Can R-2R relay attenuator act as a pot for tone controls
Hello,
Can R-2R relay attenuator act as a normal potentiometer and be used in baxandall like tone control circuit?
Generally I have only seen R-2R relay attenuator being used for volume control where one end is connected to input one end to output and one end to ground.
So can we connect a similar R-2R relay attenuator to replace a potentiometer in baxandall tone control circuit , where all the 3 ends will be connected to other components but none to ground?
So far only able to get an example like below, but here we can have one relay for one db level in tone control. So if we can use R2R relay , we may have lot more levels for few relays and have more granular tone controls instead of steps of 5db as here
thanks and regards
S Sarath
Hello,
Can R-2R relay attenuator act as a normal potentiometer and be used in baxandall like tone control circuit?
Generally I have only seen R-2R relay attenuator being used for volume control where one end is connected to input one end to output and one end to ground.
So can we connect a similar R-2R relay attenuator to replace a potentiometer in baxandall tone control circuit , where all the 3 ends will be connected to other components but none to ground?
So far only able to get an example like below, but here we can have one relay for one db level in tone control. So if we can use R2R relay , we may have lot more levels for few relays and have more granular tone controls instead of steps of 5db as here
Congratulation !
How do the passive tone controller sound ?
Mine not ready , as I am involved now in another urgent project !
I see you only connected the + dB side ...
I must find time for finishing this project .
thanks and regards
S Sarath
Last edited:
Relay controls Baxandall_Tone or Volume-Control
Hello,
I face the same problem during build up the D.Self-preamp (Linear Audio Vol #5). I contacted Jos van Eijndhoven (RelaiXed2 -- DIY balanced pre-amplifier).
His answer: "you posed an interesting question, which I also needed to seriously think about... I am afraid the answer is negative. My relay attenuator has a staged design and thereby has a compex/weird input or output resistance to signal ground. So, the 'signal ground' in my attenuator cannot act as the '3rd pin' in these Baxandall controls."
I think that -in principle- this attenuator behavior is caused by my aim for a logarithmic volume controller, whereas these Baxandall designs would be fine with a linear control."
He mentioned also: "E.g. chips like MCP4251 or TPL0102 could be fine components."
I'm not a professional Builder, but I played a little bit with the tone-control part in LTSpice. For me it seems to crack a big nut.
May be there is a professional somewhere out who can suggest a workarround.
Best regards,
Wopo
Hello,
I face the same problem during build up the D.Self-preamp (Linear Audio Vol #5). I contacted Jos van Eijndhoven (RelaiXed2 -- DIY balanced pre-amplifier).
His answer: "you posed an interesting question, which I also needed to seriously think about... I am afraid the answer is negative. My relay attenuator has a staged design and thereby has a compex/weird input or output resistance to signal ground. So, the 'signal ground' in my attenuator cannot act as the '3rd pin' in these Baxandall controls."
I think that -in principle- this attenuator behavior is caused by my aim for a logarithmic volume controller, whereas these Baxandall designs would be fine with a linear control."
He mentioned also: "E.g. chips like MCP4251 or TPL0102 could be fine components."
I'm not a professional Builder, but I played a little bit with the tone-control part in LTSpice. For me it seems to crack a big nut.
May be there is a professional somewhere out who can suggest a workarround.
Best regards,
Wopo
sarath Sarath, why don't u just use TDA7439 smd one which still in production , if u want u will get smd on mouser India and element14....
for doing the RR version we need to manually match each resisitor... And it would be best to import. The resisitors.
Other option is using 3 pga2311 such that ... You will have 3 different slave select line on spi.. So you can control each pga2311... Which much better ... I m just helping u not doing blunders.... ..
for doing the RR version we need to manually match each resisitor... And it would be best to import. The resisitors.
Other option is using 3 pga2311 such that ... You will have 3 different slave select line on spi.. So you can control each pga2311... Which much better ... I m just helping u not doing blunders.... ..
in the datasheet its recommended to connect the wiper pin to a high impedance may be like a opamp. The same is mentioned with other digital pots.
This requirement is covered well when we use them for volume control where we follow up wiper pin with output buffer. But in tone control wiper pin will be connected to resistor or capacitors and not high impedance like opamp..
I'm pretty sure an R-2R network is linear. You could probably recalculate the network to make it logarithmic but then it wouldn't be 'R-2R'.
G²
I agree, and this is a big pet peeve of mine.
These days, folks in the DIY audio world use the term "R-2R" to describe any sort of ladder with a pile of resistors and (usually) relay switches. If it's a 7 stage "modern R-2R" ladder, then it's built from 14 different resistor values. Not sure why the kids felt the need to corrupt a perfectly self-explanatory term and use it to mean "a ladder with more resistors than I'm comfortable solving using Ohm's law", but that's what happens.
