I am looking to upgrade the volume control on my current Buffalo II DAC. Presently I am using the Volumite board for digital volume control. I recently did a listening test and found that I prefer a simple resistor voltage divider for low volume listening as it seemed more realistic or natural. It was subtle, but noticeable.
As further background information, I am building a circuit board to provide an op-amp based high/low filters for subwoofer output and line switching for home theatre bypass.
I am seeking an analog volume control now and there seem to be three different options:
1. Stepped attenuator (TPA Joshua Tree). This has the benefit of being something I could easily remote control. It requires a balanced to single ended converter and an output buffer before going to the filter board. There are a number of options here as well besides the one named above.
2. Transformer Volume Control (Sowter 9335). This seems to eliminate the need for both the bal-se conversion and output buffer. I have been reading very good things about transformers lately and am interested in experimenting with them. It is much more expensive which would suck if I am not happy with it. It also requires a relatively expensive switch to work.
3. Something like a Light-Dependant Resistor. This sounds fairly complicated and I am just including it for completeness. I would need a very good reason to try this. Also requires the bal-se and buffer.
Is there anyone here who has tried the above and can comment on the sonic qualities of the different methods and their preference? Is there anything I missed that I should consider?
Also, I am using a Legato 3.1 output stage if that makes a difference. Amplification is by Emotiva and I will be adjusting the Legato for ~0 dBu output to match the amps sensitivity.
Forgive me if this has been asked before… if so, good links would be appreciated. Thanks for your help.
As further background information, I am building a circuit board to provide an op-amp based high/low filters for subwoofer output and line switching for home theatre bypass.
I am seeking an analog volume control now and there seem to be three different options:
1. Stepped attenuator (TPA Joshua Tree). This has the benefit of being something I could easily remote control. It requires a balanced to single ended converter and an output buffer before going to the filter board. There are a number of options here as well besides the one named above.
2. Transformer Volume Control (Sowter 9335). This seems to eliminate the need for both the bal-se conversion and output buffer. I have been reading very good things about transformers lately and am interested in experimenting with them. It is much more expensive which would suck if I am not happy with it. It also requires a relatively expensive switch to work.
3. Something like a Light-Dependant Resistor. This sounds fairly complicated and I am just including it for completeness. I would need a very good reason to try this. Also requires the bal-se and buffer.
Is there anyone here who has tried the above and can comment on the sonic qualities of the different methods and their preference? Is there anything I missed that I should consider?
Also, I am using a Legato 3.1 output stage if that makes a difference. Amplification is by Emotiva and I will be adjusting the Legato for ~0 dBu output to match the amps sensitivity.
Forgive me if this has been asked before… if so, good links would be appreciated. Thanks for your help.
Hi,
All of these have advantages and disadvantages. All can be good choices. All can be made to operate balanced.
Advantage - dirt cheap to make. Disadvantage you always have a ton of relays stringed up and loads of resistors at any sensible level. This maximises the audibility of any possible problems created by them. In many ways a really good potentiometer can sound better. Then again, Conrad Johnson uses them in very highly reviewed line stages, so they cannot be THAT bad.
Still, a better option would be to just use a single fixed series resistor and to use as many relays as you will want steps and suitable shunt resistors (which gets a bit difficult at 71 1dB Steps, but managable at for example 32 2dB steps). This is like the old Audio Synthesis "Passion" Passive Pre.
More complex matrixes may be used to trade off number of resistors and relays involved for a given number of steps.
Sitches can be made via relays or motorised rotary switches to be remote controllable. I personally like TVC's very well, but not all transformers are made equal. Some have measured results that better most (if not all) active and pure resistor solutions, others measure much worse.
Cost is invariably very high compared to all other options. If cost is no objective and I can have the best available it would be my choice.
These work quite well and are nowhere near as daunting to implemented (type "Lightspeed" into the search here). They have a big advantage in that they lack any mechanical contacts in the audio circuits and can be used with a cheap generic (and easily remoted) pot as control.
On the downside, their resistance is heavily voltage dependent and thus makes them problematic when handling AC signals. Normally very high levels distortion are observed. There are possible circuit tricks to cancel much of this, but they are getting complex. Another issue is that LED's and resistors (or whole LED/LDR combo's) need to be matched to bet a good level of balance. across the adjustment range.
