UNO - which Idss ?
Come on Nuuk, you lazy bones
it's neither of both
Just read the marking on the JFET. Is it 2SK170GR or 2SK170 BL or 2SK170V ?
Or (if it's unreadable) pick up the volt-meter and measure the voltage across the 75 Ohm resistor (one which connects JFET's drain to +40 V supply rail) with no music playing.
It would help a lot to know this (here i can buy only type V - so knowing what you got is the only way for me to make it work as it should)
Thanks a lot !
Nuuk said:Juma, please choose one of the following statements:
This question is far too simple for me to bother with it.
This question is far too technical for me to answer it.
Carlos - HELP!
Come on Nuuk, you lazy bones
it's neither of both
Just read the marking on the JFET. Is it 2SK170GR or 2SK170 BL or 2SK170V ?
Or (if it's unreadable) pick up the volt-meter and measure the voltage across the 75 Ohm resistor (one which connects JFET's drain to +40 V supply rail) with no music playing.
It would help a lot to know this (here i can buy only type V - so knowing what you got is the only way for me to make it work as it should)
Thanks a lot !
Nuuk said:Juma,
I have taken time from my busy schedule to strip down my pre-amp and look at the Jfets. They are the 2SK170GR type.
Now somebody is very conspicuous by their absence! Where is the man from Lisboa?
Thanks a lot Nuuk !
That's really helpfull. It means that JFET works in penthode (saturation) regime and being of low Idss current type, is controlled only by Vgs voltage. Everything is all right as long as impedance of previous stage is low and the impedance of next stage is high enough (20 KOhm should be driven without a problem).
I'll build it and see what happens, but I would surely like to get rid of the coupling cap.
Thanks again Nuuk, you are a sport!
Cheers all !
powersupply for preamp
what is the optimal power supply for this preamp? (seemed like someone wanted to try double regulation etc... but my question is more fundamental than that)
i really have no idea how much filter capacitors the regulators need, so i copied nuuks power supply configuration for his buffer.
however i noticed that nuuk only used 4 diodes to regulate the power supply coupled with the grounds being shared early on, what is the motivation for this configuration? (i just used 4 diodes for regulation for each channel since i already started the circuit like this)
well i guess my bigger question is, what are the motivations for the different configurations such as the chipamp powersupply versus nuuks buffer power supply.
chipamp power supply with split gnds and 4 diodes for rectification
thanks,
what is the optimal power supply for this preamp? (seemed like someone wanted to try double regulation etc... but my question is more fundamental than that)
i really have no idea how much filter capacitors the regulators need, so i copied nuuks power supply configuration for his buffer.
An externally hosted image should be here but it was not working when we last tested it.
however i noticed that nuuk only used 4 diodes to regulate the power supply coupled with the grounds being shared early on, what is the motivation for this configuration? (i just used 4 diodes for regulation for each channel since i already started the circuit like this)
well i guess my bigger question is, what are the motivations for the different configurations such as the chipamp powersupply versus nuuks buffer power supply.
chipamp power supply with split gnds and 4 diodes for rectification
thanks,
frdchang, there will always be a variety of circuits that will do the same job. We have to decide how complicated we want to make them and that decision will be based on factors such as cost, space, and most importantly, how they sound.
Some folks reckon that the extra stage of regulation in that supply is overkill and some will say that two rectifiers bridges is also unnecessary.
The design shown on DD is what I arrived at after a bit of trial and error. I really got to the stage where I thought the supply sounded good enough and my money was better spent elsewhere.
I also target DD at the novice so I didn't want to land up with something that was too complicated to understand. As you have found, once the basic circuit is understood, it isn't too difficult to modify it.
The second caps shown as 0.1 uF should be 0.01 uF.
There is nothing to stop you trying that circuit (keeping in mind the corrctions mentioned above) and it should sound fine!
Some folks reckon that the extra stage of regulation in that supply is overkill and some will say that two rectifiers bridges is also unnecessary.
The design shown on DD is what I arrived at after a bit of trial and error. I really got to the stage where I thought the supply sounded good enough and my money was better spent elsewhere.
I also target DD at the novice so I didn't want to land up with something that was too complicated to understand. As you have found, once the basic circuit is understood, it isn't too difficult to modify it.
The second caps shown as 0.1 uF should be 0.01 uF.
There is nothing to stop you trying that circuit (keeping in mind the corrctions mentioned above) and it should sound fine!
An observation
I have build three versions of the CFM AD815 preamp. The first was as drawn, but used 480 ohm resistors for feedback and shunt to yield a gain of two. This circuit was built in a hurry, there were leads and parts stuffed all over, looked like rats built it.
The sonics floored me. The itch was there, so another was built, this time as a buffer, unity gain, no shunt resistor off the inverting input. The sonics were still fine, but a little less than the first.
So another was cooked up, this time only using one side per chip to seperate the power supplies per channel. Better than the first but;
There was a roughness exhibited when the volume was increased. A first blamed in on input offset. Worked on it, problem seemed reduced. But when cranking up the volume to mid rotation there were signs of non-linearity, dynamics were too large, and bass was over prominent. I kind of enjoyed this distortion so ignored it.
