50mW on the resistor and 100mW on the LED. More than that is not so great on either but it seems that the resistor can handle more and that the datasheet is being over protective here. What if you use an LDR in series as ON/OFF and a relay shunted to ground? The relay is never active in the signal path until you are NOT using that signal.
Uriah
Uriah
Remote control of Lightspeed configuration
Hi Marcus,
The current draw will depend on the chosen current range for the LEDs. When set for maximum safe LED current the total current draw will be around 100ma.
For series drive of LEDS in the balanced configuration the supply voltage will need to be between 21 volts and 24 volts to feed the on-board 18 volt regulator. Allowing for ripple voltage your proposed supply should be OK unless you suffer from low mains voltage in your house. This is worth checking and allowing for.
The 18v reg and the bias shunt regs will give a reasonably low noise level. The power system was designed this way to keep cost at a reasonable level. If you want really low noise you can fit my low noise high performance regulators in place of the on-board three terminal regulator and the TL431 shunt bias regulators.
Regards
Paul
Hi Marcus,
The current draw will depend on the chosen current range for the LEDs. When set for maximum safe LED current the total current draw will be around 100ma.
For series drive of LEDS in the balanced configuration the supply voltage will need to be between 21 volts and 24 volts to feed the on-board 18 volt regulator. Allowing for ripple voltage your proposed supply should be OK unless you suffer from low mains voltage in your house. This is worth checking and allowing for.
The 18v reg and the bias shunt regs will give a reasonably low noise level. The power system was designed this way to keep cost at a reasonable level. If you want really low noise you can fit my low noise high performance regulators in place of the on-board three terminal regulator and the TL431 shunt bias regulators.
Regards
Paul
Not regulating the psu will be cheaper...
Last sim I did resulted in 15mV ripple (peak to peak) when regulated. Adding more caps would probably make it even better.
But I'll run the numbers with 100mA and compare regulated vs. unregulated. Being somewhat of a perfectionist I'll probably go regulated just because I can.
Last sim I did resulted in 15mV ripple (peak to peak) when regulated. Adding more caps would probably make it even better.
But I'll run the numbers with 100mA and compare regulated vs. unregulated. Being somewhat of a perfectionist I'll probably go regulated just because I can.
I have Lithium Ion batteries I can use, and no one "golden ears" can reliably detect the difference between them and the standard mains regulated supply that I do with the production Lightspeed Attenuator, and this is on a very high rez system.
Even Sam Tellig "golden ears from Stereophile" said the same, as he bought the batteries as well, thinking they had to be better (pure dc) than the regulated mains but no he said if there is a difference it's miniscule and not detectable.
Cheers George
Even Sam Tellig "golden ears from Stereophile" said the same, as he bought the batteries as well, thinking they had to be better (pure dc) than the regulated mains but no he said if there is a difference it's miniscule and not detectable.
Cheers George
LDR power regulation
I am not an expert to comment about the technicality of Bal,rugulated v/s unregulated PS.
I tried 4 different combination and auditioned infront of some people does not know about audiophile sound and with some piano and saxaphoen experts.
1 Tried from wall wart 12 VDC 1A to 7805 regulation
2 From walwart 18 vdc to 7812 to 7805
3 from a Ct transformer regulated to 12 VDC 300m A to LM 317T ( R1 220+ Radioshack trim resister)
4 from a Ct transformer 12 VDC 300mA to LM 317 ( usinf R1 220+ 1.5k) 9.78V and to 7805 . ( fixed resister is far better than trim at least for LDR passive unless you go for very good ones ). So far I have seen varaible is not my cup of tea for LDR Passive
Combination was No 3 was not for me. highs shooting up and unnatural and all my friend agreed to that.
The best smooth sound was No#1 and #2
Crystel clean sound a little shy of bass was # 4.
So my assumption is that GEORGE IS ALWAYS RIGHT.Simple is beautiful and listenable as long as you have the right matched LDRs. To my ears LDR like 7805. I have built one ( combination # 4 very clean and clear)that directly goes to input of the amp and plan to add Uria's or pauls's bal board on the same case to compare. Short of money. Will do that after Fed-Tax payment.
The idea of remote is good. Waiting for that solution
I am not an expert to comment about the technicality of Bal,rugulated v/s unregulated PS.
I tried 4 different combination and auditioned infront of some people does not know about audiophile sound and with some piano and saxaphoen experts.