Some people have claimed sonic miracles for LDR based Attenuators, the ones I have encountered, both DIY types and those build into commercial gear had more drawbacks than I percieved sonic advantages...
Some electronic Chip attenuators (not the common ones though) can perform also very well if implemented right, certainly up to the level of a "Joshua Tree" implemented with generic resistors and relay
Ciao Ts.
All of these have advantages and disadvantages. All can be good choices. All can be made to operate balanced.
Advantage - dirt cheap to make. Disadvantage you always have a ton of relays stringed up and loads of resistors at any sensible level. This maximises the audibility of any possible problems created by them. In many ways a really good potentiometer can sound better. Then again, Conrad Johnson uses them in very highly reviewed line stages, so they cannot be THAT bad.
Still, a better option would be to just use a single fixed series resistor and to use as many relays as you will want steps and suitable shunt resistors (which gets a bit difficult at 71 1dB Steps, but managable at for example 32 2dB steps). This is like the old Audio Synthesis "Passion" Passive Pre.
More complex matrixes may be used to trade off number of resistors and relays involved for a given number of steps.
Sitches can be made via relays or motorised rotary switches to be remote controllable. I personally like TVC's very well, but not all transformers are made equal. Some have measured results that better most (if not all) active and pure resistor solutions, others measure much worse.
Cost is invariably very high compared to all other options. If cost is no objective and I can have the best available it would be my choice.
These work quite well and are nowhere near as daunting to implemented (type "Lightspeed" into the search here). They have a big advantage in that they lack any mechanical contacts in the audio circuits and can be used with a cheap generic (and easily remoted) pot as control.
On the downside, their resistance is heavily voltage dependent and thus makes them problematic when handling AC signals. Normally very high levels distortion are observed. There are possible circuit tricks to cancel much of this, but they are getting complex. Another issue is that LED's and resistors (or whole LED/LDR combo's) need to be matched to bet a good level of balance. across the adjustment range.
Some people have claimed sonic miracles for LDR based Attenuators, the ones I have encountered, both DIY types and those build into commercial gear had more drawbacks than I percieved sonic advantages...
Some electronic Chip attenuators (not the common ones though) can perform also very well if implemented right, certainly up to the level of a "Joshua Tree" implemented with generic resistors and relay
Ciao Ts.
Thanks for the info. I appreciate it. Do you have any experience with the Sowter or further recommendations for TVC? I haven't been able to find to much info on them.
Hi,
I had earlier on the chance to compare Sowter (I think a 9335) and S&B TX-102. In those comparisons I preferred S&B. These are now no longer available for DIY, Silk in Thailans have been making comparable items but I do not know precisely how they compare, as I was never able to put them side by side, so no idea how they stack up.
Ciao T
I had earlier on the chance to compare Sowter (I think a 9335) and S&B TX-102. In those comparisons I preferred S&B. These are now no longer available for DIY, Silk in Thailans have been making comparable items but I do not know precisely how they compare, as I was never able to put them side by side, so no idea how they stack up.
Ciao T
Hi ThorstenL, I found a really old post by you describing some history to the transformer volume controls. Pretty interesting read. It is too bad the S&B TX-102 are out of business. I am still leaning towards a TVC, but the Sowters have been compared to the S&B a few times and it seems the S&B were always ahead.
Any other thoughts on the TVC? Anybody try the Silk?
Any other thoughts on the TVC? Anybody try the Silk?
Hi,
I have been exposed to Silks quite regularly. They seem fine (do what I expect TVC's to do) but I never had the chance to make a 1:1 comparison to S&B, so I cannot comment on that score.
Ciao T
I have been exposed to Silks quite regularly. They seem fine (do what I expect TVC's to do) but I never had the chance to make a 1:1 comparison to S&B, so I cannot comment on that score.
Ciao T
Hi
I use a TVC for years now and I'm happy with it. For me it's the best volume control you can use but you have to take care about some rules:
Ratio of output impedance of your DAC regarding input impedance of your amplifier (for example 600 ohm into 10Kohm). Choose an TVC which owns a high inductivity in the bass frequencies that the dynamics will be preserved.
Best is to use an autoformer which is a transformer using only one winding of a transformer and you get a better signal transformation because of the absence of the magnetic coupling from primary to secondary.
To answer your question ahead I use the Tribute Volume Control. Its inductivity is 150H and it works well to 5Vrms.
