New one on me. Just ordered an new batch of the 4562's today. I also just got several cheap MCoy kits from e-bay so I can practice on my SMT skills. Should be fun as I am old and half blind.
Luigi,
I found the data sheet. It is attached here.
Read page 7 of the BB data sheet for the DAC- last two paragraphs.
What you may be hearing is most likely, your transformer saturating with RF from the DAC. The data sheet suggests that the DAC be followed with a high-order filter (a very steep slope 24-36db/oct-9th to 12th order LPF around 22-25kHz)which is probably what you bypassed in the Yamaha. If you have access to a scope you will see all kinds of HF artifacts mixed with the audio signal.
What do you mean by balanced? You only have +signal and GND as outputs. To feed a balanced transformer input with an unbalanced source means you are unbalancing the input. A true balanced signal is + signal, -signal and GND. The negative (or 180 DEG phase reverse) is produced, usually by an inverting op-amp/ driver. This is a separate topic from trying to get a DAC to work without a LPF of some kind.
I get the feeling that you can't really do what you want to, in this case. Perhaps the better way to go would be to optimize the output section of the Yamaha unit with the best parts money can buy. Also, get rid of any electrolytic caps in the audio path or design a "servo"-style output which eliminates DC and the need for caps in the audio path in the first place, etc, etc. Op amps are not that evil, if done correctly.
Ciao,
Gregg
I found the data sheet. It is attached here.
Read page 7 of the BB data sheet for the DAC- last two paragraphs.
What you may be hearing is most likely, your transformer saturating with RF from the DAC. The data sheet suggests that the DAC be followed with a high-order filter (a very steep slope 24-36db/oct-9th to 12th order LPF around 22-25kHz)which is probably what you bypassed in the Yamaha. If you have access to a scope you will see all kinds of HF artifacts mixed with the audio signal.
What do you mean by balanced? You only have +signal and GND as outputs. To feed a balanced transformer input with an unbalanced source means you are unbalancing the input. A true balanced signal is + signal, -signal and GND. The negative (or 180 DEG phase reverse) is produced, usually by an inverting op-amp/ driver. This is a separate topic from trying to get a DAC to work without a LPF of some kind.
I get the feeling that you can't really do what you want to, in this case. Perhaps the better way to go would be to optimize the output section of the Yamaha unit with the best parts money can buy. Also, get rid of any electrolytic caps in the audio path or design a "servo"-style output which eliminates DC and the need for caps in the audio path in the first place, etc, etc. Op amps are not that evil, if done correctly.
Ciao,
Gregg
My blog page has plenty of info about op amps and mod's
opamp - opamp construction
Supplying power supply with surface mount transistors in a emitter follower configuration all on board the Brown Dog adapter is my latest venture, See schematic and blog
Stereo separation improves immensely,
Happy listening
Cheers / Chris
opamp - opamp construction
Supplying power supply with surface mount transistors in a emitter follower configuration all on board the Brown Dog adapter is my latest venture, See schematic and blog
Stereo separation improves immensely,
Happy listening
Cheers / Chris
Attachments
DAC follower
Dear Luigi,
I have to second, thats it would be a good idea with a kind of 'follower' after Your Dac.
DAC's normally delievers current out with high impedance and a form of impedance adaptor would therefore be a good idea.
I do not recommend You to Use the 4562 and derivatives however. If find them very precise and clean, but also a bit 'lifeless' - bit like most Japanese cars really... - To perfect for their own good. There's a reason a lot of these impedance adaptors/follower has previously been done with NE5534's and 5532 for economy. A good samples of these will do a very competent job here with loads of musicallity. Best is probably the old Signetics version in Cerdip, but a BB/Ti one will do the job...
You can also venture into doing something discrete here, since all You need is impedance transformation. A current generator/source and an emmitter or source follower not to mention delievering the signal from the DAC to a virtual ground point, should do just the trick. There will most probably be simplified schematic's/curcuits avaliable on this site and/or if You google around a bit You should be able to find something suitable.
