I've read the first 37 pages so far. 37 out of 302... pardon me if I've missed the answer to my question. If it's been asked before?
Hopefully it's not to OT...
How difficult would it be to controll the optocoupler in the 5Mohm to 5kohm range?
It looks to be some very small currents in this range.
Hopefully it's not to OT...
How difficult would it be to controll the optocoupler in the 5Mohm to 5kohm range?
It looks to be some very small currents in this range.
Markus,
Its not to bad controlling the LDR up into a half of a megohm but past that and it gets pretty silly. 5k is easy and very stable. The lower the resistance of the LDR then the more stable it is.
If you have read the first 5 pages you know all you need to build the Lightspeed.
Uriah
Its not to bad controlling the LDR up into a half of a megohm but past that and it gets pretty silly. 5k is easy and very stable. The lower the resistance of the LDR then the more stable it is.
If you have read the first 5 pages you know all you need to build the Lightspeed.
Uriah
Yeah, building the original is a no-brainer. However a 7kohm semi constant impedance "pot" feels a little low and not perfectly suited for my application? So, I'm toying with alternatives... For example, the nsl-32sr3 might be a better solution if I opt for a 25kohm constant impedance 60dB "attenuator".
I could probably make a +47kohm unit with the nsl-32sr2 if I allow the impedance to vary (grow). It all comes down to personal taste and application.
Problems driving the optocouplers close to the end points will of course make things a little more tricky. Knowledge is the key here.
However there is a huge gap in my knowledgebank!
I don't know what a typical ouput impedance is for a cd or dac?
Or a typical input inpedance for a pre-amp?
I haven't been doing this in a long time so I've forgotten a lot of the basic stuff.
I could probably make a +47kohm unit with the nsl-32sr2 if I allow the impedance to vary (grow). It all comes down to personal taste and application.
Problems driving the optocouplers close to the end points will of course make things a little more tricky. Knowledge is the key here.
However there is a huge gap in my knowledgebank!
I don't know what a typical ouput impedance is for a cd or dac?
Or a typical input inpedance for a pre-amp?
I haven't been doing this in a long time so I've forgotten a lot of the basic stuff.
A lot of the time you can find the output impedance of your source or input of your amp by checking the users manual. My Azur 640C cdp says it has <50R output. Didnt know that til last week and figured it was much higher. But anyway, I found it on the user manual. With that output impedance I can use just about any impedance on my passive pre as I want. With an average Lightspeed we have easily driven a amp with 47k input impedance. Lower than that may work depending out output of your source but might benefit from a buffer. There are a lot of good ones out there and I think in the first 37 pages you read George probably showed an easy tube buffer he recommends if you actually need a buffer. If your amp is 100k which is very common then this is easy and you can get away with almost anything without needing a buffer.
The NSL-32SR3 is fine but harder to match as they have a huge range of resistance. Its nice to have a smaller range when you are trying to match and who uses a megohm pot anyway? So if you want a 25k pot then you can make one by tweaking the circuit a little. Myself and a few others have suggested sticking a 100R trimmer in series with the 5VDC from the supply then on to the control pot. You can get your 25k this way.
You are not limited with the SR2 version. Trust me. Its a good way to go. With good matching you can have a pretty stable impedance pot.
Glad you are on this thread with questions. I like helping and the thread needs a boost of activity, so ask away.
Uriah
The NSL-32SR3 is fine but harder to match as they have a huge range of resistance. Its nice to have a smaller range when you are trying to match and who uses a megohm pot anyway? So if you want a 25k pot then you can make one by tweaking the circuit a little. Myself and a few others have suggested sticking a 100R trimmer in series with the 5VDC from the supply then on to the control pot. You can get your 25k this way.
You are not limited with the SR2 version. Trust me. Its a good way to go. With good matching you can have a pretty stable impedance pot.
Glad you are on this thread with questions. I like helping and the thread needs a boost of activity, so ask away.
Uriah
I've been trying to figure out a smart follow up question to answer all my needs but alas those perfect questions are most often thought up in hindsight when you know the answer.
I'll try to explain my situation and maybe we can find a nice solution.
