AD844 common base stage subjective comments
I have been listening to this AD844 common base stage in a TDA1541A NOS DAC. The output is dc-coupled followed by passive attenuation/JLH/Quad ESL-57. A few subjective comments follow.
First impressions are of an open and unrestrained sound with remarkably copious ambience and presence. Real musical communication! There is some roughness in the treble and a metallic quality which reveals even the slightest of faults on a recording. While there is ample ambient information I am not as convinced by it as when using the AD844 in the more conventional role with feedback, the latter arrangement producing a wider 'stage' and a more relaxed sound, albeit more distant (and therefore less involving for some).
In some respects the piano sounds very real to me (I am a pianist) but there seem to be inconsistencies between almost every note which amount to more than what I hear in real life and more than what I hear from the same recordings when played through the AD844 (and other op-amps) used conventionally.
Some unique (?) strengths but weaknesses also. A very stimulating listen but a little too unrefined/inconsistent for me.
Tim.
I have been listening to this AD844 common base stage in a TDA1541A NOS DAC. The output is dc-coupled followed by passive attenuation/JLH/Quad ESL-57. A few subjective comments follow.
First impressions are of an open and unrestrained sound with remarkably copious ambience and presence. Real musical communication! There is some roughness in the treble and a metallic quality which reveals even the slightest of faults on a recording. While there is ample ambient information I am not as convinced by it as when using the AD844 in the more conventional role with feedback, the latter arrangement producing a wider 'stage' and a more relaxed sound, albeit more distant (and therefore less involving for some).
In some respects the piano sounds very real to me (I am a pianist) but there seem to be inconsistencies between almost every note which amount to more than what I hear in real life and more than what I hear from the same recordings when played through the AD844 (and other op-amps) used conventionally.
Some unique (?) strengths but weaknesses also. A very stimulating listen but a little too unrefined/inconsistent for me.
Tim.
You have basically outlined the weakness of TDA 1541 NOS design, which are not necessarily weakness of diamond output stage.
I have extensively compared TDA1541 NOS DAC with AD844 diamond configuration, and modified Wadia’s 27ix Decoding Computer DAC. Your remarks regarding piano notes are noticed by me and others as well. We have also spent countless hours trying to figure out if piano notes produced by NOS DAC are more faithful than those produced by Wadia…. My opinion is that Wadia gives closer to reality piano sound – and that’s it for Wadia! NOS DAC does everything else better, or much better than Wadia.
My latest incarnation of TDA1541 NOS DAC is “point to point wired” with off board power supply and reclocking at 50 MHz. The whole design is based around copper ground plane supporting receiver and DAC chips. It is by far the best DAC I’ve heard.
Boky
Boky,
I think I could have saved you a lot of time
Personally I have noticed nothing untoward wrt piano tone from a NOS DAC vs OS DAC. For me a NOS DAC reproduces the piano very faithfully indeed, even down to slightly out-of-tune unisons - the two strings 'beating' once every three or four seconds - clearly more intelligible from the NOS DAC. Naturally it is better to have the intrument in tune to start with, but this all adds to the realism! Tonal shading, dynamic gradations and a host of other characteristics are all better for me through the NOS DAC.
To recap, my criticism is with regard to the AD844 used in common base mode with the TDA1541A, where the piano becomes far less than convincing tonally/dynamically than when using the AD844 in the conventional manner.
What I don't know is whether this is due to the distortion characteristics of the AD844 stage when used in common base mode or to possible non-linearities from the TDA1541A caused by its voltage compliance being considerably exceeded as a result of the 50R input impedance of the AD844.
Boky, I am very curious to know whether you are using the AD844 in your latest TDA1541 DAC?
Tim.
I think I could have saved you a lot of time
Personally I have noticed nothing untoward wrt piano tone from a NOS DAC vs OS DAC. For me a NOS DAC reproduces the piano very faithfully indeed, even down to slightly out-of-tune unisons - the two strings 'beating' once every three or four seconds - clearly more intelligible from the NOS DAC. Naturally it is better to have the intrument in tune to start with, but this all adds to the realism! Tonal shading, dynamic gradations and a host of other characteristics are all better for me through the NOS DAC.
To recap, my criticism is with regard to the AD844 used in common base mode with the TDA1541A, where the piano becomes far less than convincing tonally/dynamically than when using the AD844 in the conventional manner.
What I don't know is whether this is due to the distortion characteristics of the AD844 stage when used in common base mode or to possible non-linearities from the TDA1541A caused by its voltage compliance being considerably exceeded as a result of the 50R input impedance of the AD844.
Boky, I am very curious to know whether you are using the AD844 in your latest TDA1541 DAC?
Tim.
An update....
Previous listening was with a 2mA constant current source on the AD844 output. With this removed the results are turned around completely - the AD844 now sounding considerably better as common base stage. The sound is wonderful with effortless, airy treble. Instruments simply float in the air over a firm bass. The tonal variety is almost intoxicating at times. I hope to try an OPA860 later but in the mean time this is superb. Pedja, if you are listening thankyou very much.
Tim.