Please have a look on their website and I attached an image from the unit because it's unice that the switch is included:
Tribute audio transformers
I use a TVC for years now and I'm happy with it. For me it's the best volume control you can use but you have to take care about some rules:
Ratio of output impedance of your DAC regarding input impedance of your amplifier (for example 600 ohm into 10Kohm). Choose an TVC which owns a high inductivity in the bass frequencies that the dynamics will be preserved.
Best is to use an autoformer which is a transformer using only one winding of a transformer and you get a better signal transformation because of the absence of the magnetic coupling from primary to secondary.
To answer your question ahead I use the Tribute Volume Control. Its inductivity is 150H and it works well to 5Vrms.
Please have a look on their website and I attached an image from the unit because it's unice that the switch is included:
Tribute audio transformers
Attachments
Last edited:
Hi,
Well, the DAC Chip in your Buffalo has the ability to output 6V Peak-Peak from a very low source impedance (50 Ohm IIRC) and can drive a dead short, so it is perfect to connect directly to the TVC and to cut any added stages out, they cannot improve the sound quality and basically have no real job any longer. In this case you cannot use a AVC but a real TVC is needed, yet the results are quite exceptional.
I have not measured the Silk's primary inductance, but I can do so, they claim over 1KH (> 1,000H) which if it can be confirmed puts them well past good and bad for most applications, IIRC the S&B ones where 800H...
As a rule, the higher your source impedance the greater the potential mismatch issues. I found at least with S&B several KOhm source impedance where no real limitation.
Ciao T
Well, the DAC Chip in your Buffalo has the ability to output 6V Peak-Peak from a very low source impedance (50 Ohm IIRC) and can drive a dead short, so it is perfect to connect directly to the TVC and to cut any added stages out, they cannot improve the sound quality and basically have no real job any longer. In this case you cannot use a AVC but a real TVC is needed, yet the results are quite exceptional.
I have not measured the Silk's primary inductance, but I can do so, they claim over 1KH (> 1,000H) which if it can be confirmed puts them well past good and bad for most applications, IIRC the S&B ones where 800H...
As a rule, the higher your source impedance the greater the potential mismatch issues. I found at least with S&B several KOhm source impedance where no real limitation.
Ciao T
The Silk TVC has an interesting selection of taps. As I a lot of my music listening at low volumes the lowest end of the scale is important. I was considering using the Legato with an I/V resistor set to produce low enough voltage so that I could achieve the right output for night listening. This is one area that made me consider a stepped attenuator I this first place. Given the amount of attenuation I use, I am already concerned about the reduced range available on a TVC compared to a resistive attenuator.
Also, as I am considering a balanced to single ended conversion a strength of the TVC an autotransformer would have this going against it. It would have to be really good to overcome this.
Also, as I am considering a balanced to single ended conversion a strength of the TVC an autotransformer would have this going against it. It would have to be really good to overcome this.
Hi,
A lot said here lacks any reference...
Let's for example take a Speaker with 87dB/2.83V/1m sensitivity and an Amplifier with 32dB Gain and 200W RMS/8 Ohm maximum power... This produces rated power with 1V.
For 3m listening distance and the normal levels of crest factor this means the average (not peak) SPL will be around 93dB and need 6dB Attenuation.
With 48dB attenuation the SPL will be around 51dB... This is pretty quiet.
You can try by rigging up a quick resistor attenuator with 46dB Attenuation (10K + 51R) and see if this is quiet enough.
Ciao T
A lot said here lacks any reference...
Let's for example take a Speaker with 87dB/2.83V/1m sensitivity and an Amplifier with 32dB Gain and 200W RMS/8 Ohm maximum power... This produces rated power with 1V.
For 3m listening distance and the normal levels of crest factor this means the average (not peak) SPL will be around 93dB and need 6dB Attenuation.
With 48dB attenuation the SPL will be around 51dB... This is pretty quiet.
You can try by rigging up a quick resistor attenuator with 46dB Attenuation (10K + 51R) and see if this is quiet enough.
Ciao T
Looks good for an Inductive Volume Control
And what's the inductance below 1Khz? Interesting is the inductivity at, for example, 20Hz.
Tribute measure, and specifies the inductance at low frequency - see my posting above.
Agree here ...
Joao,
Autofrmer Volume Control, Inductive Volume Control, same thing, different sticker...
