There are a number of ways to do this. Using digital controlled analog is perhaps the best solution, such as digital controlled resistors or MDAC's.
But I would question the purpose of such a filter. I assume you are not satisfied with your current xover frequency and want to experiment to find the optimum. And when you have found this frequency, the need for adjustable xover is no longer there.
I would suggest that you buy/borrow a digital xover, find the optimum frequency and modify your current xover for that frequency.
But I would question the purpose of such a filter. I assume you are not satisfied with your current xover frequency and want to experiment to find the optimum. And when you have found this frequency, the need for adjustable xover is no longer there.
I would suggest that you buy/borrow a digital xover, find the optimum frequency and modify your current xover for that frequency.
Hi,
I´d also suggest a digital filter (IIR) for the first shots. If You´ve found a optimum configuration You may build a analog filter after that fashion.
If I had to control 10 pots at the same , I´d think about two ways:
1) use digital Pots, controlled by a microcontroller. But similar to analog pots this is not exact since the pots vary quite alot in end-value.
2) useage of jumpers, DIP-sockets, or DIP-switches to modify the filter resistors. This may be ighly precise, but requires increased effort in complexity, parts number count and cost.
I tend to say that an analog filter only makes sense nowadays if it requires only a few simple filter structures and if it is not implemented using OP-amps.
If those requirements are not fulfilled a digital X-over is probabely superior an any way.
jauu
Calvin
I´d also suggest a digital filter (IIR) for the first shots. If You´ve found a optimum configuration You may build a analog filter after that fashion.
If I had to control 10 pots at the same , I´d think about two ways:
1) use digital Pots, controlled by a microcontroller. But similar to analog pots this is not exact since the pots vary quite alot in end-value.
2) useage of jumpers, DIP-sockets, or DIP-switches to modify the filter resistors. This may be ighly precise, but requires increased effort in complexity, parts number count and cost.
I tend to say that an analog filter only makes sense nowadays if it requires only a few simple filter structures and if it is not implemented using OP-amps.
If those requirements are not fulfilled a digital X-over is probabely superior an any way.
jauu
Calvin
I would say the easiest way would be to use a bunch of digital since they cost about the same as a regular pot any more,sometimes cheaper.
Or you can use something like this circuit to control more than one variable resistance opto-isolators,
Use a photoelectric-FET optocoupler as a linear voltage-controlled potentiometer | EDN
I had many sleepless nights for many many years thinking about this for a particular application similar to yours.
Even though this is not my circuit it is something along the lines that I was thinking about, since at the time digital pots where quite costly and hard to get back then.
I was building VCA's that had better performance than the opamp types using devices like these in the early 80's and late 70's.
jer
Or you can use something like this circuit to control more than one variable resistance opto-isolators,
Use a photoelectric-FET optocoupler as a linear voltage-controlled potentiometer | EDN
I had many sleepless nights for many many years thinking about this for a particular application similar to yours.
Even though this is not my circuit it is something along the lines that I was thinking about, since at the time digital pots where quite costly and hard to get back then.
I was building VCA's that had better performance than the opamp types using devices like these in the early 80's and late 70's.
jer
Last edited:
Some pro audio active crossovers used VCAs (voltage controlled amplifiers) to replace the pots that determined the tuning frequency. You may still be able to get the Analog Devices SSM2164 (quad low-cost VCA). Or there's the better SSM2018 which is still available to buy from Digikey - but it's a single VCA and over $7 each.
I believe it's also possible to use multiplying DAC chips as digital gain controls. Parallel the digital inputs and drive them with dipswitches or thumbwheel switches, so no microcontroller is involved (and there's no possibility of noise from digital activity).
I believe it's also possible to use multiplying DAC chips as digital gain controls. Parallel the digital inputs and drive them with dipswitches or thumbwheel switches, so no microcontroller is involved (and there's no possibility of noise from digital activity).
Last edited:
I've built active crossovers using trimpots for the resistors so they'd be variable - it works fine, it's just irritating to change their values.
I have since moved onto to a self-built DSP solution. The DCX2496 isn't perfect, it is also inexpensive, and I can highly advocate using something like it as the flexibility gained by doing isn't something to sniff at.
I have since moved onto to a self-built DSP solution. The DCX2496 isn't perfect, it is also inexpensive, and I can highly advocate using something like it as the flexibility gained by doing isn't something to sniff at.
thank you so much for your replies...
ok...i try in proteus sim a digital pot solution...
i use 1 AD5206 (6-channel pot) , 1 AD5204 (4-channel pot) and a pic16f88
to control them using SDI interface...
Having an external pot connected on an analog input of pic should do the job..
I have stuck on programming the pic but ok the basic is to find the general idea...
Hmmm also i think that for stereo cross i have to control totally 20 pots...
see attachment
ok...i try in proteus sim a digital pot solution...
i use 1 AD5206 (6-channel pot) , 1 AD5204 (4-channel pot) and a pic16f88
to control them using SDI interface...
Having an external pot connected on an analog input of pic should do the job..
I have stuck on programming the pic but ok the basic is to find the general idea...
Hmmm also i think that for stereo cross i have to control totally 20 pots...
see attachment
How about a 10-gang rotary switch? I did something very similar using a 4-position, 28 pole switch scavenged from an old style computer KVM (Keyboard Video Mouse switchbox). You may be able to find an old KVM or printer switch box with a similar switch inside.
A digital pot works in a very similar way to a standard pot; there is no processing per se. The signal is definitely not converted to digital. The variable resistances are (in general) implemented with FETs, controlled by a digital circuit which is usually controlled by an external microcontroller (some, rarely, have up/down button inputs instead). Because they have active circuitry in the signal path, they will add more noise and distortion than a normal pot. Also, the resistance through the "wiper" is usually some 10's to 100's of Ohm, and there is usually some non-linearity as well. This means that they like to be driven by a low impedance, and like to be followed by a high impedance next stage, in order to reduce distortion. Consider these limitations before choosing to use them.
A digital pot works in a very similar way to a standard pot; there is no processing per se. The signal is definitely not converted to digital. The variable resistances are (in general) implemented with FETs, controlled by a digital circuit which is usually controlled by an external microcontroller (some, rarely, have up/down button inputs instead). Because they have active circuitry in the signal path, they will add more noise and distortion than a normal pot. Also, the resistance through the "wiper" is usually some 10's to 100's of Ohm, and there is usually some non-linearity as well. This means that they like to be driven by a low impedance, and like to be followed by a high impedance next stage, in order to reduce distortion. Consider these limitations before choosing to use them.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.