Oeh, my brain hurts. Does the pot not change the light and reflect its distortion in that manner? If not how does one change the volume? I am not getting it.
The other thing I am trying to grasp here is that I have many switches in my system. I count 5 switches without knowing the details of each circuit (DAC, Active Xover, 2 monoblocks for higher frequencies, and 1 amp for lower frequencies). I have gotten rid of the pots in my active xover by swapping in resistors that are connected with screw clamps. I am guessing that the response will be that these other switches are buffered with PS capacitors and the like. And that the attenuator described here is more critical. But still the proposal seems to be that the opto-controlled resistor somehow blocks distortion and passes a more pure signal. So, while that may be true, why would that be?
Semantics. How does the electrical distortion from the pot which controls the intensity of the light end up not distorting your signal? That is the basis of the proposal here. I get it only if the optical-resistor preferentially buffers distortion over the music signal. Where is the data?
It doesn't block distortion. If your cdp or dac sends distortion every single component down the line will welcome that distortion. There is no device that will reduce distortion that is sent to it because it wont be able to differentiate between distortion and signal. Capacitors store DC and pass AC. There is no buffering going on in/with a cap.
If you want to learn about building just buy a kit with good instructions. If you want to learn about circuits I would start with a battery a resistor or potentiometer and a led. At the same time build a chip amp and learn schematics.
If you want to learn about building just buy a kit with good instructions. If you want to learn about circuits I would start with a battery a resistor or potentiometer and a led. At the same time build a chip amp and learn schematics.
The pot sets the level of constant bias current in the diodes, which then emit a constant level of light
that shines on the LDR, which sets the operating resistance of the LDR, which sets the attenuation level
and hence the volume. The pot has nothing to do with the audio signal directly. There's no distortion
due to the pot in the light going to the LDR. The light is NOT modulated with the audio signal.
However, the LDRs themselves DO generate significant distortion because they are not as linear as
conventional resistors. Some measurements can be seen here:
http://www.diyaudio.com/forums/anal...uator-new-passive-preamp-134.html#post1520950
Last edited:
Only when over driven, if the signal is at Redbook CD source level outputs, it remains way under .1% (even on heavy transients) and that is mainly the "nice" tube sounding one 2nd harmonic. Which is maybe why so many like the sound of the Lightspeed. Also another reason maybe, it's conductor instead of being metal film or carbon as in resistors or pots, is Cadmium Sulphide, and no contacts like resistors or pots need, which are dissimilar materials and very light weight contacts, which they need to change their resistances.
And Rama is right bcc1955, a led does not flicker (is what I think your getting at) when driven from a good power supply and has no modulating effect on the LDR. Beside LDR's are very slow to react to light changes and in a way are self damping to that as well.
Cheers George
Last edited:
Udaily,
I continue to think it is semantics.
You say "it doesn't block distortion. If your (source)..sends distortion every ...component down the (passes it on). no device...will reduce distortion....because it wont be able to differentiate between distortion and signal"
That is precisely my comment. There appears to be a claim on this thread that the signal is passed directly to the speakers without any influence from the potentiometer that controls the LED light emission. That the LED light emission is absolutely constant regardless of current distortions caused by the pot this device claims to remedy.
Rayman and George,
"The pot sets the level of constant bias current in the diodes, which then emit a constant level of light that shines on the LDR, which sets the operating resistance of the LDR, which sets the attenuation level and hence the volume"
"A led does not flicker (is what I think your getting at) when driven from a good power supply and has no modulating effect on the LDR. Beside LDR's are very slow to react to light changes and in a way are self damping to that as well"
Me: Yeah all that sound great, there is a "constant level of light" and "LEDs do not flicker" if the power supply is perfect. But the premise is that this system is not subject to distortions from traditional potentiometers. So you are both claiming that distortions from poor pot connections are not reflected in the LED light produced. This is why I was asking if there is a buffer effect that prevents that transfer of distortion in current (changing resistance at the metal to metal pot interface). To which George has given a reasonable answer. If what he says is correct the music signal may truly be somewhat isolated (or buffered) from the volume level distortion produced by the pot.
"LDR's are very slow to react to light changes and in a way are self damping", which is what I meant when I said "buffered".
I continue to think it is semantics.
You say "it doesn't block distortion. If your (source)..sends distortion every ...component down the (passes it on). no device...will reduce distortion....because it wont be able to differentiate between distortion and signal"
That is precisely my comment. There appears to be a claim on this thread that the signal is passed directly to the speakers without any influence from the potentiometer that controls the LED light emission. That the LED light emission is absolutely constant regardless of current distortions caused by the pot this device claims to remedy.
Rayman and George,
"The pot sets the level of constant bias current in the diodes, which then emit a constant level of light that shines on the LDR, which sets the operating resistance of the LDR, which sets the attenuation level and hence the volume"
"A led does not flicker (is what I think your getting at) when driven from a good power supply and has no modulating effect on the LDR. Beside LDR's are very slow to react to light changes and in a way are self damping to that as well"
Me: Yeah all that sound great, there is a "constant level of light" and "LEDs do not flicker" if the power supply is perfect. But the premise is that this system is not subject to distortions from traditional potentiometers. So you are both claiming that distortions from poor pot connections are not reflected in the LED light produced. This is why I was asking if there is a buffer effect that prevents that transfer of distortion in current (changing resistance at the metal to metal pot interface). To which George has given a reasonable answer. If what he says is correct the music signal may truly be somewhat isolated (or buffered) from the volume level distortion produced by the pot.