Well, even without measuring I can tell that the measured inductance at low frequencies is greater than at 1KHz. BTW, me writing 1kH is 1 KiloHenry, nothing to do with 1KHz.
Yes, most people do, you get bigger numbers that way, but you also understand better how things will behave at LF.
Ciao T
Autofrmer Volume Control, Inductive Volume Control, same thing, different sticker...
Well, even without measuring I can tell that the measured inductance at low frequencies is greater than at 1KHz. BTW, me writing 1kH is 1 KiloHenry, nothing to do with 1KHz.
Yes, most people do, you get bigger numbers that way, but you also understand better how things will behave at LF.
Ciao T
Thorsten
I had a look at S&B website, but no information regarding the inductance of their TVC ...
Here you can find the SILK data:
SACThailand
As well, no comment about the frequency they measure the inductance of > 1KHenry?
Can you provide your measurements of the S&B?
Ate proxima
Joao
I had a look at S&B website, but no information regarding the inductance of their TVC ...
Here you can find the SILK data:
SACThailand
As well, no comment about the frequency they measure the inductance of > 1KHenry?
Can you provide your measurements of the S&B?
Ate proxima
Joao
Shouldn't the inductance be constant throughout the frequency range? Variable inductance would make a non-linear device, which I would think would be detrimental to the sound. The impedance will change, but there should be very little change in the inductance.
Inductance = L (Henries)
Impedance (reactance) = 2 * PI * f * L (Ohms)
Am I missing something here?
Inductance = L (Henries)
Impedance (reactance) = 2 * PI * f * L (Ohms)
Am I missing something here?
You gave the answer yourself in the formula. F = frequency is the variable so the impedance increases with the increased frequency (capacitors are contrary)...That's why I look for the inductance at low frequency.
not to start an arguement, but the inductance by definition is measured in Henries, defined by the symbol L and independant of frequency. If the actual inductance is 1100 H, then the impedance (X sub L) at 20 Hz is 138 kOhms. At 1 kHz, the impedance would be 6.9 MOhms.
Either way, I do not think I can afford a two transformer solution (one for bal-se and one for AVC/TVC), so the Tribute is out of the question.
At the moment, I am going between a Silk TVC and a stepped attenuator followed by a symmetric B1 buffer. I haven't decided yet.
Either way, I do not think I can afford a two transformer solution (one for bal-se and one for AVC/TVC), so the Tribute is out of the question.
At the moment, I am going between a Silk TVC and a stepped attenuator followed by a symmetric B1 buffer. I haven't decided yet.
Hi,
There used to be an appliation note for the TX-102 that I wrote, it listed that...
I think Silk would have measured at 50Hz if they used a standard LCR analyser.
As for the iductance not being constant wih frequency, it is a feature of the fact that the transformer is an iron or nickel cored choke not air cored. This does necesarily cause distortions, so it rarely receives coverage outside works specially covering measuring transformers.
Ciao T
There used to be an appliation note for the TX-102 that I wrote, it listed that...
I think Silk would have measured at 50Hz if they used a standard LCR analyser.
As for the iductance not being constant wih frequency, it is a feature of the fact that the transformer is an iron or nickel cored choke not air cored. This does necesarily cause distortions, so it rarely receives coverage outside works specially covering measuring transformers.
Ciao T
Sorry my post didn't make too much sense. Sometimes typing on the iPhone can make it hard to go back and proof what was written.
I worked out the same figures as you. I actually did a test with 46 dB of attenuation when testing some of the fixed resistors combo's. When I listen to music at night, that is typical of the level I would listen at. My wife has sensitive ears I guess... plus our house/neighborhood is quiet.
Anyways, I find it odd that Tribute doesn't have any dealers, nor does it seem like they have a way to sell online.
How do you get these things if you did want one? Anyone know how much?
Thorsten,
sorry I could not find these application note, is it possible you provide it to me? Thanks.
Eeven if Silk measures with standard equipment I am sceptical regarding the specs ...
Correct, Tribute has no dealers you have to send an E-Mail.
Hello,
anyone have tested this?
http://silvercore.de/index.php?grundlagen-1
At the moment I'm using LDR but I's not so easy.
Ciao
Guglielmo
anyone have tested this?
http://silvercore.de/index.php?grundlagen-1
At the moment I'm using LDR but I's not so easy.
Ciao
Guglielmo
- Status
- Not open for further replies.