OK. I had thought of something like 2sc2240 / 2sa970 but anyway ...
Both the 2sc2240 and 2sa970 are obsolete and end of life.
1 - What is the ideal voltage amplitude on the load? Or varies the power you want? Ex: 6v amplitude / 60 ohms load.
The higher the amplitude of the load, will have more definition / fidelity?
The load is fixed by your choice of headphones. The voltage is then fixed by your choice of volume. Your question makes no sense to me.
In terms of efficiency, power output, and distortion, the circuit is optimized for 60 ohm loads, as this gives about equally good performance for 16 ohm and 300 ohm headphones.
AD797, DIY-990, and others
The AD797 had 6 mv of offset in one channel and less in the other channel. There was no sign of instability. In a word, the sound is clean. The highs were extended as were the lows with no embellishment. It lent itself to long listening sessions and put one in the room with the performers. And it had good pacing. It was quieter than the other monolithic op amps.
The DIY-990 has 35 mv of offset in one channel and 49 mv in the other, after adjustment. That involved swapping DOAs between channels to get the 3 mv extra to get it below 50 mv. Like the other DOAs, the DIY-990 puts one on the stage with the performers. It seems to be able to bring out more texture in the instruments as well as provide more impact to crescendos. There seems to be more depth the to silence, yet it's no quieter than the AD797 and it doesn't have the blackness of the SG-DOA-2. The DIY-990 makes it easy to distinguish the cymbals used as well as the change of technique in playing them. Getting lost in a performance is too easy with this op amp. And, as cliched as this sounds, I was hearing parts of songs that I'd missed before with the other op amps.
The others -
The AD843 was quite the interesting op amp. Tangent reviewed this in the early 2000's. There was no offset to report with the Sapphire. It had very good extension, with good highs. But it has too much bass. I kept wanting to shelve the 40 Hz to 100 Hz down 1 dB to 1.5 dB. When I swapped it out it felt like having a weight removed from my head. And the AD797 seemed very light. Still, it brought the bass players up front. There was an intense sense of all the textures on each bass note and drum hit.
I had high hopes for the Burson "Supreme Sound Op-amp". Indeed it had very nice air on the highs. It was easy to listen to and there was no offset to worry about. It has one downside when used in the Sapphire - ignoring it's 1.6" height, and that is the bass. While it is extended, it seems flat or 2-Dimensional.
The gar2520 was tried because I was curious about how they voice a DOA to sound like those from the API consoles from the 1970s. True to their word, it was forward on vocals and the mid-range. While the bass was extended, the voicing caused the dynamics brought with bass to be subdued. And the highs were hard and harsh. A good op amp for it's intended purpose.
The ML-918 is a the DIY-990's older sibling. Fewer parts and slower, it's a nice op amp to live with in the Sapphire if the voltage is raised to +/-15 Vdc. While it works well at +/-12 Vdc, complex music can cause some breakup. It has good bass extension and provides the intimacy of being on the stage or at least in the first row.
The AD797 had 6 mv of offset in one channel and less in the other channel. There was no sign of instability. In a word, the sound is clean. The highs were extended as were the lows with no embellishment. It lent itself to long listening sessions and put one in the room with the performers. And it had good pacing. It was quieter than the other monolithic op amps.
The DIY-990 has 35 mv of offset in one channel and 49 mv in the other, after adjustment. That involved swapping DOAs between channels to get the 3 mv extra to get it below 50 mv. Like the other DOAs, the DIY-990 puts one on the stage with the performers. It seems to be able to bring out more texture in the instruments as well as provide more impact to crescendos. There seems to be more depth the to silence, yet it's no quieter than the AD797 and it doesn't have the blackness of the SG-DOA-2. The DIY-990 makes it easy to distinguish the cymbals used as well as the change of technique in playing them. Getting lost in a performance is too easy with this op amp. And, as cliched as this sounds, I was hearing parts of songs that I'd missed before with the other op amps.
The others -
The AD843 was quite the interesting op amp. Tangent reviewed this in the early 2000's. There was no offset to report with the Sapphire. It had very good extension, with good highs. But it has too much bass. I kept wanting to shelve the 40 Hz to 100 Hz down 1 dB to 1.5 dB. When I swapped it out it felt like having a weight removed from my head. And the AD797 seemed very light. Still, it brought the bass players up front. There was an intense sense of all the textures on each bass note and drum hit.
I had high hopes for the Burson "Supreme Sound Op-amp". Indeed it had very nice air on the highs. It was easy to listen to and there was no offset to worry about. It has one downside when used in the Sapphire - ignoring it's 1.6" height, and that is the bass. While it is extended, it seems flat or 2-Dimensional.