Best of hunting
DocO
Dear Luigi,
I have to second, thats it would be a good idea with a kind of 'follower' after Your Dac.
DAC's normally delievers current out with high impedance and a form of impedance adaptor would therefore be a good idea.
I do not recommend You to Use the 4562 and derivatives however. If find them very precise and clean, but also a bit 'lifeless' - bit like most Japanese cars really... - To perfect for their own good. There's a reason a lot of these impedance adaptors/follower has previously been done with NE5534's and 5532 for economy. A good samples of these will do a very competent job here with loads of musicallity. Best is probably the old Signetics version in Cerdip, but a BB/Ti one will do the job...
You can also venture into doing something discrete here, since all You need is impedance transformation. A current generator/source and an emmitter or source follower not to mention delievering the signal from the DAC to a virtual ground point, should do just the trick. There will most probably be simplified schematic's/curcuits avaliable on this site and/or if You google around a bit You should be able to find something suitable.
Best of hunting
DocO
Well, went through your blog. using pass transistor makes sense.but then I have seen in old amps and preamps, that mfg used different powersupply for each channel. Also is recommended to use a pair of pass transistor stage for each channel instead of using it with every op-amp.
I got a more doubt after reading your blog. what was the selection craiteria used that made the new generation opamp sound better over the one of yesteryears. unity gain bw? noise? CMRR? PSRR? slew rate or something else?
Hi Aucosticraft
This idea, has provided lots of discussion in the Solid State section of the Forum under Powering Opamps. It is causing a lot of comment particularly concerning " subjective- what people are hearing " vs measured what is revealed by oscilloscopes.
I am providing some scope measurements later today, from the same circuit connected to a Marantz CD80.
My thoughts on how to describe it, and you would think someone after 58 years of transistor use would have connected a transistor this way, is that its closest description is that of a transistor connected diode, however it seems to improve with resistance between the collector and base, as I have designed, and improves again with load being drawn from it.
It also follows V+ V- less transistor junction, alluding ( where there has been much heated discussion ) , where I initially described it as an emitter follower, about its correct description. I have amended website description last night, awaiting better clarification
Subjectively when involved with opamps it improves consistently the spatial stereo information, that in my experience of op amps is a very good sign of improvement.
Cheers / Chris
This idea, has provided lots of discussion in the Solid State section of the Forum under Powering Opamps. It is causing a lot of comment particularly concerning " subjective- what people are hearing " vs measured what is revealed by oscilloscopes.
I am providing some scope measurements later today, from the same circuit connected to a Marantz CD80.
My thoughts on how to describe it, and you would think someone after 58 years of transistor use would have connected a transistor this way, is that its closest description is that of a transistor connected diode, however it seems to improve with resistance between the collector and base, as I have designed, and improves again with load being drawn from it.
It also follows V+ V- less transistor junction, alluding ( where there has been much heated discussion ) , where I initially described it as an emitter follower, about its correct description. I have amended website description last night, awaiting better clarification
Subjectively when involved with opamps it improves consistently the spatial stereo information, that in my experience of op amps is a very good sign of improvement.
Cheers / Chris
pcm 63
Dear Gregg,
Many thanks for your last answer regarding the pcm56 cdp yamaha, but now I wanted to ask you for information instead PCM63 (denon dcd3520)
and 'can bypass this opamp dac?
or i 'can use an output transformer without opamp?
Luigi
Dear Gregg,
Many thanks for your last answer regarding the pcm56 cdp yamaha, but now I wanted to ask you for information instead PCM63 (denon dcd3520)
and 'can bypass this opamp dac?
or i 'can use an output transformer without opamp?
Luigi
Attachments
pcm64 not pcm63
sorry ......
the dac is pcm 64 (denon dcd 3520) and not pcm 63
the question is same .
thanks luigi
sorry ......
the dac is pcm 64 (denon dcd 3520) and not pcm 63
the question is same .
thanks luigi
That is a great op amp, I have tried it and use it in a home made pre. The Burr-Brown OPA2134 is one that is at the top of my list. But often, it is more of how the op amp integrates into the circuit that matters rather than how good the op amp itself is. But op amp rolling is fun. Also tried the OPA2604, but not had good results with it.
