loud, not to make good sounds. It was used about 60 years ago.
 

Thread Starter

andyhof

Joined Jun 22, 2011
13
It is not a high frequency driver. Its very narrow frequency response cuts lows and cuts highs.
Its horrible old frequency response is fairly flat from 250Hz to 1700Hz much worse than an old telephone.
It is made to be very loud, not to make good sounds. It was used about 60 years ago.
I stand corrected - compression driver.
Yes, this is old technology.
Have a read:
http://www.theatreorgans.com/hammond/faq/mystery/mystery.html
Dating back in principle to 1939.

The original compression driver (Jensen v21) is being removed and replaced with the atlas. This “new” compression driver has slightly higher frequency response (undesirable) and is like to “roll it off”
 

Audioguru

Joined Dec 20, 2007
11,248
The Atlas compression driver already cuts frequencies above 1500Hz. From its peak of 330Hz to 1500Hz then 4kHz is down -12dB. 8kHz is down to -27dB. You could add an inductor in series with it to cut high frequencies a little more.
 

Thread Starter

andyhof

Joined Jun 22, 2011
13
The Atlas compression driver already cuts frequencies above 1500Hz. From its peak of 330Hz to 1500Hz then 4kHz is down -12dB. 8kHz is down to -27dB. You could add an inductor in series with it to cut high frequencies a little more.
The Atlas compression driver already cuts frequencies above 1500Hz. From its peak of 330Hz to 1500Hz then 4kHz is down -12dB. 8kHz is down to -27dB. You could add an inductor in series with it to cut high frequencies a little more.
I would absolutely like to cut high freq a little more... the frequencies drop off nicely until about 4.8kHZ... it's that spike at 5kHZ i'd like to get rid of .(i'd like to continue that nice slope drop-off that starts at 1500Hz

So an RL circuit where the output is taken across the resistor ?
 

Audioguru

Joined Dec 20, 2007
11,248
A resistor parallel with the speaker wastes amplifier power.
An inductor in series with a speaker cuts high frequencies at -6dB per octave which is very gradual. A 1mH inductor with a reactance of 16 ohms at 2.5kHz reduces 2.5kHz by -3dB and reduces the 5kHz peak by -6dB.
Since the crossover network is second-order then it creates a notch filter at the cutoff frequency unless the phase of one speaker is reversed, then it makes a peak of 3dB at the crossover frequency. I did a quick calculation of the crossover frequency and they are very different for the low and high frequencies. Test reversing the phase of one speaker to pick what sounds best.
 

Thread Starter

andyhof

Joined Jun 22, 2011
13
A resistor parallel with the speaker wastes amplifier power.
An inductor in series with a speaker cuts high frequencies at -6dB per octave which is very gradual. A 1mH inductor with a reactance of 16 ohms at 2.5kHz reduces 2.5kHz by -3dB and reduces the 5kHz peak by -6dB.
Since the crossover network is second-order then it creates a notch filter at the cutoff frequency unless the phase of one speaker is reversed, then it makes a peak of 3dB at the crossover frequency. I did a quick calculation of the crossover frequency and they are very different for the low and high frequencies. Test reversing the phase of one speaker to pick what sounds best.

picked this up and will install in series.
http://www.erseaudio.com/Products/EQCoils18Gauge/EAC33-18-1000
 

Thread Starter

andyhof

Joined Jun 22, 2011
13
Since the crossover network is second-order then it creates a notch filter at the cutoff frequency unless the phase of one speaker is reversed, then it makes a peak of 3dB at the crossover frequency. I did a quick calculation of the crossover frequency and they are very different for the low and high frequencies. Test reversing the phase of one speaker to pick what sounds best.
Now since my existing diving network is 2nd order.. does this new inductor coil in series change any of the phase ?
 
Last edited:
Top