Good point Andrew. Others cannot use a name that might cause confusion either, witness McDonalds pprotecting their trademark against any and all McSomethings. Sorry about that, George I didn't mean to push your trademark towards being generic, I knew better. Zen Mod correctly avoided calling his thread Lightspeed. http://www.diyaudio.com/forums/analog-line-level/134316-poor-serbian-man-optical-volume-control.html
Last edited:
Dac with LDR attenuator
I recently finished my DAC with Doedes DDDAC with 60x TDA1543 chips, I2S input from the Transport, separate 11V power supply, Uriahs LDR attenuator kit and a IR remote controlled poti kit for the volume. Sounds fantastic and no need for preamp, just directly into my 300B SET power-monos.
Thanks to this thread, which introduced me to the LDR's , I can now enjoy fantastic sound with the least possible components in the signal path.
This is what's all about, isn't it?
Thanks to this thread, which introduced me to the LDR's , I can now enjoy fantastic sound with the least possible components in the signal path.
This is what's all about, isn't it?
I recently finished my DAC with Doedes DDDAC with 60x TDA1543 chips, I2S input from the Transport, separate 11V power supply, Uriahs LDR attenuator kit and a IR remote controlled poti kit for the volume. Sounds fantastic and no need for preamp, just directly into my 300B SET power-monos.
Thanks to this thread, which introduced me to the LDR's , I can now enjoy fantastic sound with the least possible components in the signal path.
This is what's all about, isn't it?
Another one that has seen the Light
"KISS" that's what it's all about.Cheers George
If I'm to OT just say so and I'll drop the matter.
It would be interesting to talk a little about what matters and how much it matters.
Talk about balanced operation has been very dogmatic and "black and white".
Iirc balanced operation is mainly used to reduce distortion picked up in the cables. I don't have the correct words but I hope you get the idea anyhow.
This is done by cancellation and differential amplification.
What happens if the + and - is of different amplitude?
The rf interference picked up in the cables should be roughly the same.
It's possible the reduction in electric fields are not perfect since the amplitude are not equal?
Still, I wonder...
Is it even noticeable?
The L/R difference in output swing can be corrected with channel balance correction. (easily done)
I guess my question is:
Does it even matter if the balanced operation is a little off? Naturally you want the LDR to track well but it's mostly a matter of conveniance. If you don't mind correcting the channel balance you could pretty much use any un-matched LDRs?
It would be interesting to talk a little about what matters and how much it matters.
Talk about balanced operation has been very dogmatic and "black and white".
Iirc balanced operation is mainly used to reduce distortion picked up in the cables. I don't have the correct words but I hope you get the idea anyhow.
This is done by cancellation and differential amplification.
What happens if the + and - is of different amplitude?
The rf interference picked up in the cables should be roughly the same.
It's possible the reduction in electric fields are not perfect since the amplitude are not equal?
Still, I wonder...
Is it even noticeable?
The L/R difference in output swing can be corrected with channel balance correction. (easily done)
I guess my question is:
Does it even matter if the balanced operation is a little off? Naturally you want the LDR to track well but it's mostly a matter of conveniance. If you don't mind correcting the channel balance you could pretty much use any un-matched LDRs?
Last edited:
Hi,
balanced connection's main advantage is interference attenuation.
It achieves this using balanced impedance connections, not balanced voltage connections.
It is the impedance matches that result in the good attenuation of interference.
If impedances are not balanced then the major reason for using balanced connections has gone.
balanced connection's main advantage is interference attenuation.
It achieves this using balanced impedance connections, not balanced voltage connections.
It is the impedance matches that result in the good attenuation of interference.
If impedances are not balanced then the major reason for using balanced connections has gone.
You'll find many explanations of the circuit on the web but there's a typically articulate one on Rod Elliott's site:
Balanced Line Driver & Receiver
As an aside and IMHO, the clamour to over-elaborate and "improve" the Lightspeed circuit misses a key point. Better power supplies (as with e.g. The Lighter Note) definitely make for better performance and many users appreciate remote control but I know of no successful LDR-based attenuator that has improved on George's basic circuit.
Many have suggested "better" designs but none to date (and it's been years rather than months) have come to fruition.
I don't find this surprising - with a device as quirky (as non-linear) as an LDR, getting it to work surely means that simplicity is of the essence.
The fact is that this simple circuit, despite a succession of arm-chair theoreticians explaining why it shouldn't work, does work and, as we know, works well. This is not a trivial point.