Good luck.
Good luck.
help gregg for pcm 56
Hi Greeg thanks for your answer please but if use step-dawn transformer 4:1
2500 ohm / 600 ohm after the pcm56p
can good matched dac/imput preamplifier with low impedence ?
how you think ?
thanks luigi
Hi Greeg thanks for your answer please but if use step-dawn transformer 4:1
2500 ohm / 600 ohm after the pcm56p
can good matched dac/imput preamplifier with low impedence ?
how you think ?
thanks luigi
Foxye,
The LME49720HA (metal can) is the best opamp of the group you listed! Be sure to use a good regulator though and run the device at 15-19VDC rails...with a clip-on heatsink on the device. Do not use any plastic package devices.
I just modified a D/A preamp last week (that I designed many years ago at National-now TI) by replacing all the plastic DIPs with metal cans. Several engineers who heard it were surprised that the same die in a metal can package could sound that different! (I actually did the same demo for Bob Pease during my first stint at National and he was shocked that he could hear the difference. Wish he was still with us!) The only plastic opamp I would play with right now is the LME49990. 0.9nV of noise!
Of course my favorite opamp is still the CFB LME49713HA...but since I did specify that part I am a bit biased! It is a Current Feedback Opamp though and requires a few special considerations when using it, like a 1.2k to 1.5k feedback resistor.
Best National...I mean TI regards,
Audioman54
The LME49720HA (metal can) is the best opamp of the group you listed! Be sure to use a good regulator though and run the device at 15-19VDC rails...with a clip-on heatsink on the device. Do not use any plastic package devices.
I just modified a D/A preamp last week (that I designed many years ago at National-now TI) by replacing all the plastic DIPs with metal cans. Several engineers who heard it were surprised that the same die in a metal can package could sound that different! (I actually did the same demo for Bob Pease during my first stint at National and he was shocked that he could hear the difference. Wish he was still with us!) The only plastic opamp I would play with right now is the LME49990. 0.9nV of noise!
Of course my favorite opamp is still the CFB LME49713HA...but since I did specify that part I am a bit biased!
It is a Current Feedback Opamp though and requires a few special considerations when using it, like a 1.2k to 1.5k feedback resistor.Best National...I mean TI regards,
Audioman54
Audioman
A lot of the so-called evidence for the sonic qualities of different op-amps comes from anecdotal sources or poorly designed comparisons. I think your example of swapping plastic case for metal can versions of the same op-amp comes into that category I'm afraid because it relies on your memory of what the DAC sounded like before you made the swap, and that's not 100% reliable. Are you aware of a properly designed and conducted double blind test where the basic circuit remains the same and only the op-amp is different e.g. two identical DACs using different op-amps switched blind while playing music in a reference system.
Did National carried out double-blind tests like this to evaluate their audio op-amps when you were there? If so, were any statistically significant differences between sonic qualities of the different op-amps tested found?
A lot of the so-called evidence for the sonic qualities of different op-amps comes from anecdotal sources or poorly designed comparisons. I think your example of swapping plastic case for metal can versions of the same op-amp comes into that category I'm afraid because it relies on your memory of what the DAC sounded like before you made the swap, and that's not 100% reliable. Are you aware of a properly designed and conducted double blind test where the basic circuit remains the same and only the op-amp is different e.g. two identical DACs using different op-amps switched blind while playing music in a reference system.
Did National carried out double-blind tests like this to evaluate their audio op-amps when you were there? If so, were any statistically significant differences between sonic qualities of the different op-amps tested found?
Last edited:
I have a hypothesis for why there might be a sound quality difference between the devices in the metal can vs the plastic ones. Would you be able to describe what the differences sounded like to you? Also did Bob hear the same effects?