I will be using a active tube pre-amp and monoblocks.
The preamp has roughly a 1Mohm input impedance I think and an output impedance of 85ohm.
I haven't got a clue about the CD or monoblocks?
So, I want to build the preamp as a generic unit that'll work with any source and load you can find.
I have two questions that will have a big impact on the following solution.
1. How much should the input impedance be? Lowest/highest?
2. Does it really matter if the input impedance isn't constant? what does it matter if it starts at 47kohm and rises to 470kohm? (The 47 is just a nice middle of the road kind of number)
The original is said to mimic a 7kohm pot fairly well. I'm worried that 7kohm is to low.
A single shunt cirquit will result in lower distortion figures according to the Silonex website. Why not make it simpler and just go for a single shunt solution?
With a shunt only cirquit you can pick the low end of the input impedance and it'll grow as you turn the knob.
A pi-cirquit will make for a constant input impedance but it'll put the LDR in series with the signal and it'll require a very high resistance to make for a good input impedance. So high that it becomes silly really, definately not practical in any way. 25kohm -> 25Mohm for a 60dB attenuation.
I guess some of you will think I'm daft for sticking with an active pre but I have my reasons. If you feel this thread should be passive only I'll start my own thread but imho it's still on topic since we're discussing a passive attenuation with LDR's.
I'll try to explain my situation and maybe we can find a nice solution.
I will be using a active tube pre-amp and monoblocks.
The preamp has roughly a 1Mohm input impedance I think and an output impedance of 85ohm.
I haven't got a clue about the CD or monoblocks?
So, I want to build the preamp as a generic unit that'll work with any source and load you can find.
I have two questions that will have a big impact on the following solution.
1. How much should the input impedance be? Lowest/highest?
2. Does it really matter if the input impedance isn't constant? what does it matter if it starts at 47kohm and rises to 470kohm? (The 47 is just a nice middle of the road kind of number)
The original is said to mimic a 7kohm pot fairly well. I'm worried that 7kohm is to low.
A single shunt cirquit will result in lower distortion figures according to the Silonex website. Why not make it simpler and just go for a single shunt solution?
With a shunt only cirquit you can pick the low end of the input impedance and it'll grow as you turn the knob.
A pi-cirquit will make for a constant input impedance but it'll put the LDR in series with the signal and it'll require a very high resistance to make for a good input impedance. So high that it becomes silly really, definately not practical in any way. 25kohm -> 25Mohm for a 60dB attenuation.
I guess some of you will think I'm daft for sticking with an active pre but I have my reasons. If you feel this thread should be passive only I'll start my own thread but imho it's still on topic since we're discussing a passive attenuation with LDR's.
Last edited:
Markus,
I think you are wrong about your 25k-25M. The LDR CAN do that but it doesnt HAVE TO. In my LDR attenuator right now I turn the control pot, then the LDRs respond and move from 50R up to 15k. No problem.
Your preamp will have <100R output and massive input. Good. This is what you want. Stick the LDR attenuator on either side of it. You are fine. I would put it on the amp side but with try to see what you like. This LDR attenuator is just a pot. A fancy pot. A wonderful sounding pot.... but just a pot.
Dont want 7k? To low? Read my previous post. Stick the 100R trimmer in series with the power to the control pot and dial it to around 10R or maybe 30R. You will have a 47k pot if thats what you want.
The input impedance will not move around like you mentioned. With your tube pre you are going to be fine with any Lightspeed iteration at almost any impedance. Really and honestly you nearly have the perfect set up for using this and trust me 7k will be fine if thats all you get from it. 7k will leave you with a very detailed sound. That mixed with tubes will be heaven. Remember you can use that trimmer pot to get a higher impedance if you want.
Uriah
Oh, forgot. The dual LDR attenuator sounds better and has even impedance where the other you mentioned does not.
I think you are wrong about your 25k-25M. The LDR CAN do that but it doesnt HAVE TO. In my LDR attenuator right now I turn the control pot, then the LDRs respond and move from 50R up to 15k. No problem.
