Pot applications
Yes. 7K is a quite fine value for between two amplifiers.
On the other hand it would be nice to be able to put them everywhere... For example between 2k to 250k units in active crossovers.
On each input of 6 channel amps for gain-matching (making the load 7k/3 because they're in parallell. And so on...
I'm thinking that if they don't drift much, then very exact attenuators could be built.
To make it really messy but possible to fine-calibrate:
a 24 or why not >100-position switch with a trimpot at each position replaces the 100kohm pot.
Each step would be fine tuned for resistance in the series and shunt element, and the channels could have better matching than with 1% tolerance resistor ladder-attenuators...
A lot of trimming, but the leds and the LDR would not have to be matched at all, right?
What do you think? Are there reliable LDRs with higher resistance than the ones your using, for such a circuit?
Yes. 7K is a quite fine value for between two amplifiers.
On the other hand it would be nice to be able to put them everywhere... For example between 2k to 250k units in active crossovers.
On each input of 6 channel amps for gain-matching (making the load 7k/3 because they're in parallell. And so on...
I'm thinking that if they don't drift much, then very exact attenuators could be built.
To make it really messy but possible to fine-calibrate:
a 24 or why not >100-position switch with a trimpot at each position replaces the 100kohm pot.
Each step would be fine tuned for resistance in the series and shunt element, and the channels could have better matching than with 1% tolerance resistor ladder-attenuators...
A lot of trimming, but the leds and the LDR would not have to be matched at all, right?
What do you think? Are there reliable LDRs with higher resistance than the ones your using, for such a circuit?
Changed value of voltage control pot
At first I was using two stereo 50K linear pots. Measured around 8K. The sonics were great, but the volume control for low level listening was very touchy.
Replaced the two 50K pots with a pair of stereo 100K linears. Now the impedances measure about 25k. At first I did not like the resulting sound as much. Semed to color the sonics and slow things down.
Swapping back and forth, decided the sound was not worse, just different. The 100K pots allow great low level control. Since mine uses two controls. the series resistors are now 200 ohms.
I get a little different measurements compared to a real Lightspeed. The Lightspeed uses a single stereo pot and it is a 100K log. But mine tracks real well and I like the taper.
I have used mono volumes controls since 1989. It takes a little time to get used to. Also if you wire the source selector with a mute inbetween every source it minimizes opening and closing the volume controls.
George
At first I was using two stereo 50K linear pots. Measured around 8K. The sonics were great, but the volume control for low level listening was very touchy.
Replaced the two 50K pots with a pair of stereo 100K linears. Now the impedances measure about 25k. At first I did not like the resulting sound as much. Semed to color the sonics and slow things down.
Swapping back and forth, decided the sound was not worse, just different. The 100K pots allow great low level control. Since mine uses two controls. the series resistors are now 200 ohms.
I get a little different measurements compared to a real Lightspeed. The Lightspeed uses a single stereo pot and it is a 100K log. But mine tracks real well and I like the taper.
I have used mono volumes controls since 1989. It takes a little time to get used to. Also if you wire the source selector with a mute inbetween every source it minimizes opening and closing the volume controls.
George
Reading this post with great interest. I am currently using DACT ct2 as passive attenuators with my aksa100n power amp. The power amp input impedance is 47K. Will I get better sound switching from DACT to lightspeed? All the nice amp kit that I come across have below 100K input impedance. The Simple Killer Amp's GB150 power input impedance is 27K. Aussie Amplifier NX150 is 34K.
Far out!! Just realise that those I listed plus George are all Aussie.
Must be something in the water
Sam
Far out!! Just realise that those I listed plus George are all Aussie.
Must be something in the water
Sam
SamL said:Reading this post with great interest. I am currently using DACT ct2 as passive attenuators with my aksa100n power amp. The power amp input impedance is 47K. Will I get better sound switching from DACT to lightspeed? All the nice amp kit that I come across have below 100K input impedance. The Simple Killer Amp's GB150 power input impedance is 27K. Aussie Amplifier NX150 is 34K.
Far out!! Just realise that those I listed plus George are all Aussie.