Previous listening was with a 2mA constant current source on the AD844 output. With this removed the results are turned around completely - the AD844 now sounding considerably better as common base stage. The sound is wonderful with effortless, airy treble. Instruments simply float in the air over a firm bass. The tonal variety is almost intoxicating at times. I hope to try an OPA860 later but in the mean time this is superb. Pedja, if you are listening thankyou very much.
Tim.
Hello Tim,
If I remember well, the constant current source was at the input of the AD844 and it was there to compensate the 2mA DC offset at the output of the TDA1541. If you remove the current source, how can you compensate for that DC offset? Or does the AD844 work well even with that offset?
Best regards.
Paul
If I remember well, the constant current source was at the input of the AD844 and it was there to compensate the 2mA DC offset at the output of the TDA1541. If you remove the current source, how can you compensate for that DC offset? Or does the AD844 work well even with that offset?
Best regards.
Paul
Paul,
In fact I had two constant current sources, one on the op-amp input to null the offset, the other on the output to bias the op-amp into class-A (remaining from use with earlier voltage feedback op-amps which all sounded a little more refined to some degree as a result of this). Needless to say the offset nulling CCS on the op-amp input remains.
Tim.
In fact I had two constant current sources, one on the op-amp input to null the offset, the other on the output to bias the op-amp into class-A (remaining from use with earlier voltage feedback op-amps which all sounded a little more refined to some degree as a result of this). Needless to say the offset nulling CCS on the op-amp input remains.
Tim.
Hi TimA,
AD844 output section / buffer is designed to provide high slew rate. Its output stage current, and point of operation, was sent off by your 2mA current source and you have noticed it as harsh sound.
The input impedance is stable and would not change even if you start using the OP in a conventional manner and close the loop.
I have tried many configurations and would always go back to diamond configured AD844.
Boky
AD844 output section / buffer is designed to provide high slew rate. Its output stage current, and point of operation, was sent off by your 2mA current source and you have noticed it as harsh sound.
The input impedance is stable and would not change even if you start using the OP in a conventional manner and close the loop.
I have tried many configurations and would always go back to diamond configured AD844.
Boky
Boky,
With the ADD844 in common base mode the impedence seen by the preceeding DAC is high enough to exceed the +/-25mV voltage compliance specified in the TDA1541A datasheet by a considerable margin, I measure 130mV (peak). When used as conventional I/V stage my understanding is that the nfb causes a virtual earth to be seen by the DAC output, therefore removing the possibility for errors caused by the exceeded voltage compliance.
Tim.
With the ADD844 in common base mode the impedence seen by the preceeding DAC is high enough to exceed the +/-25mV voltage compliance specified in the TDA1541A datasheet by a considerable margin, I measure 130mV (peak). When used as conventional I/V stage my understanding is that the nfb causes a virtual earth to be seen by the DAC output, therefore removing the possibility for errors caused by the exceeded voltage compliance.
Tim.
George,
I tried to get the OPA860 working in my circuit but complications prevented completion.
Recently, I have been taking the output directly from pin 5 of the AD844, thereby bypassing its internal output buffer. The value of I/V resistor is now affected slightly by the impedance of the following attenuator, however the level of transparency has improved considerably.
Tim.
I tried to get the OPA860 working in my circuit but complications prevented completion.
Recently, I have been taking the output directly from pin 5 of the AD844, thereby bypassing its internal output buffer. The value of I/V resistor is now affected slightly by the impedance of the following attenuator, however the level of transparency has improved considerably.
Tim.
TimA said:
It is nice to see people still have interest in this design. As for the nfb - that is correct, the OP input pin, nfb resistor, and the OP output pin work as I to V... the "other resistor end" connected to the OP output, sits at virtual earth potential.
I still have to hear a pre-amp or DAC design where the buffer actually brings benefits to the sound, unless the interconnects / amp input impedance combination is so capacitive that you loose all the hf extension. I always bypass buffers as well.
Regards,
Boky
To overcome the AD844 high 50 ohms input impedance , this IC should be pushed to + - 18V at all times.
This IC needs high complementary pair current mirror' BIAS CURRENT to perform better - at lower distortions. Hence the reason for piano notes not sounding as natural as they should with + and - 15V supply. From the (very) limited internal circuit diagram I can not be 100% sure if the high Vcc / Vss actually increases this bias current. The fact, however, is that this IC sounds better if powered by absolute maximum of + and - 18V.
The output at pin 6 is 2V pp. Original 1.5kohms resistor will give you this peak output voltage X 1.01 at pin 5 - the buffer has voltage gain of 0.99
Regards,
Boky
This IC needs high complementary pair current mirror' BIAS CURRENT to perform better - at lower distortions. Hence the reason for piano notes not sounding as natural as they should with + and - 15V supply. From the (very) limited internal circuit diagram I can not be 100% sure if the high Vcc / Vss actually increases this bias current. The fact, however, is that this IC sounds better if powered by absolute maximum of + and - 18V.
The output at pin 6 is 2V pp. Original 1.5kohms resistor will give you this peak output voltage X 1.01 at pin 5 - the buffer has voltage gain of 0.99
Regards,
Boky
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.