These "modern R-2R ladders" are series connected attenuators that switch so many dB at each stage. By providing 32dB, 16dB, 8dB, and so on stages, one can cook up a pile of fine grained attenuations with only 7 or 8 stages. However, building one is a horrible annoyance - two oddball values are needed for each stage, which will probably have to be built from a couple of series/parallel E192 values, or $40 custom wire wounds.
Several decades ago, the term "R-2R ladder" referred to a ladder that was built with two resistor values: R, and twice that, or 2R. That was why the phrase "R-2R" was used. Seems obvious to me, but apparently people just sling fancy sounding phrases around without thinking much.
Because of these values, a "real E-2R ladder" is a linear ladder, whose attenuation is linear, one part in 2^N for each binary switch code, where N is the size of the ladder in bits. So, half the codes get used for the first 6dB of attenuation, the next half of the remaining codes get used for the next 6dB of attenuation, and so on. Such a ladder can be used for an audio attenuator, but you need a lot of bits to get fine grained attenuation, and some fancy resistors to avoid a number of well known problems, and have to accept coarse resolution at high attenuations.
However annoying the "real R-2R ladder" is, it was the only thing feasible on an IC way back when, since monolithic ladder DAC ICs had to be built with the resistors available, and there was no way to control the absolute value of a resistor on an IC - it could vary by 20-30% easily.
If you wanted to build an 8 bit ladder, which requires resistors good to ~0.2% you could use one resistor value for R and the same resistor, two in series, to make 2R. Because the ratio of a pile of identical resistors on an IC is basically as good as your lithography, it wasn't hard to make an 8 or 10 bit ladder that's 1/2 LSB accurate, even back in the stone age, because the ladder accuracy has nothing to do with the absolute value of the resistors used, only their ratio. Making a pile of the same resistor value, whatever that was, is not that hard on an IC. The DAC-08 is a great example - cheap, reliable, and remarkable for 1978 or whenever it was designed.
I wish kids would stop using "R-2R" to mean "a large pile of incomprehensible resistors", but they do. Not sure why either, but audio has always been filled with so many cool sounding terms that get applied inappropriately, so it's nothing new really. "Phase" is another mostly misapplied phrase, but I won't start that rant.
OK, I've settled down now… 
I'd like to suggest a "multiplying DAC" to do a linear ladder for a tone control. The annoyance is finding one that can handle high supply and signal voltages, and you have to deal with the built-in transistor switches and their potential distortion, but if they're carefully applied, they can sometimes work pretty well for this sort of programmable filter application.
In general, digital pots (DPOTs) are basically the same thing as an MDAC, with the addition of a handy EEPROM, serial control, and a lot of nice features, but they're more annoying to use, simply because they are usually low voltage parts.
One could spin up their own switches and resistor network, but that's a lot of work. IIRC, there have been some programmable filter designs done using analog VCAs, which could be adapted to a modern programmable attenuator without thinking too much. The advantage is that some of the programmable attenuators are actually quite clean compared to MDACs or DPOTs, and very easy to apply.
So, while a traditional circuit might use a linear divider, this could be replaced with a programmable attenuator and a slightly different circuit. Sorry to be vague, but search for some VCA controlled filter designs to get a starting point. Perhaps some old SSM application notes?
I'd like to suggest a "multiplying DAC" to do a linear ladder for a tone control. The annoyance is finding one that can handle high supply and signal voltages, and you have to deal with the built-in transistor switches and their potential distortion, but if they're carefully applied, they can sometimes work pretty well for this sort of programmable filter application.
In general, digital pots (DPOTs) are basically the same thing as an MDAC, with the addition of a handy EEPROM, serial control, and a lot of nice features, but they're more annoying to use, simply because they are usually low voltage parts.
One could spin up their own switches and resistor network, but that's a lot of work. IIRC, there have been some programmable filter designs done using analog VCAs, which could be adapted to a modern programmable attenuator without thinking too much. The advantage is that some of the programmable attenuators are actually quite clean compared to MDACs or DPOTs, and very easy to apply.
So, while a traditional circuit might use a linear divider, this could be replaced with a programmable attenuator and a slightly different circuit. Sorry to be vague, but search for some VCA controlled filter designs to get a starting point. Perhaps some old SSM application notes?
Start here
Logarithmic Attenuator Calculator
8 stages with 1% resistors (E96) can give LARGE range and accuracy
G²
Last edited:
Vout = Vref* bitnumber/(2^n bits)
For a tone control you don't need large range, Morgan says that the original Baxandall had +/-20dB which was way too much.
If you make the tone control with discretes you can calculate the values which would give a more reasonable range.
BTW, if you are switching into a reasonably high impedance next stage, the distortion of the Analog Devices switches is next to negligible.
Also -- a bag of Yaego 1% film resistors from Digikey costs about as much as a beer, and you will be able to match a large percent to better than 1%.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.