I am currently building a 3 band phase linear crossover filter summed to stereo.
The kit that I bought, the outputs of the three 50k logarithmic pots are 0dB when the pots are all the way up. I want to modify the circuit so that at the halfway point (12 oclock) on the pots, the output is 0dB, all the way to the left is -infinity, and all the way to the right is +6 or +9 dB.
Please help? Not sure how to proceed.
Here is the circuit schematic for the 3 band crossover. The pots are located just before the outputs. My hunch is I need to add additional non-inverting amps with specific gains to each output?
http://analogmetric.com/download/CO3 Three Band Phase-Linear Crossover Filter schematic.pdf
The kit that I bought, the outputs of the three 50k logarithmic pots are 0dB when the pots are all the way up. I want to modify the circuit so that at the halfway point (12 oclock) on the pots, the output is 0dB, all the way to the left is -infinity, and all the way to the right is +6 or +9 dB.
Please help? Not sure how to proceed.
Here is the circuit schematic for the 3 band crossover. The pots are located just before the outputs. My hunch is I need to add additional non-inverting amps with specific gains to each output?
http://analogmetric.com/download/CO3 Three Band Phase-Linear Crossover Filter schematic.pdf
Last edited:
You can add a non-inverting gain stage at each output. Use 18k nfb and 10k to ground.
For +6dB, use 10k nfb and 10k to ground.
Adjust the position of the knob on the shaft so the pointers are straight up when set for 0dB gain.
This is unlikely to be the center of rotation, though. If necessary, you can adjust that with a resistor
to ground connected to each of the wipers. This assumes that the outputs at the center of rotation
are too high, which is likely to be the case with log pots.
Last edited:
Either kind, but log pots will work better for this particular purpose.
A setting of 0dB in this case would be above the center of rotation with a log pot,
or else would be below the center of rotation with a linear pot.
A linear pot can be "law faked" by adding a resistor from the wiper to ground, though.
Last edited:
I modelled the circuit in TINA using their 5534 models and I am getting proper tracking (with a bit of a problem with the low mids - maybe need to tweak the components a bit that came with the package?).
You can see the hump in the summing here:
Attachments
You can just as well boost right after the pot. There's an under-used opamp there. It affects all three outputs the same, which is probably what you want.
However... this is the SAME thing as your main volume control. Why do you need another?
To begin to do this we *need* to know the actual pot taper. "log" is not a definite thing (no ordinary pot is truly logarithmic). "Audio taper" has been applied to several different tapers. Commonly specified as percent resistance at mid-turn. You can probably read this directly with a digital ohm meter in-circuit. (A big old analog Simpson would open-up the junctions in the chips and give a wrong reading in-circuit.)
If you are content with "+6dB" max and can replace the pot, a good Linear pot plus a 2X gain will do -inf, 0dB, +6dB.
If you are stuck with 10% "Audio taper", the circuit tends to want to be +20dB at max. And the hiss will be brought up.
It is possible to "load" the pot to get nearer what you want at three places (ends and middle). But that's much matchbook work and starts with knowing the actual taper of the pot you have or can get.
However... this is the SAME thing as your main volume control. Why do you need another?
To begin to do this we *need* to know the actual pot taper. "log" is not a definite thing (no ordinary pot is truly logarithmic). "Audio taper" has been applied to several different tapers. Commonly specified as percent resistance at mid-turn. You can probably read this directly with a digital ohm meter in-circuit. (A big old analog Simpson would open-up the junctions in the chips and give a wrong reading in-circuit.)
If you are content with "+6dB" max and can replace the pot, a good Linear pot plus a 2X gain will do -inf, 0dB, +6dB.
If you are stuck with 10% "Audio taper", the circuit tends to want to be +20dB at max. And the hiss will be brought up.
It is possible to "load" the pot to get nearer what you want at three places (ends and middle). But that's much matchbook work and starts with knowing the actual taper of the pot you have or can get.
Attachments
Last edited:
To answer exactly what you asked for:
1) ditch the kit Logarithmic pots which should be about -20dB at 12 o clock (or on "5" in a 0-10 scale) and replace them with Lin curve ones, which are guaranteed -6dB on "5" (halfway rotation).
2) mod the unity gain input stage by cutting the track between output and -In , adding a 10k resistor between now separated pads, and another 10k resistor from -In to Ground.
Now full crossover still has 0dB gain as before, if pots set to "5", and range goes from +6dB to zero output.
Enjoy.
1) ditch the kit Logarithmic pots which should be about -20dB at 12 o clock (or on "5" in a 0-10 scale) and replace them with Lin curve ones, which are guaranteed -6dB on "5" (halfway rotation).
2) mod the unity gain input stage by cutting the track between output and -In , adding a 10k resistor between now separated pads, and another 10k resistor from -In to Ground.
Now full crossover still has 0dB gain as before, if pots set to "5", and range goes from +6dB to zero output.
Enjoy.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.