Your preamp will have <100R output and massive input. Good. This is what you want. Stick the LDR attenuator on either side of it. You are fine. I would put it on the amp side but with try to see what you like. This LDR attenuator is just a pot. A fancy pot. A wonderful sounding pot.... but just a pot.
Dont want 7k? To low? Read my previous post. Stick the 100R trimmer in series with the power to the control pot and dial it to around 10R or maybe 30R. You will have a 47k pot if thats what you want.
The input impedance will not move around like you mentioned. With your tube pre you are going to be fine with any Lightspeed iteration at almost any impedance. Really and honestly you nearly have the perfect set up for using this and trust me 7k will be fine if thats all you get from it. 7k will leave you with a very detailed sound. That mixed with tubes will be heaven. Remember you can use that trimmer pot to get a higher impedance if you want.
Uriah
Oh, forgot. The dual LDR attenuator sounds better and has even impedance where the other you mentioned does not.
Also the dual (series&shunt) LDR has a lower distortion at the null point (which happens to be around the average listening position on the control) than just a shunt LDR, as the series/shunt LDR's null out each other distortion. Not that the very small distortion <.1% at either end of the volume range is objectionable, it gives a tube 2hd type of distortion, which is pleasant to the ear.
Tube amps with minimal or no negative feedback (which all with ears prefer over high feedback designs) give much more than this, and the speakers themselves, well they are up into the percent figure.
Just build it as I designed it and listen, your ears will be the judge, and forget the analyzing, you CDP is 85ohms this is fine, and your power amp will probably be more than 47k which is also fine, unless it's a chip amp or a chopper amp (digital amp) then you may need a buffer after the Lightspeed.
Cheers George
hornperfect, you seem to have forgot you have been warned not to make trouble here
Your intentions are obviously not good
Its got to stop NOW
no more warnings, you may be facing bintime now, a long one
http://www.diyaudio.com/forums/anal...-photoresistor-attenuator-preamp-general.html
Your intentions are obviously not good
Its got to stop NOW
no more warnings, you may be facing bintime now, a long onehttp://www.diyaudio.com/forums/anal...-photoresistor-attenuator-preamp-general.html
Thanks for the response to my questions.
I realize i still have some reading to do before I get everyting sorted out.
From your replies I gather that the input impedance of the pot doesn't really matter all that much? And as such, any fluctuations due to nonlinearity and so on doesn't really matter? If it doesn't matter if the circuit shows 7kohm, 15kohm or 47kohm then my questions are really off target.
georgehifi> I'm not trying to be difficult but could you please comment on the distortion figures? Looking at the diagrams at the silonex website the series/shunt cirquit show upward 4% distortion. The shunt cirquit has a THD+N(%) too low to measure with their measuring set-up.
To my eyes this would indicate that the series/shuncircuit has way more noise/distortion than the shunt circuit?
Am I reading it wrong or what? Are their measurements faulty?
udailey> I get the cirquit but I think I'll have to do some experimenting before I can truly understand how the resistance curves interact. Need some data to plot it out.
I realize i still have some reading to do before I get everyting sorted out.
From your replies I gather that the input impedance of the pot doesn't really matter all that much? And as such, any fluctuations due to nonlinearity and so on doesn't really matter? If it doesn't matter if the circuit shows 7kohm, 15kohm or 47kohm then my questions are really off target.
georgehifi> I'm not trying to be difficult but could you please comment on the distortion figures? Looking at the diagrams at the silonex website the series/shunt cirquit show upward 4% distortion. The shunt cirquit has a THD+N(%) too low to measure with their measuring set-up.
To my eyes this would indicate that the series/shuncircuit has way more noise/distortion than the shunt circuit?
Am I reading it wrong or what? Are their measurements faulty?
udailey> I get the cirquit but I think I'll have to do some experimenting before I can truly understand how the resistance curves interact. Need some data to plot it out.
Your cdp, almost any cdp, will drive the Lightspeed easily. 7, 15, 47, whatever. To the CDP the higher the better but then the higher the Lightspeed the higher the input of amp must be to work well with the whole situation. You have an advantage with that tubed pre. So you can stick the lightspeed on the input side of the preamp and this will work perfectly or you can stick it on the output side and depending on the amps input impedance things will work well or not but with 85Ohm out of the pre you are probably going to be fine with almost any value input on your amp.
