passive 30Hz-20kHz Bandpass filter for tube preamp?

[erer4]

Hi,
I'm working on designing a bandpass filter for a tube microphone preamp. About 30Hz to 20kHz

Tried various calculators, ltSpice, etc... still not sure I'm doing it right.

Amp has 200ohm input impedance, 50ohm-500ohm output impedance. I'd like 12db/octave roll-off. Butterworth looks like the best, IMO.

These look right? Which of the two would be the best choice?

http://www.wa4dsy.net/cgi-bin/lc_fil...units=HZ&Z=200

Thanks!

[stratus46]

Originally Posted by erer4

Hi,
I'm working on designing a bandpass filter for a tube microphone preamp. About 30Hz to 20kHz

Tried various calculators, ltSpice, etc... still not sure I'm doing it right.

Amp has 200ohm input impedance, 50ohm-500ohm output impedance. I'd like 12db/octave roll-off. Butterworth looks like the best, IMO.

These look right? Which of the two would be the best choice?

http://www.wa4dsy.net/cgi-bin/lc_fil...units=HZ&Z=200

Thanks!

Just my opinion. Why a filter? And why 30Hz and 20KHz? The analog guys want north of 20KHz and will likely tell you the filter will damage the audio. Perhaps they're right. The pipe organ guys will wonder why you'd throw away the 32' stop with a 30 Hz filter. My personal mic preamp is flat to below 5 Hz though the way low end needs large format condensers like Neumanns or AKG to even consider that low. Of course that opens you up to air handling noise.

Butterworth filters are a nice compromise between passband flatness and phase shifts but Good Grief. 6 poles? That's 36dB/octave but you said 12 dB so 2 poles. And inductors? You're not worried about hum pickup with those coils? Have you considered an active filter (though it may not be happy with hollow state)? It's s snap with opamps AFTER the initial gain block. Are you implying you'd hang that filter at the input of the preamp so a long cable drives it and a transformer input amp is the load? I'd be plenty concerned about RFI if you try to use it near a broadcast station.

General rule of thumb on mic preamps. Try to keep the input network simple as there are a LOT of unknowns when you get different cables and mics involved.

G²

[DF96]

Keep it simple at the input. It is too easy to lose some signal and pick up some noise/interference. At most have 6dB/octave before the preamp, then do whatever you want afterwards. Remember that real inductors have significant series resistance so this needs to be taken into account when designing filters. By comparison, capacitors are near-enough perfect.

Unless you intend using the microphone right next to a radio transmitter you don't need a sharp HF cutoff immediately next to the audio band.

[erer4]

Thanks for the responses guys.

RC will work fine. Now that I look at it, that seems the be the better choice.

I was planning to patch this filter after the preamp. This pre was usually followed by a 30Hz-20kHz passive filter, and worked fantastically. I just want to see how it sounds with the right filter, and it's no big loss if I don't like it.

Impedance compatability was my main concern, but most simulators dont address the issue...

[speakerfritz]

what on line calculator are you using.....those values do not look right at all.

[stratus46]

Originally Posted by DF96

Keep it simple at the input. It is too easy to lose some signal and pick up some noise/interference. At most have 6dB/octave before the preamp, then do whatever you want afterwards. Remember that real inductors have significant series resistance so this needs to be taken into account when designing filters. By comparison, capacitors are near-enough perfect.

Unless you intend using the microphone right next to a radio transmitter you don't need a sharp HF cutoff immediately next to the audio band.

A real problem with RF is that you may 'detect' (demodulate) it due to overloading an input section with the RF. Once it's in there, NO filter will remove it as it's now part of the audio. I've seen it happen with AM, FM and near airports, air traffic radios. Fortunately the RF is so much higher frequency than any audio you're interested in that a 500 KHz filter at the input can be very effective for the RF and transparent to the audio.

G²