1 Tried from wall wart 12 VDC 1A to 7805 regulation
2 From walwart 18 vdc to 7812 to 7805
3 from a Ct transformer regulated to 12 VDC 300m A to LM 317T ( R1 220+ Radioshack trim resister)
4 from a Ct transformer 12 VDC 300mA to LM 317 ( usinf R1 220+ 1.5k) 9.78V and to 7805 . ( fixed resister is far better than trim at least for LDR passive unless you go for very good ones ). So far I have seen varaible is not my cup of tea for LDR Passive
Combination was No 3 was not for me. highs shooting up and unnatural and all my friend agreed to that.
The best smooth sound was No#1 and #2
Crystel clean sound a little shy of bass was # 4.
So my assumption is that GEORGE IS ALWAYS RIGHT.Simple is beautiful and listenable as long as you have the right matched LDRs. To my ears LDR like 7805. I have built one ( combination # 4 very clean and clear)that directly goes to input of the amp and plan to add Uria's or pauls's bal board on the same case to compare. Short of money. Will do that after Fed-Tax payment.
The idea of remote is good. Waiting for that solution
Just wondering...
George,
I have a single-ended tube amp with ~220kO input impedance, fed by the 200 O analog output of my CD player. I am wondering if the lightspeed would work for me? Thanks for the reply.
Phil
George,
I have a single-ended tube amp with ~220kO input impedance, fed by the 200 O analog output of my CD player. I am wondering if the lightspeed would work for me? Thanks for the reply.
Phil
Hi Phil, seeing you've got 220k at the poweramp the cd 's output of 200ohm should be OK.
I have had customers with 200ohm tube phono stages running into the Lightspeed and then into 100k amps and they say it's the best their phono has ever sounded.
Cheers George
I have had customers with 200ohm tube phono stages running into the Lightspeed and then into 100k amps and they say it's the best their phono has ever sounded.
Cheers George
Thanks George
....for sharing your ideas and your experiences with the DIY community. I really appreciate this.
Now, I received my matched pairs from Uriah and some random LDRs.
With 4 random LDRs, I just tried a MkII test setup.
The total current from the 5V source to all 4 LEDs was only 0,2mA (pot 12 o´clock) to 20mA (pot full up or down). So each LED is driven by not more than 10mA. Hope this is fine.
As a first impression the setup worked quite nicely with my 300R source and only (!) 10k load (Linkwitz Orion active crossover).
With "medium-volume" recordings, for normal listening, the pot was set 9 o´clock.
I could also achieve "normal listening level" by turning the pot to the recommended 12 o´clock position and reducing the supply voltage from 5V to 3V. Is this reducing THD further?
I´m really looking forward doing more testing with parallel Rs and LDRs, also trying the effect of lowering the total impedance from 7k to 5k.
Really interesting.
Cheers, Frank
....for sharing your ideas and your experiences with the DIY community. I really appreciate this.
Now, I received my matched pairs from Uriah and some random LDRs.
With 4 random LDRs, I just tried a MkII test setup.
The total current from the 5V source to all 4 LEDs was only 0,2mA (pot 12 o´clock) to 20mA (pot full up or down). So each LED is driven by not more than 10mA. Hope this is fine.
As a first impression the setup worked quite nicely with my 300R source and only (!) 10k load (Linkwitz Orion active crossover).
With "medium-volume" recordings, for normal listening, the pot was set 9 o´clock.
I could also achieve "normal listening level" by turning the pot to the recommended 12 o´clock position and reducing the supply voltage from 5V to 3V. Is this reducing THD further?
I´m really looking forward doing more testing with parallel Rs and LDRs, also trying the effect of lowering the total impedance from 7k to 5k.
Really interesting.
Cheers, Frank
Hi,
Is there an explanation for your observations ? Could this be ground loop problem ?
I presume that noise sound be lower with items 3 and 4 . Have you scoped the power supply ?
Thanks for sharing .
kp93300
There is no need nor would you want to ground the supply if built as I have described.
In the production Lightspeed Attenuator, I isolate the regulated power supply and all it's components from the chassis and ground plane of the interconnects and RCA's.
So all you have is a non grounded floating powersupply that has no reference to chassis or RCA grounds, this way the only connection of the regulated power supply to the LDR's and signal RCA input or output is via the light of the led's themselves, which as we know light is non conductive so there is no way powersupply induced hum or noise can be transmitted to the audio signal, unless the regulated supply itself becomes faulty or the mains drops so low that the regulators stop regulating.
Cheers George
In the production Lightspeed Attenuator, I isolate the regulated power supply and all it's components from the chassis and ground plane of the interconnects and RCA's.