The gar2520 was tried because I was curious about how they voice a DOA to sound like those from the API consoles from the 1970s. True to their word, it was forward on vocals and the mid-range. While the bass was extended, the voicing caused the dynamics brought with bass to be subdued. And the highs were hard and harsh. A good op amp for it's intended purpose.
The ML-918 is a the DIY-990's older sibling. Fewer parts and slower, it's a nice op amp to live with in the Sapphire if the voltage is raised to +/-15 Vdc. While it works well at +/-12 Vdc, complex music can cause some breakup. It has good bass extension and provides the intimacy of being on the stage or at least in the first row.
Excuse me for the bad English, I expressed myself badly.
What I wanna know is: If you believe that IC buffers, can not reach the level of buffers based on discrete components?
Or do you have a view to a low-cost circuit?
My intention of an amplifier or headphones, is to extract the purest class A possible. But what I see is more comments on the LME49600, but I honestly believe more in discrete components. However, I like to know your honest opinion.
"extract the purest class A possible"
How do you know 99.99% class A sounds better than 99.9% class A? How could anyone even begin to quantify that? The mind boggles...
Since you asked, though, yes, the problem with IC buffers is they have very low bias current though the output stage, so they will indeed operate in class AB in a typical headphone amp circuit.
Since I wanted my buffer to run open loop, without using global feedback from the op amp open loop gain to reduce the distortion, I have to make sure the bias current is high enough that the output always runs in class A. The easiest way to do that is to build a discrete output circuit where I can set the bias current as I like.
How do you know 99.99% class A sounds better than 99.9% class A? How could anyone even begin to quantify that? The mind boggles...
Since you asked, though, yes, the problem with IC buffers is they have very low bias current though the output stage, so they will indeed operate in class AB in a typical headphone amp circuit.
Since I wanted my buffer to run open loop, without using global feedback from the op amp open loop gain to reduce the distortion, I have to make sure the bias current is high enough that the output always runs in class A. The easiest way to do that is to build a discrete output circuit where I can set the bias current as I like.
Last edited:
You tested Burson "Supreme Sound Op-amp" is standart Sapphire or with any modification? By datasheet Burson have very high Input Bias Current (180uA) and Input Offset Curren (70uA). It must have really big DC offset - or I am wrong? On page 54 is writen, that 5uA on Sparko SS3601 was big problem until modification.
Hi RJM
I am very interesant about this amplifier, I have in plan order PCBs. Please, I have any questions:
- Cas I use 2N5401/2N5551 instead BC327/BC337?
- Most Opamps sound better with higher voltage. Can I use 16V or 17V in circuit? What I must modify in Z-reg for this?
- How much is max. output current from opamp? I have any OPA128SM, what is excellent opamp, but have only 10mA output current.
I am very interesant about this amplifier, I have in plan order PCBs. Please, I have any questions:
- Cas I use 2N5401/2N5551 instead BC327/BC337?
- Most Opamps sound better with higher voltage. Can I use 16V or 17V in circuit? What I must modify in Z-reg for this?
- How much is max. output current from opamp? I have any OPA128SM, what is excellent opamp, but have only 10mA output current.
@Wishmaster
Yes, you can use those transistors, but it is your responsibility to confirm the pin assignments.
The maximum output current of the op amp is quite low, in the order of <~1 mA.
"Most Opamps sound better with higher voltage." While I personally disagree with that assessment, you can increase the voltage by changing the Zener to 18V increasing the power supply voltages to +/- 25V, and increasing the rating of all the electrolytic capacitors from 25V to 35V.
/R
Yes, you can use those transistors, but it is your responsibility to confirm the pin assignments.
The maximum output current of the op amp is quite low, in the order of <~1 mA.
"Most Opamps sound better with higher voltage." While I personally disagree with that assessment, you can increase the voltage by changing the Zener to 18V increasing the power supply voltages to +/- 25V, and increasing the rating of all the electrolytic capacitors from 25V to 35V.
/R
long time since i posted as been interested in other pursuits like my motorbike.
roll on about a year maybe more or less i dont know as i have forgotten but i decided to fire up the RJM Sapphire (2.0f i think) and my lord its lovely. in the past i rolled a fair few op amps but i still return to the OPA134 as to me it has a nice musical sound to it. i listen to music to chill out the other favourite is the NE5534 but its a bit colder than the 134. something else that made a huge difference was changing the ear pads on my DT880 600ohm the pads off the 770 are great for rock, the original 880 are a bit ermm thin sounding but my favourite at the moment are the ones off the superlux HD330. the Sapphire is still work in progress but i see Richard has a v3 now so i think i will give it a go at some point in the near future.
i think what Richard has done is nothing short of outstanding in regards to how it sounds v's cost. ok so i could have gone out and spent some money on a ready made thing that might have sounded worse, same or better but i have had the fun of building it myself and to my untrained ears the results are good plus i have learnt something along the way.
