I have conductive plastic (TKD) potentiometer in my daily used 801 tube preamp.
I haven't touched it in weeks.
I haven't touched it in weeks.
No volume control at all. I found that in my system, once the right volume is set, I don't need to change it when playing different disks. That make me think that I can live without VC.
Beam deflection tube. These work by partitioning electron beam between a pair of anodes. Volume is changed by biasing the BDT. The circuit also makes a perfect phase splitter, but can be used in SE mode. It is a nice volume control.
TVC is the best in my opinion. Lowest noise and distortion of all. Speaking of square waves, properly made TVC may have self-resonance above 100 kHz and handle square waves better than best output transformers.
Beam deflection tube. These work by partitioning electron beam between a pair of anodes. Volume is changed by biasing the BDT. The circuit also makes a perfect phase splitter, but can be used in SE mode. It is a nice volume control.
TVC is the best in my opinion. Lowest noise and distortion of all. Speaking of square waves, properly made TVC may have self-resonance above 100 kHz and handle square waves better than best output transformers.
It was posted by John Swenson on Audio Asylum in 2008. Unfortunately, the link to the schematic no longer works. But if you look at 6AR8 data sheet, a schematic therein gives an idea. The volume is controlled by variable cathode resistor. Input signal is applied to control grid, and differential outputs are taken from the two anodes. At full tilt, the gain is unity; it becomes zero when the current is shut off by high enough negative bias voltage.
Transformers "need" a low impedance on one side or the other to give wide-band response. In this context, conventionally the low Z drives the volume control.
OK, well clearly there's no real interest in doing it electronically. I've used volume control chips since 1980. Honestly, they track perfectly, they have lots of pricision steps, follow a perfect "log" taper, and so on. Unless you want quick and easy, I don't know why you'd use a pot. Of course, they're talking about transformers and autoformers...Yikes, what a horrible solution. And attentuators...always the quest for more steps. It's like audio engineering in the 1950s.
How about something in between? Like a single linear taper pot driving a pair of VCAs? Tracking is perfect, the taper is either a perfect log or whatever you like, and the pot part of it is just a single linear pot.
http://www.thatcorp.com/datashts/THAT_2162_Datasheet.pdf
The ever returning challenge to many of us fanatic DIYers.. Recently a guy I know wanted to try out stepped attentuators. Again the trap of 24 steps which practically never leads to satisfaction. Khozmo produces 48 step versions also in series and ladder type. Not too expensive either. Avoid the shunt type as it has variable impedance to the source.
Electronic? Also OK but greatly depending on the IC. Distortion when buffers are omitted, clicks when changing volume etc. Many sound less optimal compared to switched resistors/potentiometers and all the electronic volume controls I built were not used. Muses volume control seems a step ahead. Sometimes the thought comes up that the remote control possibility ways heavier than the best sound quality.....
It also seems the relay types offer a lot of advantages and best of both worlds qualities but ... I never got myself into building one. Used the RK27 and Khozmo to satisfaction 🙂 and then ... there was TKD. They were praised and seemed a quality in-between solution. Not the guru endorsing type but Bruno Putzeys found some interesting things with TKD potentiometers which is why I avoided them after reading this many years ago. Now the type is unknown but the information is shocking considering the average prices and supposedly good reputation. The consumers that buy and only listen seem to like them. Bruno did not. From his blog http://www.brunoputzeys.be/r4.php?item=25
Also out are reed relays. Those wonderful red Coto's are the choice for reliable switching at low currents and voltages, but at 30mA (and again at 20kHz) I found about 7uV of 3rd harmonic across the contact, which as expected came down a bit as the coil voltage was increased.
The irony (pardon the pun) is that I'm discovering all this whilst building a relay-switched gain control to replace an esoteric audiophile Japanese potentiometer that was plainly distorting. Mentioning this to the manufacturer (TKD) elicited exactly no response at all. Mind, you can get virtually zero distortion from even a cheap linear pot if you use it as the feedback divider in an inverting amplifier. Clearly doing the same with a logarithmic pot is a waste of time. Doing so with an expensive logarithmic pot is a great way of discovering just how many relays the same amount of moolah can buy.
Sunday 23 September 2012
So, right now I can use a quality relay controlled attenuator (preferably a 10 kOhm version) myself but which one to choose? These have a good reputation over here but availability seems difficult:
http://www.vaneijndhoven.net/jos/switchr/design.html
http://www.vaneijndhoven.net/jos/relaixedpassive/index.html
https://www.tentlabs.com/Components/page31/page31.html
Electronic? Also OK but greatly depending on the IC. Distortion when buffers are omitted, clicks when changing volume etc. Many sound less optimal compared to switched resistors/potentiometers and all the electronic volume controls I built were not used. Muses volume control seems a step ahead. Sometimes the thought comes up that the remote control possibility ways heavier than the best sound quality.....