"LDR's are very slow to react to light changes and in a way are self damping", which is what I meant when I said "buffered".
Last edited:
"LDR's are very slow to react to light changes and in a way are self damping", which is what I meant when I said "buffered".
Correct
It takes 80msec for the ldr to react to a change in led dimming or brightening. Hence they are self damping, to any led flicker if there is any, which in this case there is none.
Cheers George
Correct
It takes 80msec for the ldr to react to a change in led dimming or brightening. Hence they are self damping, to any led flicker if there is any, which in this case there is none.
Cheers George
Last edited:
OK, let's assume that you are right, that the potentiometer that controls the LED's has "current distortion". How would you characterize this distortion? Is it harmonic distortion? The pot is controlling a DC voltage supplied to the LED's. What is the second harmonic of DC? The third harmonic?
OK, maybe it's intermodulation distortion. The pot is only passing DC, so what are the IM products?
Or maybe you mean noise? Fair enough, but because of the slow response of the LDRs and the filtering across the pot/LED, even a very noisy pot is unlikely to cause significant noise-based modulation of the audio signal.
Again, the audio signal does not pass through the pot.
Disclaimer: I have not heard one of these attenuators, and I do not consider the "problem" they solve to be very important, but I think I understand how they work and what they are supposed to accomplish.
The input impedance of the LDR is roughly equal to the sum of the series and shunt resistances.
For higher values of attenuation the shunt LDR needs to go low and that is set by the 68R+VR (~200r/2)
The series LDR has little effect at low levels of attenuation and thus does not need a resistance as low as the shunt LDR.
The minimum series LDR resistance is set with the 1k+VR (~5k/2).
These two sets of LED current limiting resistors can be any value that suits the builder provided that the maximum LED current does not risk damaging the LED/LDR in the long term.
But the input impedance is not significantly affected by these current limiting resistors.
It's the current fed from the tracks of the 100k volume setting pot that sets the input impedance.
More current = lower impedance. Less current = higher impedance.
These impedances are not constant. They will vary with volume setting, much more so that with a conventional vol pot feeding into a load that is 10times higher.
If you want to experiment with varying the LED current to see what effect it has on the input impedance, you can replace the fixed 5V supply with a variable supply.
I'd suggest one that can be varied from 3V to 6Vdc.
At 6V a bit more LED current than standard will flow and this will lower the input impedance and lower the output impedance.
Conversely as you lower the voltage towards 3V the LED current will drop and the input/output impedances will increase.
For higher values of attenuation the shunt LDR needs to go low and that is set by the 68R+VR (~200r/2)
The series LDR has little effect at low levels of attenuation and thus does not need a resistance as low as the shunt LDR.
The minimum series LDR resistance is set with the 1k+VR (~5k/2).
These two sets of LED current limiting resistors can be any value that suits the builder provided that the maximum LED current does not risk damaging the LED/LDR in the long term.
But the input impedance is not significantly affected by these current limiting resistors.
It's the current fed from the tracks of the 100k volume setting pot that sets the input impedance.
More current = lower impedance. Less current = higher impedance.
These impedances are not constant. They will vary with volume setting, much more so that with a conventional vol pot feeding into a load that is 10times higher.
If you want to experiment with varying the LED current to see what effect it has on the input impedance, you can replace the fixed 5V supply with a variable supply.
I'd suggest one that can be varied from 3V to 6Vdc.
At 6V a bit more LED current than standard will flow and this will lower the input impedance and lower the output impedance.
Conversely as you lower the voltage towards 3V the LED current will drop and the input/output impedances will increase.
Last edited:
>Or maybe you mean noise? Fair enough, but because of the slow response of the LDRs and the filtering across the pot/LED, even a very noisy pot is unlikely to cause significant noise-based modulation of the audio signal<
Yes, this is similar to the explanation George gave as well and it makes good sense to me. I guess my unconventional terminology and the term "buffering" (a result of my biochemistry background) caused some misunderstanding. At any rate I may end up getting one of these to place between my turntable and active crossover. I just need to think of the implication of this versus a TVC (with an expensive quality pot) for that application, and anything I might add into the mix in the future.
So all and all I greatly appreciate everyone helping explain LDR volume control to me.
Yes, this is similar to the explanation George gave as well and it makes good sense to me. I guess my unconventional terminology and the term "buffering" (a result of my biochemistry background) caused some misunderstanding. At any rate I may end up getting one of these to place between my turntable and active crossover. I just need to think of the implication of this versus a TVC (with an expensive quality pot) for that application, and anything I might add into the mix in the future.
So all and all I greatly appreciate everyone helping explain LDR volume control to me.
Sorry José Humberto, and all others.
I do not ground the - of the power supply to chassis. It floats. I got just a minute touch better noise figures this way, which were down in the 10uV (not mV) region, just visible on base line of my Tektronic scope.
Here is the amended circuit, the one I posted before was an old one.
Cheers George
I do not ground the - of the power supply to chassis. It floats. I got just a minute touch better noise figures this way, which were down in the 10uV (not mV) region, just visible on base line of my Tektronic scope.
Here is the amended circuit, the one I posted before was an old one.
Cheers George
Attachments
the LED has a Vf that varies slightly with current.
at a set current, the Vf is fixed.
One does not send in a different voltage to change the light level.
One sends a different current to change the light level. That will result in a slightly different Vf, but substanially similar to what it was before the current change.
at a set current, the Vf is fixed.
One does not send in a different voltage to change the light level.
One sends a different current to change the light level. That will result in a slightly different Vf, but substanially similar to what it was before the current change.
