Close enough
I use dual stereo 50K linear pots in mine. Use 200 ohm resistors.
Match them up as well as you can. Listening is log scale. Even 10% match is close on a log plot.
The match on the LDR is a long way from perfect. This is where dual volume pots comes in handy. If there is a channel imbalance, adjust it out.
Been listening to a new amp that requires a coupling cap to terminate the feedback loop properly. Here the Lightspeed clone really shines.
George
I use dual stereo 50K linear pots in mine. Use 200 ohm resistors.
Match them up as well as you can. Listening is log scale. Even 10% match is close on a log plot.
The match on the LDR is a long way from perfect. This is where dual volume pots comes in handy. If there is a channel imbalance, adjust it out.
Been listening to a new amp that requires a coupling cap to terminate the feedback loop properly. Here the Lightspeed clone really shines.
George
100K Attenuator
Hi, very interesting thread. I may build a standalone passive, but first if feasible I would like to substitute an LDR series/shunt circuit for the 100K 4-deck 23 step series attenuator in my diff-balanced Atma-Sphere MP-1 MkII phono stage/preamp. The phono section is 12AT7 and the LS is 6SN7 with circlotron. For the LDR volume control at the output of the phono section I would need fairly high input impedance close to the stock 100K stepped attenuator. Is this attainable with the Silonex LDRs?
Hi, very interesting thread. I may build a standalone passive, but first if feasible I would like to substitute an LDR series/shunt circuit for the 100K 4-deck 23 step series attenuator in my diff-balanced Atma-Sphere MP-1 MkII phono stage/preamp. The phono section is 12AT7 and the LS is 6SN7 with circlotron. For the LDR volume control at the output of the phono section I would need fairly high input impedance close to the stock 100K stepped attenuator. Is this attainable with the Silonex LDRs?
Think so
My system is all solid state these days, so a higher impedance volume control was not needed. With its planned application as a passive line, a 100K value is not optimal.
With Georges diagram, I seem to remember his using a 50K log pot. And 100 ohm current limiting resistors. Mine is a dual mono copy of his, so all the resistances are doubled. It measures around 20K. Maybe 17K with the control pots in mid range.
I think that if you build a version of George's posted design and use a stereo 100K log pots to run all four LDR's the resulting impdeance will be somehere between 50 -100K If it is still too low, either a 250K or a pair of 250K paralleled (125K) might get it to about a 100K.
As a volume pot this should work fine. It would KILL a passive line stage. Most passive units are closer to 10K. This is a value that preserves dynamics and still allows coupling with most sources.
I have been listening to mine today. Been swapping out with an active last couple weeks. And working on the active to get the clarity and pureness of the LS clone. Still trying to get there.
Next is replace the 25K PEC pots in the active with a LS clone voume control. Got to dig up the correct pots to control the LED's. If it improves over the PEC pots it will prove that it is the LDR elements and not just the active vs passive design that makes it so special.
George
My system is all solid state these days, so a higher impedance volume control was not needed. With its planned application as a passive line, a 100K value is not optimal.
With Georges diagram, I seem to remember his using a 50K log pot. And 100 ohm current limiting resistors. Mine is a dual mono copy of his, so all the resistances are doubled. It measures around 20K. Maybe 17K with the control pots in mid range.
I think that if you build a version of George's posted design and use a stereo 100K log pots to run all four LDR's the resulting impdeance will be somehere between 50 -100K If it is still too low, either a 250K or a pair of 250K paralleled (125K) might get it to about a 100K.
As a volume pot this should work fine. It would KILL a passive line stage. Most passive units are closer to 10K. This is a value that preserves dynamics and still allows coupling with most sources.
I have been listening to mine today. Been swapping out with an active last couple weeks. And working on the active to get the clarity and pureness of the LS clone. Still trying to get there.
Next is replace the 25K PEC pots in the active with a LS clone voume control. Got to dig up the correct pots to control the LED's. If it improves over the PEC pots it will prove that it is the LDR elements and not just the active vs passive design that makes it so special.
George
I am lost
Tom,
So you paralleled the resistive elements and series connected the LED's. I can see where this should get it back to close to a single LDR.
In fact in might make tracking a little better. From my matching about 30 with at 3 current level the offsets and slopes vary even with the matched grade.
But when I checked most stereo log pots the tracking on them is awful also. I suspect since audio loudness is a log function, looking at linear deviations is not a big factor.
I wired up a balanced LDR, matching up closely tthe four series and four shunt elements. But never got around to using it. Is there any advantage sonically to wiring it like yours?
I had planned on pulling it apart and making two stereo single ended controls out of it. But could do this too.
George
Tom,
So you paralleled the resistive elements and series connected the LED's. I can see where this should get it back to close to a single LDR.
In fact in might make tracking a little better. From my matching about 30 with at 3 current level the offsets and slopes vary even with the matched grade.
But when I checked most stereo log pots the tracking on them is awful also. I suspect since audio loudness is a log function, looking at linear deviations is not a big factor.
I wired up a balanced LDR, matching up closely tthe four series and four shunt elements. But never got around to using it. Is there any advantage sonically to wiring it like yours?
I had planned on pulling it apart and making two stereo single ended controls out of it. But could do this too.
George
George, the only benefits I see are these. Parallel R operation gives you a more friendly impedance as it halves it. It will average any resistive deviations, which should give more accurate tracking. It might also reduce the inherent distortion of the LDR by halving the signal current appearing across each unit (but I think that distortion is a product of voltage, not current?).
That is it
That is how I understood the matching. Tom did a bettter job than me, mine were done at three current levels. Still the resulting points allowed a match that tracks very nicely.
Georgehifi does a match job more like Tom's for the commercial Lightspeed. I do not need it exactly, mine has dual mono pots to make up for any small channel balance issues.
Is there still a source out there for the graded LDR's?
Badge said:
That is how I understood the matching. Tom did a bettter job than me, mine were done at three current levels. Still the resulting points allowed a match that tracks very nicely.
Georgehifi does a match job more like Tom's for the commercial Lightspeed. I do not need it exactly, mine has dual mono pots to make up for any small channel balance issues.
Is there still a source out there for the graded LDR's?
Reg supply, current limiting resistor
When I checked mine, the voltage across the LED was fairly constant when all 30 were connected to a regulated supply and dropping resistor.
I used three different resistors to get all the diodes to about the same levels. The thing measured was the value of the resistive element. Matching the LED sections might do something, but it may not have anything to do with the value of the resistive element.
I just rechecked mine, with it set up with dual stereo pots for dual mono voltage control it is higher than I remember.
The lowest value is 22.3 and 22.7. The highest is wide open at 29.3K and 30.1K. The values were closer when new, but the LDR's have at least a couple thousand hours on them.
Is increasing resistance a function of LED dimming or cadium losing a little sensitivity? I remeber reading about 20% lower overall resistances when brand new. 22K down to 17 -18K
Or it may be my supply has drifted.
When I checked mine, the voltage across the LED was fairly constant when all 30 were connected to a regulated supply and dropping resistor.
I used three different resistors to get all the diodes to about the same levels. The thing measured was the value of the resistive element. Matching the LED sections might do something, but it may not have anything to do with the value of the resistive element.
I just rechecked mine, with it set up with dual stereo pots for dual mono voltage control it is higher than I remember.
The lowest value is 22.3 and 22.7. The highest is wide open at 29.3K and 30.1K. The values were closer when new, but the LDR's have at least a couple thousand hours on them.
Is increasing resistance a function of LED dimming or cadium losing a little sensitivity? I remeber reading about 20% lower overall resistances when brand new. 22K down to 17 -18K
Or it may be my supply has drifted.