Must be something in the water
Sam
If you have a good strong source, (cd player or dac) there is no problems with the Lightspeed Attenuator driving the 47k Aska, and yes it will absolutely flog the Dact, done that comparsion and there is no comparison.
Cheers George
What's the noise performance of this attenuator like?
The reason I ask is that the spot where it makes the most sense in my system is right after I/V conversion in my DAC where the signal is only around 40mV RMS. The output impedance at this point is 50 ohms. A further 30db of amplification is needed to get near line level. Since I want to use valves for this gain, getting low impedance is going to cost something - a step down transformer or cathode follower.
The reason I ask is that the spot where it makes the most sense in my system is right after I/V conversion in my DAC where the signal is only around 40mV RMS. The output impedance at this point is 50 ohms. A further 30db of amplification is needed to get near line level. Since I want to use valves for this gain, getting low impedance is going to cost something - a step down transformer or cathode follower.
Hi guys!
I'm one of the lightspeed owner. I bought the unit on my recent trip to Sydney after reading about it in this forum n met the man himself. Nice chap. So far, here is my findings. First, I'm a very happy owner of the unit. Basically the good points has been discussed here in the forum. For me, this pre has the best dynamics compare to the other passive pre's(tvc,step). N it has speed too. Resolution is first class. Other than that, I've managed to play with power source other than the walwart. Coz getting a bit itchy. Wonder how is the sound if I push it further. But what the heck, no harm for trying. So what I did, I swapped the walwart with battery power supply. Initially I was a bit sceptical that it will change the sound. But OMG! The improvement is tremendous! Night n day difference! It is very quiet. The sound is even more sweeter n transparent n improve all the other positive areas as well. Dunno why this happened. Probably it makes the led quieter thus reducing the inteference in the optocoupler. Or maybe bcoz the grounding scheme is a common ground between the signal ground n psu ground n the battery makes the psu quieter. I only can speculate but the result speak for itself. So far, I'm very satisfied with this pre. So guys, start to heat up your soldering iron. Gud stuff George! Thumbs up for the man!
Cheers!
Eddy.
I'm one of the lightspeed owner. I bought the unit on my recent trip to Sydney after reading about it in this forum n met the man himself. Nice chap. So far, here is my findings. First, I'm a very happy owner of the unit. Basically the good points has been discussed here in the forum. For me, this pre has the best dynamics compare to the other passive pre's(tvc,step). N it has speed too. Resolution is first class. Other than that, I've managed to play with power source other than the walwart. Coz getting a bit itchy. Wonder how is the sound if I push it further. But what the heck, no harm for trying. So what I did, I swapped the walwart with battery power supply. Initially I was a bit sceptical that it will change the sound. But OMG! The improvement is tremendous! Night n day difference! It is very quiet. The sound is even more sweeter n transparent n improve all the other positive areas as well. Dunno why this happened. Probably it makes the led quieter thus reducing the inteference in the optocoupler. Or maybe bcoz the grounding scheme is a common ground between the signal ground n psu ground n the battery makes the psu quieter. I only can speculate but the result speak for itself. So far, I'm very satisfied with this pre. So guys, start to heat up your soldering iron. Gud stuff George! Thumbs up for the man!
Cheers!
Eddy.
Banned
Joined 2002
Hi,
finished my Lightspeed Pre yesterday. Im am also very satisfied with the results. Very good detail resulation and absolute noiseless. As far as I can say after short testing: Better than my 24 step attenuator.
G o o d w o r k G e o r g e !!! THX!
@flyboi
Instead of batteries you can also use two voltage regulators in series to reduce ripple to minumim. The ripple will be under noise
level of the regulator.
Arne
finished my Lightspeed Pre yesterday. Im am also very satisfied with the results. Very good detail resulation and absolute noiseless. As far as I can say after short testing: Better than my 24 step attenuator.
G o o d w o r k G e o r g e !!! THX!
@flyboi
Instead of batteries you can also use two voltage regulators in series to reduce ripple to minumim. The ripple will be under noise
level of the regulator.
