I want to hear the views of Members on using CCS control for the LED current in LDR volume adjuster.
There seem to be a number of obvious topologies to achieve current control and I'm sure there must be many not so obvious topologies.
I am going to start off this discussion by showing Wapo's version mentioned in the Lightspeed Thread.
http://www.diyaudio.com/forums/anal...nuator-new-passive-preamp-94.html#post3032839
It is based on an Analog Devices paper which in turn is using their IC version of a Howland Current Pump.
Difference Amplifier Forms Heart of Precision Current Source: Analog Dialogue: Analog Devices
We can use a conventional log law single track potentiometer to divide a 5000mV Vref voltage required at the input of the AD diff amp precision current source, then we can output a control current to the LEDs. We could just as easily use 2.5V from a TL431 or any other reference voltage.
Using 40k resistors (or 39k if going discrete) for Rg1, Rg2, Rf1, Rf2 and selecting 270r for R1,
we end up with the maximum output current from the fig4 schematic as Io(max) = 5000mV / 270r = 18.5mA.
If the log law pot is turned down to -74dB then the min Io becomes 5000 * 0.02% / 270r = 3.7uA
-74dB requires the minimum output voltage to be 0.02% of the 5V reference and this requires the output impedance of a 10k control pot to be ~2r
All the 10k pots I have measured can consistently achieve less than this.
That 3.7uA to 18.5mA range of control currents gives a wide range of usable LDR resistance for the Series LED/LDR.
We still need an inverted control current for the Shunt LED/LDR.
I see two solutions for the inverted control current.
1.) do as George did and introduce a dual track pot.
2.) electronically derive an inverse control current from the output of the single track control pot.
Let's hear what can be done?
Then we can look at listening and measuring what the various topologies can achieve.
There seem to be a number of obvious topologies to achieve current control and I'm sure there must be many not so obvious topologies.
I am going to start off this discussion by showing Wapo's version mentioned in the Lightspeed Thread.
http://www.diyaudio.com/forums/anal...nuator-new-passive-preamp-94.html#post3032839
It is based on an Analog Devices paper which in turn is using their IC version of a Howland Current Pump.
Difference Amplifier Forms Heart of Precision Current Source: Analog Dialogue: Analog Devices
We can use a conventional log law single track potentiometer to divide a 5000mV Vref voltage required at the input of the AD diff amp precision current source, then we can output a control current to the LEDs. We could just as easily use 2.5V from a TL431 or any other reference voltage.
Using 40k resistors (or 39k if going discrete) for Rg1, Rg2, Rf1, Rf2 and selecting 270r for R1,
we end up with the maximum output current from the fig4 schematic as Io(max) = 5000mV / 270r = 18.5mA.
If the log law pot is turned down to -74dB then the min Io becomes 5000 * 0.02% / 270r = 3.7uA
-74dB requires the minimum output voltage to be 0.02% of the 5V reference and this requires the output impedance of a 10k control pot to be ~2r
All the 10k pots I have measured can consistently achieve less than this.
That 3.7uA to 18.5mA range of control currents gives a wide range of usable LDR resistance for the Series LED/LDR.
We still need an inverted control current for the Shunt LED/LDR.
I see two solutions for the inverted control current.
1.) do as George did and introduce a dual track pot.
2.) electronically derive an inverse control current from the output of the single track control pot.
Let's hear what can be done?
Then we can look at listening and measuring what the various topologies can achieve.
Last edited: