Lightspeed Attenuator a new passive preamp

[EUVL]

Thank you Peter.

Patrick

[georgehifi]

Hate to burst your bubble guys, but the LDR's will not sound any different this way as it does not when batteries are used to power the led's, the way the led is lit up has no influence on the way the LDR sounds.
So long as the energy used to light up the led is stable and quite is all that matters. I have injected all sorts of garbage into the led power rails to see if it effects the sound and it doesn't, the only thing that does, as I said is uneven power (fluctuating), and very noisy power (switch mode garbage) but this was hard to detect.

All your doing is taking a very simple effective design and complicating it for no gain in sound quality, and I will repeat myself once again, series/ shunt LDR's sound much better than just shunt, and I believe it's because they have a much more constant input and output impedance and able to impedance match to the source and power amp much better, the shunt is all over the place and according to the system will only be matched up at a particular volume setting, everywhere else you'll have impedance matching problems, also the series/shunt has the added advantage of being able to reach down closer to zero volume.

Enough for now
Cheers George

[pietjers]

All your doing is taking a very simple effective design and complicating it for no gain in sound quality

George,

First this: congratulations with the fact that your topic exceeded the 1000 posts!!

Back to your reply:
How can you be so sure without actually trying it? Audible effects are not always easy to explain. For every audible effect is an explanation, but things are not always clear right away. The fact is that the amount of light a LED emits is dependent of the current flowing trough it. That is the reason why Silonex recommends current for the LED’s and the rest of the world in critical applications is driving LED’s by the means of current sources.

I made a test setup in which I was able to switch between your original design and the constant current source version in just seconds. I can assure you that the sonic differences are not small!

In my opinion the performance of a circuit is more important than simplicity. And who is afraid of just a hand full of components? I think my current source circuit is still quite simple.

[Lostcause]

Originally posted by pietjers


George,

First this: congratulations with the fact that your topic exceeded the 1000 posts!!

Back to your reply:
How can you be so sure without actually trying it? Audible effects are not always easy to explain. For every audible effect is an explanation, but things are not always clear right away. The fact is that the amount of light a LED emits is dependent of the current flowing trough it. That is the reason why Silonex recommends current for the LED’s and the rest of the world in critical applications is driving LED’s by the means of current sources.

I made a test setup in which I was able to switch between your original design and the constant current source version in just seconds. I can assure you that the sonic differences are not small!

In my opinion the performance of a circuit is more important than simplicity. And who is afraid of just a hand full of components? I think my current source circuit is still quite simple.

OK, you need to explain in more detail how this method changes the behaviour of the LDR.
The resistor is.... well... a resistor yes?
Light changes it's resistance. This resistance is not affected by anything else.... current and voltage do not have any effect on how the principal physics work. so how can a control of the 'variance' have any affect?
You need to prove that there is fluctuation (and this is the difficult part) that changes the resistance on something that has such a very low response time to change.
Don't get me wrong here, if you can prove your point then great but I just don't see it right now........

[georgehifi]

Ditto, like he said.

Cheers George

[1543]

Did some measurements concerning series resistor and THD. From a measurement point of view series resistor / shunt LDR configuration is the way to go. Nonrelevant what kind of control (voltage or current) is used.

At mid attenuation (normal listening sound level) the THD of LDR shunt attenuator is as good as THD when using a stepped attenuator.

I have compared / measured a stepped attenuator, a LDR attenuator (lightspeed configuration with series LDR + shunt LDR) and an attenuator with 22 k series resistor and shunt LDR (with 0,1 µF Bypass).

The series LDR and shunt LDR configuration adds coloration, which can enjoy of course.

Measurement at mid attenuation:


Measurement at lowest attenuation:

[georgehifi]

These are the Silonex measurements, green is the THD measurements red is the db level
As you can see the Series LDR with Shunt LDR has the best THD% at the medium to louder levels and the Series Resistor with Shunt LDR is better at lower levels, most of the listening with my Lightspeed is done in the top 1/3 of the setting after 12o'clock with cd as source.
But all this is academic as the distortion that's present is H2 and low and is not unpleasant to the ear.

Here is the Series LDR & Shunt LDR

Cheers George

[georgehifi]

Here is the Series Resistor & Shunt LDR

Cheers George

[1543]

Hi Goerge,

thanks. I know the measurement from Silonex before. The Silonex measurements shows us:

The series+shunt configuration has low THD+N at levels from 0-10 db.

The series resistor / shunt LDR configuration has better specs at attenuation level - 10 db to ~.

Depending on the gain factor of your system, the one or the other configuration is best choice. Normally I do not listen at -10 to 0 dB attenuation levels, because of do not want to anger my neighbors.
With my measurements I do not see any improvement in THD+Noise when using Series LDR / Shunt LDR at low attenuation.

With my last post I wanted to show that the shunt LDR configuration could have be an THD+N Level as good as a passive stepped attenuator. And using only 1 LDR per channel making it easier to add µC and remote functionality for the couch potatoes.

@pietjers
Good design. Not sure of having sonic improvements instead of using voltage to control the LDR (perhaps there is less drift with temperature).

@barryblue
The AD420 is perfect for controling LDR's with less parts as possible. A few month ago I started a 6-channel design with voltage out DACS , but it results more an more in an overkill (using adapted Silonex circuits for linear control and temp. compensated voltage supply). Damned that I did not know the AD420 before :-( Anyway, thanks a lot for that hint.


Regards, Arne

[Nordic]

Set phaser to kill kill kill...