The -IN input is at very high impedance and there may be a very small DC input offset. The current passing through the cursor can only be very low.
The whole point of Bruno's pre-amp is to use good circuit design to eliminate the use of fancy parts as an attenuator - by not using an attenuator.
- from Bruno's article, just a few lines down from discussing wiper distortion in pots used as normal attentuators. The G word: Demo project - A balanced volume controller | EDN
BUT my ignorance of electrical engineering could mean I'm missing something - e.g. that there can be distortional effects of a wiper in a pot's use as a sole feedback element (that Bruno forgot about.... perhaps).
Last edited:
But this is exactly what I have been trying to say all along, but somehow my message is not coming thru. Maybe because English is not my native language.
The digital potmeter should be used in exactly the same way as the analog potentiometer, not as an attenuator.
If the distortion artifacts from a lousy wiper in a cheap potmeter is eliminated in Brunos circuit (because no current is going thru the wiper) the the same could hold true in a digital potmeter, as it probably is the switches that produces the distortion,( the resistive element inside the chip does not produce distortion by itself, as far as I can see), then this distortion is as well non existing, when used as the sole feedback element in the circuit.
Peace...
The distortion in a digital pot is caused by the switches which change resistance when the voltage across them changes. The switches are non-linear resistors, albeit with very low resistance.
A thought experiment: assume an attenuator step resistor selected of 9.9k, connected through a switch with a nominal resistance of 100 ohms. If the voltage across the combination is at the signal top, say 2V, the switch is now not 100 ohms but 110 ohms. So you have an attenuator changing from 10k to 10.01k which is where the distortion comes from: the signal shape is changed because it is attenuted more at the max value than at the min value.
So it is a totally different mechanism than in a mechanical pot.
Jan
A thought experiment: assume an attenuator step resistor selected of 9.9k, connected through a switch with a nominal resistance of 100 ohms. If the voltage across the combination is at the signal top, say 2V, the switch is now not 100 ohms but 110 ohms. So you have an attenuator changing from 10k to 10.01k which is where the distortion comes from: the signal shape is changed because it is attenuted more at the max value than at the min value.
So it is a totally different mechanism than in a mechanical pot.
Jan
As far as I can see from the description of the digital potmeters, this 100 ohms resistance is the WIPER resistance eg. it is in series with the wiper, not the resistive element. If no current is drawn out of the "wiper" (switch) there is no voltage drop and no nonlinearity. There are no switches in what is the resistive element in the digital potmeter , just as in a normal potmeter.
And this is exactly why I think Brunos way of using a potmeter COULD eliminate the distortion in a digital potentiometer.
Last edited:
I changed from a linear track vol pot to a switcher, with custom values of resistors.
I posted the resistor values some while ago and gave a short report pointing out how much better the control of attenuation was with the switcher compared to the linear track.
In my view the linear track as proposed by B.Putzeys is a waste of resources.
But, I much prefer the reproduction provided by the DCB1.
The balanced does not sound clean. I suspect the transformer (a very flat pancake toroid) I used is interfering with the audio signal and degrading it.
I did plan to replace it with another, but never got round to that. Maybe this winter.
Last edited:
They do not specify the circuit used for the THD measurement, but I assume that ADI used standard equipment to measure the THD quoted, which would include an input stage with a reasonably high input impedance.
Looking at the THD curves in the DS, they are noise-dominated, but since the noise and THD are lower for the parts with higher pot resistances, the primary noise source seems to be the electronics itself.
So I guess this noise would also be present in the application, if the chip survives the voltage difference across the pot terminals, that is.
Reagrds,
Braca
Thank you for this. Probably the most useful comment from this entire thread as far as i am concerned.
Your reply implies that ADI measured the THD by using a low input impedance instrument, because then there would have been a current flow through the wiper.
I might be missing something, but I don't know of any such instrument.
Besides, I also wrote that the distortion might be coming from the electronics in the pot because the noise in the THD plot reduces with the pot resistance, and the physics requires it to be the other way around.
Regards,
Braca
I wouldn't think there could be any noise from the electronics as nothing is going on inside the chip, when the potmeter is not changing value (position) .
ADI´s THD instrument has probably a high input impedance, but unfortunately, no one describes the test circuit used for THD measurements.
What puzzles me is where does the -80dB come from, if it is not nonlinearities in the switches?
This is a very academic discussion and it should probably be ended here, until we actually can measure Brunos circuit with a digital pot involved - right?
There is some tech info here:
Jeff Rowland :: Digital Volume Control: DACs versus the Crystal Semiconductor CS3310
Jan
Jeff Rowland :: Digital Volume Control: DACs versus the Crystal Semiconductor CS3310
Jan