might just fire up the iron and get to work on the v2 again to see if i can tidy it up which might help the sound as everything is still on test pins.
roll on about a year maybe more or less i dont know as i have forgotten but i decided to fire up the RJM Sapphire (2.0f i think) and my lord its lovely. in the past i rolled a fair few op amps but i still return to the OPA134 as to me it has a nice musical sound to it. i listen to music to chill out the other favourite is the NE5534 but its a bit colder than the 134. something else that made a huge difference was changing the ear pads on my DT880 600ohm the pads off the 770 are great for rock, the original 880 are a bit ermm thin sounding but my favourite at the moment are the ones off the superlux HD330. the Sapphire is still work in progress but i see Richard has a v3 now so i think i will give it a go at some point in the near future.
i think what Richard has done is nothing short of outstanding in regards to how it sounds v's cost. ok so i could have gone out and spent some money on a ready made thing that might have sounded worse, same or better but i have had the fun of building it myself and to my untrained ears the results are good plus i have learnt something along the way.
might just fire up the iron and get to work on the v2 again to see if i can tidy it up which might help the sound as everything is still on test pins.
just catching up on the thread. i too would like to try an LDR at some point just to see what all the fuss is about.
it strikes me that there is already a 12v supply with the Sapphire so would it not just be a case of making a daughter board with some resistors to pull the 12v down to 5v. browsing the net for 12v to 5v i stumbled upon this
An externally hosted image should be here but it was not working when we last tested it.
but i dont know nowt about nowt so i dont know if it would work for the 5v
Last edited:
That's just a three terminal regulator datasheet configuration, drawn out in "electronics for dummies" style instead of a normal circuit diagram.
It's painful to look at. Like seeing music written out in "Do Re Mi" instead of notes and bars.
If the LDR uses digital logic, it might be better to make a separate little 5 V supply - well, that's the high-end way of approaching this at any rate. Or you can use an LM7805 (see Figure 9. on page 19)... connected to either the V+ or V++ rails.
It's painful to look at. Like seeing music written out in "Do Re Mi" instead of notes and bars.
If the LDR uses digital logic, it might be better to make a separate little 5 V supply - well, that's the high-end way of approaching this at any rate. Or you can use an LM7805 (see Figure 9. on page 19)... connected to either the V+ or V++ rails.
He was but there is a confusion here.
There is a discussion about ldr pot and power source for this pot.
There was a link to github.com above
https://github.com/camerontech/inclinometer/blob/master/README.md
Outlining how to get the car Voltage down from 12v to 5v and this picture is taken from github.com and is showing how to do this.
I have a feeling that people here got Co fused by the LDR in the bottom line and thought that this is an LDR pot with basic not good schematics. While this is not.
My guess is that LDR pot schematics were not published yet
Sent from my ONE A2003 using Tapatalk
There is a discussion about ldr pot and power source for this pot.
There was a link to github.com above
https://github.com/camerontech/inclinometer/blob/master/README.md
Outlining how to get the car Voltage down from 12v to 5v and this picture is taken from github.com and is showing how to do this.
I have a feeling that people here got Co fused by the LDR in the bottom line and thought that this is an LDR pot with basic not good schematics. While this is not.
My guess is that LDR pot schematics were not published yet
Sent from my ONE A2003 using Tapatalk
ignore my earlier post i was being stupid as per.
on another note, i knew that turning off my LED lamp at the mains rewarded me with better sound Q so i decided last night for the first time as its a PITA to turn everything connected to the spur off (i run a spur from the consumer unit for all my entertainments system when i moved in) and guess what.. yup sounds is spot on and its the best i have ever heard my Sapphire sing and the boards are still hooked up very roughly in test mode. might see if i can get everything wired up properly in the coming week or so.
on another note, i knew that turning off my LED lamp at the mains rewarded me with better sound Q so i decided last night for the first time as its a PITA to turn everything connected to the spur off (i run a spur from the consumer unit for all my entertainments system when i moved in) and guess what.. yup sounds is spot on and its the best i have ever heard my Sapphire sing and the boards are still hooked up very roughly in test mode. might see if i can get everything wired up properly in the coming week or so.
Every couple of years I get bitten by the "how'm I doin'?" bug and find myself buying a commercial unit to try out against my own stuff.
I wrote about the Orb Jade-1 in a blog post. In this forum post I will examine how it stacks up against my Sapphire3 as a pure headphone amplifier.
Source is my analog stage, Phonoclone 3, Denon DL-103, DP-2000/DA-207 turntable/tonearm. Headphones are Sennheiser HD600.
My best guess of the Jade-1 headphone output stage is attached below. The configuration of T1 R1 and C5 is 99% certain. Output coupling cap is Nichicon KW. T2 and R2 and the details of the biasing network are unknown.