This week I moved back on this buffer. Installed 480 ohm resistors for the shunt, to go with the 533 ohm feedback resistors already there. The magic was back. Much smoother, maybe a little less dynamic, but the bass at high levels was more realistic. Decided the circuit with gain is better than the buffer.
But another issue. My system gain is such that the unity gain pre never needs to go much past mid rotation. With some gain, the real low level matching got a little touchy.
So moved the resistors from the inverting input to the output to ground. With the 480 ohm resistors going from output to ground the sound seems much calmer and linear than it was without them. Maybe not as good as the circuit with gain, but better than without these resistors.
Thinking this through, I suspect the AD815 is being throttled by being loaded at the 100K impedance of my amp. There is a 2 ufd blocking cap infront of the 100K shunt resistor.
Looking at currents, most listening is done with a max power level of less than 200 mw, so this is about 0.1 volts out of the preamp. Into the 100K, this comes out to about 1 ua on peaks. No wonder the sound was getting screwed with, the 1 ua current could not deal with the blocking cap without some slewing effects..
The 480 ohm resistor on the load get the current coming from the AD815 up to ma levels anyway. So, lesson learned, if you use this chip as a buffer, you may consider adding a shunt resistor to at least let some current flow through. The AD815 is a high current line driver, it may not like to run at sub micro amp output levels.
I think the circuits with gain already take care of this issue, the shunt resistor is after the feedback to give a 900 - 1K impedance for loading.
George
I have build three versions of the CFM AD815 preamp. The first was as drawn, but used 480 ohm resistors for feedback and shunt to yield a gain of two. This circuit was built in a hurry, there were leads and parts stuffed all over, looked like rats built it.
The sonics floored me. The itch was there, so another was built, this time as a buffer, unity gain, no shunt resistor off the inverting input. The sonics were still fine, but a little less than the first.
So another was cooked up, this time only using one side per chip to seperate the power supplies per channel. Better than the first but;
There was a roughness exhibited when the volume was increased. A first blamed in on input offset. Worked on it, problem seemed reduced. But when cranking up the volume to mid rotation there were signs of non-linearity, dynamics were too large, and bass was over prominent. I kind of enjoyed this distortion so ignored it.
This week I moved back on this buffer. Installed 480 ohm resistors for the shunt, to go with the 533 ohm feedback resistors already there. The magic was back. Much smoother, maybe a little less dynamic, but the bass at high levels was more realistic. Decided the circuit with gain is better than the buffer.
But another issue. My system gain is such that the unity gain pre never needs to go much past mid rotation. With some gain, the real low level matching got a little touchy.
So moved the resistors from the inverting input to the output to ground. With the 480 ohm resistors going from output to ground the sound seems much calmer and linear than it was without them. Maybe not as good as the circuit with gain, but better than without these resistors.
Thinking this through, I suspect the AD815 is being throttled by being loaded at the 100K impedance of my amp. There is a 2 ufd blocking cap infront of the 100K shunt resistor.
Looking at currents, most listening is done with a max power level of less than 200 mw, so this is about 0.1 volts out of the preamp. Into the 100K, this comes out to about 1 ua on peaks. No wonder the sound was getting screwed with, the 1 ua current could not deal with the blocking cap without some slewing effects..
The 480 ohm resistor on the load get the current coming from the AD815 up to ma levels anyway. So, lesson learned, if you use this chip as a buffer, you may consider adding a shunt resistor to at least let some current flow through. The AD815 is a high current line driver, it may not like to run at sub micro amp output levels.
I think the circuits with gain already take care of this issue, the shunt resistor is after the feedback to give a 900 - 1K impedance for loading.
George
nuuk, is the power ok now?
here are my schematics:
straight carlosm preamp
layout and parts list will come when i get a chance.
here are my schematics:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
straight carlosm preamp
layout and parts list will come when i get a chance.
Wow, that picture took a long time to appear!
I had a quick glance as I am in the middle of bread-making but it looks OK apart from the small bypass caps. You don't need to change them both to 0.01 uF, you want one of them to be 0.1 and the other 0.01. The idea is to bypass each cap with a smaller one.
New PS for CFM AD815AY
Was thinking about how much Rudi liked this chip with dual regulated active power supply. Since my raw supply has been left alone for about three years it was time for a little updating.
Used dual 317HVH/337HVH three pin adjustable regs. Raw dc is right at +/- 30 volts dc. First stage dropping it down to 22 volts, then 15.2 volts at the board.
Then out through a pair of 10 mH chokes to the preamp. There is a total of 7K ufd in the preamp to termeinate the chokes.
So things learned in the last few years. No bypass caps per se. Output of last reg has a 47 ufd cap and 0.4 ohm resistor. The 4.7K first stage caps are snubbed with a 0.4 ohm resistor and 1.5 nf cap. There is a 0.1 ohm resistor buffering the ground between the preamp and the active circuitry.