The distortion figures should be looked at more closely. They are in reference to the voltage flowing through the resistor NOT the LED of the LDR. So if your CDP puts out 2V then you get the distortion at 2V on that graph. That graph looks pretty good for anything around 2-4V if I recall then on either side it starts to increase. If your cdp puts out 10V you are going to get a lot of distortion, but trust me it doesnt.
The shunt LDR is not going to add any distortion as the signal passes through this resistor to ground and never gets to your amp.
The only reason why I sound ambivalent about impedance of Lightspeed is because of your particular situation with that preamp. In other situations 7k, 15k, 47k would matter more but with your pre you have a lot of leeway to work with.
Uriah
The distortion figures should be looked at more closely. They are in reference to the voltage flowing through the resistor NOT the LED of the LDR. So if your CDP puts out 2V then you get the distortion at 2V on that graph. That graph looks pretty good for anything around 2-4V if I recall then on either side it starts to increase. If your cdp puts out 10V you are going to get a lot of distortion, but trust me it doesnt.
The shunt LDR is not going to add any distortion as the signal passes through this resistor to ground and never gets to your amp.
The only reason why I sound ambivalent about impedance of Lightspeed is because of your particular situation with that preamp. In other situations 7k, 15k, 47k would matter more but with your pre you have a lot of leeway to work with.
Uriah
They used to have many graphs, dig hard, or maybe they have removed them. As I said look series/shunt for the graph that gives the disortion at the null point, also the other distortion graphs remember the average that comes out of a cdp is 1v the max is 2v (Cannon Shot 1812 Telarc).
Cheers George
I have found a Vout DAC with 2Vrms output that works wonderfully well directly driving headphones. No output caps, no headphone amp needed. I've tried it on Senn HD650, AKG701 (slightly lower volume but loud enough for my listening), Grado HF2 cheap Sonys - from 32ohm upto 300ohm impedance. I wanted to use a LDR volume control on the output - I reckon only a series LDr will work. Any special considerations I need to consider? Is it possible to get a reasonable channel match across the full volume range needed or am I better off going with a stepped attenuator to avoid this issue?
No, as I said it doesn't need a headphone amp - the DAC drives headphones directly so the impedances in & out of the LDR are different to what is the norm here!
HD650's are a 300ohm load, most low impedance dacs say 50ohms or less will drive these if the 2v or more ouput voltage is loud enough for you, 50ohms into 300ohms is a 1 to 6 ratio which is OK, the moment you stick a passive volume controling device beteen them this 1 to 6 ratio will go the other way like ?kohms to 300. You will need a buffer after the passive to bring it back down to the 1 to 6 ratio or better like 1 to 10.
Cheers George
Cheers George
One of the ESS DACs ES9022 - also used in Nuforce uDac which drives headphones directly & uses a vol control (no op-amps used) see here: FIRST IMPRESSIONS: Nuforce uDAC USB DAC AMP with line out and S/PDIF out - Head-Fi: Covering Headphones, Earphones and Portable Audio
This ESS DAC can do 24/192 but in the NuForce, it's limited by the use of PCM2707 USB receiver both in speed & in high jitter. I have a 24/192 USB low jitter transport that I wanted to use with the DAC
This ESS DAC can do 24/192 but in the NuForce, it's limited by the use of PCM2707 USB receiver both in speed & in high jitter. I have a 24/192 USB low jitter transport that I wanted to use with the DAC
Last edited:
AKG701s are 32 ohm & the DAC drives these to a loudness which is enough for my tastes i.e. below ear bleeding. I have seen it quoted that the DAC is happy driving headphones down to 16ohm.
I have driven my Sony MDR-XD200 (70ohm) with a pot volume control on the DAC's output, so the concept is proven - I just wanted a higher quality vol control with the LDR but is it out of the question to expect close LDR matching all along the vol range for a good channel to channel balance?
Last edited: