Designing and building a mixer circuit.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
After I downloaded a few equilizer schematics, I came up with the attached circuit which is still in the beginning stages of optimisation. The emulated inductors' quality factor is too high. This resulted in a frequency response with narrow and very deep valleys which is consistent with theory. With much lower quality factors I got a maximum attentuation difference of 12dB. 'Difference' refers to the dB values at the maximum and minimum setting of a frequency channel. The attached schematic should have the following channel frequencies:

32Hz, 64Hz, 128Hz, 256Hz, 500Hz, 1kHz, 2kHz, 4kHz, 8kHz, 16kHz.

I am after a bigger attenuation, but it seems, this requires the use of a high resistance in the output of the input buffer amplifier stage. High resistances on signal paths do not bode well for a good signal to noise ratio. Therefore, I am trying to avoid this.
 

Attachments

  • corrected_equaliser.asc
    13.3 KB · Views: 36
Administrator
Joined 2007
Paid Member
Thats a very very nice simulation :) and a curious choice of opamp...

I tend to find that pure noise isn't particularly objectionable in practice, now obviously that doesn't mean it should be hissing at twenty paces but on the other hand are you chasing specs that might have little audible impact apart from being a bigger or smaller number on paper.
 
I chose that opamp to facilitate simulations until I get a circuit that simulates well. By further reducing the Q factor the frequency resonse has been flattened further but it still looks undulant.

I have to further tweak the resonant frequencies to get make them closer to what I want.

The new circuit is attached.
 

Attachments

  • corrected_equaliser.asc
    11.1 KB · Views: 34
The attached file is the latest version which should have a flatter frequency response. The response's shape is like a 'flat' hill top, although there is still room for improvement.

Logic tells me that the middle channels have a lower pass impedance because both side channels contribute to them, whereas this is not true for extremity channels. This means, adding extra resistors to counter this effect should result in a flatter response.
 

Attachments

  • corrected_equaliser.asc
    12.4 KB · Views: 39
Administrator
Joined 2007
Paid Member
That looks a lot better :)

Just a thought, Doug Self did a brilliant (imo) tone control many years back with variable turnover frequencies and he reckoned it was of more use than a simple graphic equaliser type design.

This was a simulation I did of it a long time ago, and I also built the real thing. Its great.

The sim has two versions on the same page, one a low impedance version (lol, noise ;)) but it really is extremely quiet in standard form.

You'll figure it out if you run it. The audio output is at the first opamp and as it stands its inverting of phase.
 

Attachments

  • Annotation 2019-04-22 093119.jpg
    Annotation 2019-04-22 093119.jpg
    49.6 KB · Views: 81
  • Tone Controls Low Impedance Version 2.asc
    15.8 KB · Views: 40
While research reading about gyrators, I encountered the concept of a gyrator that implements a variable inductance. Following that, I read that these can make variable frequency resonators (filters) that are more useful than graphic equalizers.

Since I am also interested in electronics, I would like to first understand how the circuits you posted work. I am assuming they use the previously mentioned concept of a variable resonator. The fact that the posted circuit is by Douglas Self, makes building it much more motivating.

This will greatly reduce the mixer's complexity.

Finally, I presume, there is a ready designed PCB for this. There are two circuits: which one would you suggest?
 
Administrator
Joined 2007
Paid Member
Not exactly, it was part of a complete preamp design (for which boards were/are available). I hand made a PCB many years ago now but I'm not certain I have the artwork as I used Diptrace with an old Vista PC. I'm not going to promise I can come up with anything there, it may be lost in the mists of time.

Anyone built/listened to D Selfs precision preamp.

The low noise version simply had the reistors scaled down and the caps scaled up. Marginally lower noise in theory but a tougher load for the opamps. I'd go with the original I think.
 
This equaliser looks good.

Searching for a PDF about equaliser design I found a paper with a generic equaliser layout. Using that as a guide, I redrawn the circuit with some modifications. The result is a very flat equaliser with a promising response. Setting Rn to 1 result in a closed frequency channel and 99999 in a fully open channel.
I am attaching the circuit.
 

Attachments

  • good-equaliser.asc
    12.6 KB · Views: 38
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.