Tinkering with a tubed EQ idea..

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Viitalahde

Well-known member
Joined
Nov 7, 2005
Messages
727
Location
Kuhmoinen, Finland
I'm thinking of getting some EF86's and prototyping the bandpass filter from this schematic: http://www.cgs.synth.net/tube/vcf.html

The sound samples at the Metasonix site are cool and the description suggests the commercial circuits are related to this one.

Now, this circuit has been discussed previously. The moment I saw it, I knew I'd like to test it in a configuration similar to the Sontec EQ circuit: http://img.photobucket.com/albums/v240/virtalahde/MEP250a_schemo.jpg

The bandpass filter gets either straight or inverted signal depending on the EQ gain pot, thus boosting or cutting when summed back in the amplifier.

What do you tube gurus think that would be a worthy thing to experiment as the main amplifier? A tube op-amp is an obvious choice, but I'm kinda concerned of the possible gain/bandwidth limitations..? Maybe it's not an issue if/when I'm not aiming for 12dB @ 20k but..

Anything else coming to your minds? If I ever went and actually built a full-blown EQ around this one, I wouldn't see sense in more than 2 or 3 paralleled bands like this.
 
> A tube op-amp is an obvious choice

No, it's a bad choice.

Opamps come from analog computers. They need to be DC-coupled to compute slowly-varying functions and to hold the output value long enough to read the meter. This requires MAJOR complication and compromise and cost.

You don't need or want DC response. Any opamp-solvable problem can be solved with only inverting opamps (indeed Burr-Brown's early tube opamps only had the "-" input available). An inverting no-DC "opamp" reduces to a simple grounded-cathode tube stage, grid in, coupling cap off the plate output. For low-Z loads you may need a cathode follower, but this is much simpler than a true opamp, yet works the same (probably better) for audio.

Look at the Baxandall tone control. It is now done with a chip opamp, but used to be done with one section of 12AX7, possibly with a cathode follower for fussy use.
 
[quote author="PRR"]No, it's a bad choice.[/quote]

Thanks for clearing that up!

Look at the Baxandall tone control. It is now done with a chip opamp, but used to be done with one section of 12AX7

baxandall.gif


Like this? What's your estimation of the distortion figures on such a circuit?

I also found this circuit from DiyAudio forums: http://img.photobucket.com/albums/v240/virtalahde/active.gif

If I'm not completely lost here, I see a simple op-amp in the first 12ax7 stage?

Instrument amps are of course not the best place to start the search, they're probably designed for opposite guidelines compared to my idea: lots of boost/cut, some audible distortion allowed - I aim for 6-8dB max boost/cut and less enough distortion to not worry about it.

Hmm, I'll check the Klein&Hummel UE-100 schematic, too.. I was going to clone that EQ, but it'd be too monstrous to do and would be more fun to do something of my "own" for a chance. But maybe it has a suitable amp for this.
 
The nice thing in the UE100 is that everything is switched via resistor sets (which comes cheaply with precision), using only one precision-capacitor set for each band. No inductors. Makes exact matching very easy.

This is a pic of what I'm currently working at:

http://www.gyraf.dk/tmp/G-UE100-3U.jpg

Jakob E.
 
i was thinking something like NYD's passive eq with a valve make up stage?

working with the 417a tubes that CJ made me think of this, they are super quiet and clean..
 
The output of the "NYD EQ" is at low impedance. I'd suggest a good stepup transformer (with sufficiently high primary inductance) to match the output to the tube grid, for best noise figure. And you get a little added voltage gain to boot :thumb:

Naturally, to preserve high frequency response, you want to avoid a really high ratio.

With the 250R +250R output pad shown in the PDF, the output of the EQ can be plugged into any mic preamp of your choosing, tube or solid-state.
 
ah ok. what do you think is too high a ratio? 1:10, 1:15?

I'm going to be getting some 1:10 trafos from the beyer order but i don't know their inductance.. i'll have to measure them when they arrive.

sounds like fun.

:thumb:
 
[quote author="gyraf"]The nice thing in the UE100 is that everything is switched via resistor sets (which comes cheaply with precision), using only one precision-capacitor set for each band. No inductors. Makes exact matching very easy.[/quote]

Hi Jakob,
I didn't find power supply data on the UE100 schematics.
Do you (or somebody else) know the voltages of two positive and one negative HV supplies?

Thanks,
Milan
 
Back
Top