Lightspeed Attenuator a new passive preamp

[gizmologic]

Hi George,

Since I am new at the DiY Audio, i am not permitted to send direct mail, nor was I able to find any URL that leads to your products.

Therefore, kindly let me know how i can get in contact with you and purchase the ready made product.

Best regards

Patrik a.k.a Gizmologic
gizmologic@gmail.com

[georgehifi]

Originally posted by gizmologic
Hi George,

Since I am new at the DiY Audio, i am not permitted to send direct mail, nor was I able to find any URL that leads to your products.

Therefore, kindly let me know how i can get in contact with you and purchase the ready made product.

Best regards

Patrik a.k.a Gizmologic
gizmologic@gmail.com

Hi Patrik, you have mail and loads of propaganda.

Cheers George

[Greyskale]

not to derail the thread, but there's been a lot of discussion here: Link about the use of one only one LED/LDR per channel as a shunt as opposed to the series/shunt setup. I am curious if anyone else has read this as well and what opinions you might have on the subject.

[georgehifi]

Originally posted by Greyskale
not to derail the thread, but there's been a lot of discussion here: Link about the use of one only one LED/LDR per channel as a shunt as opposed to the series/shunt setup. I am curious if anyone else has read this as well and what opinions you might have on the subject.

That was what my first Mk1 Lightspeed Attenuator was. But it was quickly learnt that series/shunt (2 LDR's) sounded superior to (one LDR), it also gave a much more constant input impedance and output impedance.
So all the Mk1's that I manufactured and sold, got recalled and modified up to Mk2 status (2 LDR's), with every owner saying it was far better when they got it back.

Cheers George

[Greyskale]

Well, I'm taking the plunge on your design... I've designed a PCB and had it set off to be fabricated and all parts have been ordered. I added a small filter cap just before each NSL-32SR2S in such a way to keep them optional (testing time ) but I figure it might add a little extra stability. I'll post some pics and impressions when it's all put together and tested.

[jjj333]

I would love to use a n optocoupler volume control to control MIDI instruments. Yet, since both hands are busy playing the keyboard I need to find another way to vary the volume levels for vibrato and overall dynamics (like used by flute/ accordion players etc).

A foot pedal isn't good enough, because with the foot I can't vary vibrato intensity accurately as needed. Any suggestions on how to solve this problem??

[georgehifi]

jjj333, it sound as though you want an accurate instant control over the volume continuously, the Lightspeed may not be suitable for you, as it has a slight delay from when the volume is turned up or down, kind of like steering a barge or very large boat, it responds after the fact.
This is not a problem when using it as a volume control for audio, because when you turn the control it has found it's level before you have sat back down to listen. But to use it up and down constantly, I could see some delay factors being a problem for you.
This is a time vs level setting distortion which many have totally misconstrued as sound distortion, (lost souls) and in no way affects listening to audio but with your way of wanting to use it constantly, you will have a problem.

Cheers George

[Richidoo]

Obviously the suggested method of sorting LDRs for matching curves at constant voltage works to everyone's satisfaction. But in the spirit of great mousetrap designers everywhere, I would like to ask this dumb question:

Why not drive the LEDs with current instead of voltage, as suggested by Silonex? There are LED driver chips available which will drive LEDs to variable current with precision. Like these app note from Maxim.

Any thoughts?
Thanks
Rich

[pietjers]

Why not drive the LEDs with current instead of voltage, as suggested by Silonex?

That is exactly what I was thinking for some time! Couldn’t get rid of that idea in my head so I made a test setup to be able to compare between the original Lightspeed and a constant current source (CCS) version, as a volume control for my pre-amp. I connected the LED’s of the NSL-32SR2S by a connector to be able to switch between the two versions in a few seconds.
To make a long story short……the current source version is sounding MUCH better!! I am not talking about small differences here!

There is one problem to cope with. In my design the constant current also flows through the potentiometer (100k) which sets the current through the LED’s. That gave some severe crackling at the highest volume settings. Potentiometers don’t like DC voltage and with constant current things even get worse. So I used my DACT stepped attenuator, which was already in my pre-amp, to control the current sources. With the DACT everything is dead silent.

There is one other topic here in this thread that pops up from time to time, and that is the question what is best sounding: series LDR / parallel LDR or series resistor / parallel LDR. I did some additional listening tests to cover this topic as well. In my tube pre-amp I have some gain to waste, so the series resistor is no problem. I tested with the famous Shinkoh 47k tatalum resistors. The series LDR / parallel LDR combination sounds so spatial (also due to the CCS) that in my system it is to much sometimes. I also missed some of the focus and directness that the resistor / LDR version has. And voices and instruments have a more natural timbre with the resistor version. So I prefer the series resistor / parallel LDR version in my setup. That is to say, in my system and with my ears. Someone else could judge different because both sound very good.

The overall sonic difference between the DACT volume control (in a shunt configuration in my case) I used before and CCS LDR volume control is phenomenal.

[EUVL]

Would you be kind enough to tell us how you constructed your current source that should have a range of say 20uA to 20mA ? Are you using bipolar transistors or JFETs ? With or without cascode ?


Patrick