Last edited:
Rod Elliott's saying pretty much exactly the same as what I'm trying to say.
"Only the out of phase signal is detected by the remote balanced receiver, and any in phase (common mode) signal is rejected. RF interference and other noise will be picked up equally by both wires in the cable and so will be in phase. It will therefore be rejected by the receiver."
This theoretical line of reasoning is just for fun. Understanding why things behave as they do increase the chance for success.
"Only the out of phase signal is detected by the remote balanced receiver, and any in phase (common mode) signal is rejected. RF interference and other noise will be picked up equally by both wires in the cable and so will be in phase. It will therefore be rejected by the receiver."
This theoretical line of reasoning is just for fun. Understanding why things behave as they do increase the chance for success.
the receiver must have a balanced impedance input combined with the transmitter/source having a balanced impedance output for the common mode rejection to be effective.
If the impedance is not balanced at both ends of the balanced cable the the interference rejection falls well short of what the balanced connection has primarily been adopted for.
The voltage levels in the two halves of the balanced connection are of much less consequence, because the receiver is designed as a differential amplifier. It amplifies the difference between the two signal of the balanced connection.
One could be xVac + DC offset. The other could be yVac + DC offset.
x and y need to be different for a signal to exist.
X=-y is one case that works. But, x and 0 would work just as well. 0 and y would also work just as well. x and 3x would also work.
The receiver amplifies the difference between the two signal lines.
If the impedance is not balanced at both ends of the balanced cable the the interference rejection falls well short of what the balanced connection has primarily been adopted for.
The voltage levels in the two halves of the balanced connection are of much less consequence, because the receiver is designed as a differential amplifier. It amplifies the difference between the two signal of the balanced connection.
One could be xVac + DC offset. The other could be yVac + DC offset.
x and y need to be different for a signal to exist.
X=-y is one case that works. But, x and 0 would work just as well. 0 and y would also work just as well. x and 3x would also work.
The receiver amplifies the difference between the two signal lines.
There's the other half of my initial thought, the voltage levels aren't important.
The source before the LDR's and the load after the LDR's should be easy enough to line up the correct impedances.
The only component left is the Lightspeed inspired unit and the only part that matters is the impedances.
All systems have non-linear behaviour in varying degrees. As in most cases the question becomes: how much non-linearity is acceptable?
At this point we're just interested in how much non-linearity we can accept for the balanced operation to work satisfactory?
The source before the LDR's and the load after the LDR's should be easy enough to line up the correct impedances.
The only component left is the Lightspeed inspired unit and the only part that matters is the impedances.
All systems have non-linear behaviour in varying degrees. As in most cases the question becomes: how much non-linearity is acceptable?
At this point we're just interested in how much non-linearity we can accept for the balanced operation to work satisfactory?
Mike
Why not just try it the way it is first? I have used down to 47k input impedance with no problem. A few people have experienced problems and if you are having an issue then just remove the lightspeed and do your work inside the amp.
There is an input resistor going from signal to ground inside your amp. Its the first thing the signal encounters unless there is a small 1k or so series resistor as well. So take that resistor going to ground and replace it with 100k.
Someone shoot me if I'm wrong.
Uriah
Why not just try it the way it is first? I have used down to 47k input impedance with no problem. A few people have experienced problems and if you are having an issue then just remove the lightspeed and do your work inside the amp.
There is an input resistor going from signal to ground inside your amp. Its the first thing the signal encounters unless there is a small 1k or so series resistor as well. So take that resistor going to ground and replace it with 100k.
Someone shoot me if I'm wrong.
Uriah
Ask MF. I would be inclined to say, yes.
You would need to analyse the input stage and filtering and also look at the output offset voltage and input offset current. Is it AC or DC coupled?
Does it have a DC servo? is there DC detect and/or input muting?
Im using my home built version with an Aleph 5 [10k] sounds good to my ears.
Currently tried it with a DIYparadise Eva 2. Insufficient volume and some rolled off highs. Rather than add a buffer, I am considering increasing impedance. I believe that the Eva has a higher output impedance than the Lighter Note and with my 31K amp, I guess that's enough of a difference. I may switch to the LN if I can't get the Eva and my amp to work together.
Thanks,
Mike
It makes sense that MF decided that a config with 31K sounded best. I have a query in to MF now. I don't think I will be tinkering with the amp unless it's just a simple resistor change.
Thanks,
Mike
Hmmm. Perhaps it's the Eva 2. Sound like either the LS or LN doesn't have these problems.
Thanks,
Mike