It also seems the relay types offer a lot of advantages and best of both worlds qualities but ... I never got myself into building one. Used the RK27 and Khozmo to satisfaction 🙂 and then ... there was TKD. They were praised and seemed a quality in-between solution. Not the guru endorsing type but Bruno Putzeys found some interesting things with TKD potentiometers which is why I avoided them after reading this many years ago. Now the type is unknown but the information is shocking considering the average prices and supposedly good reputation. The consumers that buy and only listen seem to like them. Bruno did not. From his blog http://www.brunoputzeys.be/r4.php?item=25
Never as planned
Anyone seriously looking at distortion in resistors should take care to test at high frequencies. At 5 euros a pop (Farnell price), the Vishay MPM divider series are reassuringly expensive but I've just had to eliminate it from a preamplifier circuit as at 20dBu and 20kHz it produced 0.001% THD. That's a stupendous amount coming from a precision resistor.Also out are reed relays. Those wonderful red Coto's are the choice for reliable switching at low currents and voltages, but at 30mA (and again at 20kHz) I found about 7uV of 3rd harmonic across the contact, which as expected came down a bit as the coil voltage was increased.
The irony (pardon the pun) is that I'm discovering all this whilst building a relay-switched gain control to replace an esoteric audiophile Japanese potentiometer that was plainly distorting. Mentioning this to the manufacturer (TKD) elicited exactly no response at all. Mind, you can get virtually zero distortion from even a cheap linear pot if you use it as the feedback divider in an inverting amplifier. Clearly doing the same with a logarithmic pot is a waste of time. Doing so with an expensive logarithmic pot is a great way of discovering just how many relays the same amount of moolah can buy.
Sunday 23 September 2012
So, right now I can use a quality relay controlled attenuator (preferably a 10 kOhm version) myself but which one to choose? These have a good reputation over here but availability seems difficult:
http://www.vaneijndhoven.net/jos/switchr/design.html
http://www.vaneijndhoven.net/jos/relaixedpassive/index.html
https://www.tentlabs.com/Components/page31/page31.html
Last edited:
It's not cheap but not a stupid price either. AN says "The range consists of a 100 K Ohm standard logarithmic taper potentiometer and complementing balance control. Both feature a conductive track made from thick silver, a wiper made from solid silver and heavily silver plated connection pins. We use these in all of our more affordable products".
I was hoping for just a good quality conductive plastic one. Looks like Bourns does one which you can buy from Buerklin in MOQ of 270 so no good there, and there's the TT P092N from Mouser.
Pure/solid silver is often used in niche/boutique products but it is not durable and will lead to scratching sounds etc. That is why alloys are normally used in quality switches.
Don't be surprised that you fork out a high amount of money for a hand made Yatsushiru 48 step attenuator with solid silver contacts and have scratching sounds after a few months.
Don't be surprised that you fork out a high amount of money for a hand made Yatsushiru 48 step attenuator with solid silver contacts and have scratching sounds after a few months.
A remote cut-off valve or a few of them could serve as a VCA in a valve amplifier. The distortion won't be very low, but that holds for most audiophile valve amplifiers.
According to the datasheet it adds 0.09 - 0.2% THD. Is that what you want ?
No. The non linearities in a transformer appear as a variable resistance, so distortion due to this variation depends on the source impedance. In good quality transformers it
becomes almost non existent. Distortion due to saturation is a much bigger problem but even this can be reduced to very low levels with an optimum design using the best materials.
Why do you start your post with "No"? The low-frequency roll-off PRR wrote about exists, as does the distortion you wrote about.
Both get less when the driving impedance is very small or preferably slightly negative, but I think you can get the same effect when the load impedance is low or slightly negative, like a virtual ground input. As least I can't think of a reason why that wouldn't work.
Both get less when the driving impedance is very small or preferably slightly negative, but I think you can get the same effect when the load impedance is low or slightly negative, like a virtual ground input. As least I can't think of a reason why that wouldn't work.
Wide frequency band. I said nothing about linearity.
Lo-Z on one side or the other helps bass extension but also allows optimization of treble extension.
Another possibility is an active gain control. You get to use a linear pot or stepped attenuator of whatever quality you like, and there are distinct noise advantages.
But I'm not aware of how practical this may be with valve stages.
But I'm not aware of how practical this may be with valve stages.