The +12 and -12 V rails are generated by 10,000 uF cap, LM7x12, 10,000 uF cap in the power supply. The rails are shared between channels.
So first let me tell you what I heard, then I'll speculate why it sounded like it did.
*****
With a good analog signal instead of the internal DAC, the Jade-1 sound is pleasant, smooth and warm. No trace of midrange hardness or glare. Good bass extension. The first indication of trouble was I noted I could not hear the needle running in the groove of the LP. That hiss of surface noise was absent, even in the lead in and lead out. the second indication of trouble was that I couldn't hear "into" the music. Sounds are presented in some semblance of stereo, but there is no space between them, and, crucially, no depth.
The best analogy is to say it sounded "tired/heavy". It's all smudged out: no air, no splash, no rhythm. All the treble information is sucked out of the music leaving only dull smoothness. If you are familiar with the recording and have multiple references to compare, it's brutally obvious that the Jade-1 is giving a very odd rendering. It's just plainly off.
The Sapphire3 puts the life back into recordings. It really is night and day. Stereo image seems to extend twice was wide, and there is proper sense of depth. But more than that the tone is back to what I know to be on the LP. Instruments sound like they are supposed to sound.
*****
So, what went wrong?
I suggest most of the problem is the power supply. Nice R-core transformer, but at most a 3A bridge rectifier feeding 40,000 uF!? LM7812/7912? I know, it's not uncommon to find this in high end equipment, but I think its just totally the wrong way to go about it.
The other point is the RC stages locally at the output stage. RC stages on the rails often seem to dull the sound, at least in solid state applications.
Shared power supply does nothing for the stereo imaging.
A large electrolytic output coupling cap often results in an overly slow, damped sound.
In my experience the single ended follower output stage does not necessarily sound better the more current you send through it, and is rather dark-sounding at the best of times. The Jade runs at over 250 mA bias.
Anyway, all very fascinating. I've played around with followers as headphone amps before, with the J-Mo II for example, and while there are some tonal similarities I can recognize, the the J-Mo has for more clarity and definition than the Orb design. So in this instance I think most of the blame falls on the power supply, and designing it on paper for lowest noise and energy storage rather that listening evaluation.
I wrote about the Orb Jade-1 in a blog post. In this forum post I will examine how it stacks up against my Sapphire3 as a pure headphone amplifier.
Source is my analog stage, Phonoclone 3, Denon DL-103, DP-2000/DA-207 turntable/tonearm. Headphones are Sennheiser HD600.
My best guess of the Jade-1 headphone output stage is attached below. The configuration of T1 R1 and C5 is 99% certain. Output coupling cap is Nichicon KW. T2 and R2 and the details of the biasing network are unknown.
The +12 and -12 V rails are generated by 10,000 uF cap, LM7x12, 10,000 uF cap in the power supply. The rails are shared between channels.
So first let me tell you what I heard, then I'll speculate why it sounded like it did.
*****
With a good analog signal instead of the internal DAC, the Jade-1 sound is pleasant, smooth and warm. No trace of midrange hardness or glare. Good bass extension. The first indication of trouble was I noted I could not hear the needle running in the groove of the LP. That hiss of surface noise was absent, even in the lead in and lead out. the second indication of trouble was that I couldn't hear "into" the music. Sounds are presented in some semblance of stereo, but there is no space between them, and, crucially, no depth.
The best analogy is to say it sounded "tired/heavy". It's all smudged out: no air, no splash, no rhythm. All the treble information is sucked out of the music leaving only dull smoothness. If you are familiar with the recording and have multiple references to compare, it's brutally obvious that the Jade-1 is giving a very odd rendering. It's just plainly off.
The Sapphire3 puts the life back into recordings. It really is night and day. Stereo image seems to extend twice was wide, and there is proper sense of depth. But more than that the tone is back to what I know to be on the LP. Instruments sound like they are supposed to sound.
*****
So, what went wrong?
I suggest most of the problem is the power supply. Nice R-core transformer, but at most a 3A bridge rectifier feeding 40,000 uF!? LM7812/7912? I know, it's not uncommon to find this in high end equipment, but I think its just totally the wrong way to go about it.
The other point is the RC stages locally at the output stage. RC stages on the rails often seem to dull the sound, at least in solid state applications.
Shared power supply does nothing for the stereo imaging.
A large electrolytic output coupling cap often results in an overly slow, damped sound.
In my experience the single ended follower output stage does not necessarily sound better the more current you send through it, and is rather dark-sounding at the best of times. The Jade runs at over 250 mA bias.
Anyway, all very fascinating. I've played around with followers as headphone amps before, with the J-Mo II for example, and while there are some tonal similarities I can recognize, the the J-Mo has for more clarity and definition than the Orb design. So in this instance I think most of the blame falls on the power supply, and designing it on paper for lowest noise and energy storage rather that listening evaluation.
Attachments