So all you have is a non grounded floating powersupply that has no reference to chassis or RCA grounds, this way the only connection of the regulated power supply to the LDR's and signal RCA input or output is via the light of the led's themselves, which as we know light is non conductive so there is no way powersupply induced hum or noise can be transmitted to the audio signal, unless the regulated supply itself becomes faulty or the mains drops so low that the regulators stop regulating.
Cheers George
Just a slightly O/T question for UK LDR builders.. does anyone know a source for good quality enclosures fo it? I have already looked at the usual suspects like modushop etc. I was hoping to get one like the little tripath amps use that are always for sale on ebay.
Thanks, Steve
Thanks, Steve
Depending on price Diy HiFi Supply | Take Control has some super looking enclosures. So does that company that sells the Goldpoint attenuators that are slaughtered by the Lightspeed 🙂 They both have nice boxes but neither are especially cheap.
I like using a 2U rack enclosure because I can fiddle around with different supplies, buffers, pre amps since there is so much room in there.
Uriah
I like using a 2U rack enclosure because I can fiddle around with different supplies, buffers, pre amps since there is so much room in there.
Uriah
Drool indeed! that is exactly the sort of thing I am looking for, but a little pricey when shipping to the UK is included. Tempting though, very tempting.
Steve
Hi fbee
with my method you can achieve any input impedance you want between 1k and 1 Meg! And this with unmatched devices and with full remote control! Just seek a little bit in this thread. I have posted some info around Nov 2008!
After my holidays I will make a PCB in order to get some professional results with 5 LDR-driven inputs and an 10k attenuator circuit.
with my method you can achieve any input impedance you want between 1k and 1 Meg! And this with unmatched devices and with full remote control! Just seek a little bit in this thread. I have posted some info around Nov 2008!
After my holidays I will make a PCB in order to get some professional results with 5 LDR-driven inputs and an 10k attenuator circuit.
What are you doing? Reading my mind LOL
Been working on the LDR switcher and will have it done around the same time. Have been selling Lighter Note which is able to select impedance as well. Fun stuff. Different impedances really sound very different dont they?
Whats your favorite Tolu? I prefer around 6k total resistance.
Uriah
Been working on the LDR switcher and will have it done around the same time. Have been selling Lighter Note which is able to select impedance as well. Fun stuff. Different impedances really sound very different dont they?
Whats your favorite Tolu? I prefer around 6k total resistance.
Uriah
Matching LDRs
Hi all,
I've been playing around trying to match these LDRs on and off for months.
Good fun, but so far all I have determined I don't have a suitable method that produces accurate and repeatable results. I suspect I am being to fussy. In my batch of 50 LDRs I am surprised at the large variability between individual LDRs and the sensitivity to temperature.
How many current settings are you guys using to match these LDRs?
Once the data is collected in a spreadsheet has anyone developed a statical method of analyzing which are a close match or are poeple just eyeballing the figures.
Are you matching over the entire range or weighing the results for a close match over the "normal" volume setting.
As a matter of interest, has anyone ever seen 2 identical LDRs?
regards
Hi all,
I've been playing around trying to match these LDRs on and off for months.
Good fun, but so far all I have determined I don't have a suitable method that produces accurate and repeatable results. I suspect I am being to fussy. In my batch of 50 LDRs I am surprised at the large variability between individual LDRs and the sensitivity to temperature.
How many current settings are you guys using to match these LDRs?
Once the data is collected in a spreadsheet has anyone developed a statical method of analyzing which are a close match or are poeple just eyeballing the figures.
Are you matching over the entire range or weighing the results for a close match over the "normal" volume setting.
As a matter of interest, has anyone ever seen 2 identical LDRs?
regards
Greg if you quad (for a good progressive logarithmic volume increase feel) match them at 2.5-5-10-15-20 mA and then you still have to fine calibrate after that with a scope and frequency generator then you should be able to get within +-1 or 2db. Which is still better than a Blue Alps pot.
I have never seen two identical LDR's but if you have bought as I do NSL32SR2S "sorted" and ask them from the same "batch" (A B C D E F or G) as I do, you will achieve the figures above.
Cheers George
I have never seen two identical LDR's but if you have bought as I do NSL32SR2S "sorted" and ask them from the same "batch" (A B C D E F or G) as I do, you will achieve the figures above.
Cheers George
Thanks George.
Looks like I need a bit more testing. I am currently using 330, 10k, 20k, 39k, 56k, 100k which, combined with a test voltage of only 4.5v, gives a lower spread of currents with a profile pretty much opposite to your suggestions.
regards
Looks like I need a bit more testing. I am currently using 330, 10k, 20k, 39k, 56k, 100k which, combined with a test voltage of only 4.5v, gives a lower spread of currents with a profile pretty much opposite to your suggestions.
regards