Arne
Attachments
Banned
Joined 2002
Nice work there Arne, very compact, should fit into a matchbox nicely, hope your going to box it up with a power supply, as some of them the ends are not 100% light tight.
If you shine a torch on them some can read the torch light. When you've had some more listening give us a bit more critique on the sound?
Cheers George
If you shine a torch on them some can read the torch light. When you've had some more listening give us a bit more critique on the sound?
Cheers George
flyboi said:
I was waiting for this post.
After reading that changing control pot values changed the sound I
was pretty sure that the LED supply is ultra critical and this means
any junk getting into the LED will end up modulating the LDR and
hence your music.
The bestl solution is a CCS fed shunt reg that has huge PSR.
Don't bother with ultra finicky super reg type circuits because
the OP Z doesn't matter. What is required is ultra low noise
and ultra high rejection from wal wart supply.
Time to toss the three term regs.
Cheers
Terry
Ok I was sceptical about the power supply needing to be any better than it already is in the production Lightspeed.
So I will in the next few days conduct blind listening test with so called golden eared experts (x stands for nothing and spurt is a drip under pressure) to see if they can reliably detect the difference between the production Lightspeed power supply and a battery operation, I will post the results soon after the comparison.
Cheers George
So I will in the next few days conduct blind listening test with so called golden eared experts (x stands for nothing and spurt is a drip under pressure) to see if they can reliably detect the difference between the production Lightspeed power supply and a battery operation, I will post the results soon after the comparison.
Cheers George
Guys, does LED brightness in the LDR unit vary according to a change in voltage or a change in current?
Info on the Perkin Elmer site shows a life span of as high as (and likely higher than) 50,000 hours. The only change noted was an increase in cell resistance over time, the amount or rate of increase depending on low or high current use.
As to psu quality, any variation in voltage will create a variation in cell resistance ... not good for sound. Personally, I expected Flyboi's results.
Info on the Perkin Elmer site shows a life span of as high as (and likely higher than) 50,000 hours. The only change noted was an increase in cell resistance over time, the amount or rate of increase depending on low or high current use.
As to psu quality, any variation in voltage will create a variation in cell resistance ... not good for sound. Personally, I expected Flyboi's results.
Power supply
I do not think that the power supply is ultra critical here. The reason iis the LED is more like a tube heater. The voltage across the LED controls the output, and then the resistive element.
There is a long thermal lag in this system. I have cranked the voltages up and down while monitoring the resistance. It changes immediately and then slowing stabilizes. This is in the couple ohm range. Once set, the resistance is dead stable after a couple minutes. Passing audio through the LDR section does not effect the resistance, just current through the LED.
I think the long time period for resistance changes makes the attenuator less critical of power supply artifacts. I use a wallwart and dual LT1086 regs. Have not put my scope on it, but the voltage measures 5.01 volts to the regs all the time. Maybe I should and post pics!
Either way you look at it, with the variable resistance of the control pots, the actual impedance of the power supply gets masked. Even when cranked wide open, there is a 100 ohm resistor in series between the power supply and the leds. When the volume is attenuated, there are thousands of ohms inbetween.
Flyboi, I think what you are hearing is feeding the circuit with a different voltage. I assume the stock wallwart is a 12 volt supply. May be 12, if a charger it is 16-17. The battery is more like 13.5 volts if a 12 volt model.
When you replaced the wallwart, I assume also the power supply still goes to George's dual reg board prioir to hitting the actual circuits, it not, the impdeance could be much different from the stock 7K.
Also. pics of mine are shown on post 100 and 101.
George
I do not think that the power supply is ultra critical here. The reason iis the LED is more like a tube heater. The voltage across the LED controls the output, and then the resistive element.
There is a long thermal lag in this system. I have cranked the voltages up and down while monitoring the resistance. It changes immediately and then slowing stabilizes. This is in the couple ohm range. Once set, the resistance is dead stable after a couple minutes. Passing audio through the LDR section does not effect the resistance, just current through the LED.
I think the long time period for resistance changes makes the attenuator less critical of power supply artifacts. I use a wallwart and dual LT1086 regs. Have not put my scope on it, but the voltage measures 5.01 volts to the regs all the time. Maybe I should and post pics!
Either way you look at it, with the variable resistance of the control pots, the actual impedance of the power supply gets masked. Even when cranked wide open, there is a 100 ohm resistor in series between the power supply and the leds. When the volume is attenuated, there are thousands of ohms inbetween.
Flyboi, I think what you are hearing is feeding the circuit with a different voltage. I assume the stock wallwart is a 12 volt supply. May be 12, if a charger it is 16-17. The battery is more like 13.5 volts if a 12 volt model.
When you replaced the wallwart, I assume also the power supply still goes to George's dual reg board prioir to hitting the actual circuits, it not, the impdeance could be much different from the stock 7K.
Also. pics of mine are shown on post 100 and 101.
George
George, the LED power supply connects your audio signal to the wall. Some quantity of line disturbances definitely greater than zero will modulate the audio signal ... I don't see any other way of viewing the matter. Instead of insulting Flyboi by telling him he's hearing something other than what he says he's hearing---I assume he can differentiate between greater clarity, etc. and greater loudness (I mean, really)---open your analysis to considering that just maybe AC grunge is sufficiently pernicious to find its way into the audio signal.
Clarifications
I hope flyboi was not insulted. The reason I said that changing the voltage seen by the circuit altered the sound of mine significantly.
Changing the input voltage will effect the current levels and resulting impedances. When a 100K pot was used instead of 50K the sound slowed down and lost a little of the slam. With a 50K pot the circuit was 8 - 8.5k impedance. When the 100K linear pots were used it measures 22 - 25K depending on the pot position. A 8K passive pot will sound much different than a 25K.
As far as purity of supplied power, it has to have some effect. But I suspect it is less than an actual circuit. The time constant of the attenuators should smooth out any high frequency hash.
My opinions are based on actually building one and playing around with it. But mine is little different. The power supply is not connected to the audio circuit in mine. So there is no leakage of noise from the power supply into a shared ground buss. The LED circuit is totally independent electrically from the LDR's, so the two can either be connected or left floating.
Course if the two grounds are connected, replacing a fairly noisey two stage three pin reg setup with batteries should remove a level of hash and crud. It might sound even better with the signal ground completely disconnected from the audio ground.
I apologize if these comments offend anyone. I am just sharing my thoughts and short experiance with this really sweet setup.
Thanks again to George for sharing his experiances and thoughts.
George
I hope flyboi was not insulted. The reason I said that changing the voltage seen by the circuit altered the sound of mine significantly.
Changing the input voltage will effect the current levels and resulting impedances. When a 100K pot was used instead of 50K the sound slowed down and lost a little of the slam. With a 50K pot the circuit was 8 - 8.5k impedance. When the 100K linear pots were used it measures 22 - 25K depending on the pot position. A 8K passive pot will sound much different than a 25K.
As far as purity of supplied power, it has to have some effect. But I suspect it is less than an actual circuit. The time constant of the attenuators should smooth out any high frequency hash.
My opinions are based on actually building one and playing around with it. But mine is little different. The power supply is not connected to the audio circuit in mine. So there is no leakage of noise from the power supply into a shared ground buss. The LED circuit is totally independent electrically from the LDR's, so the two can either be connected or left floating.
Course if the two grounds are connected, replacing a fairly noisey two stage three pin reg setup with batteries should remove a level of hash and crud. It might sound even better with the signal ground completely disconnected from the audio ground.
I apologize if these comments offend anyone. I am just sharing my thoughts and short experiance with this really sweet setup.
Thanks again to George for sharing his experiances and thoughts.
George
There maybe another reason way Flyboi heard a difference, the 12vdc regulated wallwart that I supplied is Australian 240/250, in Singapore the mains is 220v if the threshold was too low for the wallwart to regulate properly his only regulation was comming from the second 5v regulator inside the Lightspeed itself, and wasn't getting the advantage of dual regulation. One must look at all options before drawing a conclusion.
Cheers George
Cheers George