George
Was thinking about how much Rudi liked this chip with dual regulated active power supply. Since my raw supply has been left alone for about three years it was time for a little updating.
Used dual 317HVH/337HVH three pin adjustable regs. Raw dc is right at +/- 30 volts dc. First stage dropping it down to 22 volts, then 15.2 volts at the board.
Then out through a pair of 10 mH chokes to the preamp. There is a total of 7K ufd in the preamp to termeinate the chokes.
So things learned in the last few years. No bypass caps per se. Output of last reg has a 47 ufd cap and 0.4 ohm resistor. The 4.7K first stage caps are snubbed with a 0.4 ohm resistor and 1.5 nf cap. There is a 0.1 ohm resistor buffering the ground between the preamp and the active circuitry.
George
Attachments
![dsc01285[1].jpg](/community/data/attachments/70/70772-f4cffa2cc0bdb0bc8d4ee0ef30afe8e2.jpg)
Sound,
This shows my old board, single stage of active regulation feeding a two stage LC filter. New one is two stage active fedding a single passive.
The sound is definately better with the new board. Sound is more rounded, less electronic. I never pay attention to imaging, so dynamics, spectral balance, and warmth are the factors judged from.
Could be dual regs are better. My previous experiance with them was not good, but this was using them without a choke on the outpu. Also had bypassed caps inbetween and on output.
May just be better implimentation of the parts. Either way, this is really getting there now. Time to work over my sources again.
Thanks to all who throw out ideas on how to build better audio!
George
This shows my old board, single stage of active regulation feeding a two stage LC filter. New one is two stage active fedding a single passive.
The sound is definately better with the new board. Sound is more rounded, less electronic. I never pay attention to imaging, so dynamics, spectral balance, and warmth are the factors judged from.
Could be dual regs are better. My previous experiance with them was not good, but this was using them without a choke on the outpu. Also had bypassed caps inbetween and on output.
May just be better implimentation of the parts. Either way, this is really getting there now. Time to work over my sources again.
Thanks to all who throw out ideas on how to build better audio!
George
Attachments
![dsc01287[1].jpg](/community/data/attachments/70/70779-53558e0c7050a492855c777decbc543a.jpg)
i was using a carlos type AD 815 for preamp and headphone duty.. i I love the sound of the AD815 especially when it has to drive a long wire
built a kookaburra and i am using it right now. instead of the AD
the kook has issues at the moment because i have +10DB equipment and it clips at +- 3.5v or so . so i can not turn it up much. most people will not have this problem with consumer gear.
at lower levels i feel the kook sounds better but as it gets louder
I still prefer the output of the AD815 especially for phones.
AD just has more "beef" to it than headphones off a little op amp (dual one even)
but i like the kooks remote voltage control pot with no audio going through it. very versatile! the kook has less noise for me as well.
I think its time to make a hybrid of these. use the kookaburra controller PIC if code is downloadable or if TwistedPear will sell the chips seperately because the way it is programmed code is beutiful..
only if they are ok with the idea.
i would use the PGA 2310 instead of the 2311 although the distortion is 0.0002 more it has +- 15v instead of +-5v supply and a lot more dynamic headroom.. i would rather have 0.0002 or whatever distortion than a hard clip in the middle of the chain i am not expecting.
then it has a fixed output impedance so the DC offset fix from carlos will work on the AD.
input > optional buffer (board made to fit singles, duals and surface opamps) > PGA2310 or PGA2311 powered by twisted pear > does it need a buffer? > AD815 >space for optionaol caps
the PGA would have seperate regulators, regulators per chip or a jumper depending if you want PGA10 or 11
what does anyone think of that?
built a kookaburra and i am using it right now. instead of the AD
the kook has issues at the moment because i have +10DB equipment and it clips at +- 3.5v or so . so i can not turn it up much. most people will not have this problem with consumer gear.
at lower levels i feel the kook sounds better but as it gets louder
I still prefer the output of the AD815 especially for phones.
AD just has more "beef" to it than headphones off a little op amp (dual one even)
but i like the kooks remote voltage control pot with no audio going through it. very versatile! the kook has less noise for me as well.
I think its time to make a hybrid of these. use the kookaburra controller PIC if code is downloadable or if TwistedPear will sell the chips seperately because the way it is programmed code is beutiful..
only if they are ok with the idea.
i would use the PGA 2310 instead of the 2311 although the distortion is 0.0002 more it has +- 15v instead of +-5v supply and a lot more dynamic headroom.. i would rather have 0.0002 or whatever distortion than a hard clip in the middle of the chain i am not expecting.
then it has a fixed output impedance so the DC offset fix from carlos will work on the AD.
input > optional buffer (board made to fit singles, duals and surface opamps) > PGA2310 or PGA2311 powered by twisted pear > does it need a buffer? > AD815 >space for optionaol caps
the PGA would have seperate regulators, regulators per chip or a jumper depending if you want PGA10 or 11
what does anyone think of that?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.