The 4.7K actually do come into play after the 100uf caps and do effect the corner frequency.
Had a second look and I see it now: 4K7 + 4K7 in parallel with the 100K.
Yes. The CCS is the place to do it if one is using balanced output. Of course some amps /balanced input stages will have the facility to null offset, but transformers don't so it's a good idea in order to avoid core saturation. Drifts a little with temp, but does the job.
If you omit the relay and the BAL/SE stage and want to use high output swing (say 4VMRS) then if you increase the resistors accordingly you can actually run the legato with rails around ~20V with no issues and somewhat increased dynamic range. But please only ever try something like this is you are absolutely confident in what you are doing.
Cheers!
Russ
Readers on this thread may recall that I was playing with Legato output swing as a way to tweak dynamics to my system's very particular needs. After finalizing my settings and enjoying listening for a while, I found another very surprising way to augment response dynamics and for a very minimal investment. It took my 'mostly TPA' rig to a completely new level. I had no inkling of its potential. My learning experiences are chronicled in a conversation beginning with: http://www.diyaudio.com/forums/mult...ng-impressions-techniques-56.html#post2428295
I mention it with appreciation here because my results would not have been possible without the *superlative design and performance* of B2/Legato and my other TPA stuff. Post queries over there, not here...
Frank
Since it's you, Russ, I'll post here once more.
I've never heard any digital source sound like this. I'd like to sit you where I'm sitting, hand you a lovely beer, and watch your face when some wicked transients or typically muddy, turgid signal comes along in the music. Hearing harmonics in the piano's *hammers* as well as the strings? No sweat.
Overture amps w/ 24 V transformers put out, what, 100 W @ 8 ohms? Pfffff... Mono Opus -> Overture -> goofy painted JBL Pro 15" drivers sound cleaner and tighter than amps rated at 4-5X the Overture output.
I entertained friends and played (among other things) some 96/24 Minnesota Orch. recordings made locally by RR. The guests knew the concert hall well. We all agreed from experience that B2/Legato2 -> (different) Overture* -> goofy painted JBL Pro midrange and tweeter sounded better than sitting in a live audience where the recording was made. *[25 V supply and 1/2" solid silver inter-board connections.]
I used to rant (and my evidence supported the idea) that redbook engineering generally sucked. That's why I was all gung-ho to get into HD source files. And they were/are sweeter, of course. But I was deceived about the information in 44/16 music files. All those tunes recently rose from the dead. I didn't realize that the real life behind the fidelity is, like, ~ -30 db below all the detailed 'up front' sound which we are constantly striving to improve.
B2/Legato2 puts out *amazingly* realistic background ambiance if you can transduce it. The low noise balanced amps aren't hurting either... It's just that the redbook I/O through B2 and the balanced amps was causing ambiance-canceling and dynamic-canceling "surface distortion" in MY transducer cones (which, unmodified, were some of the best transient reproducers I had previously heard!). Reproduction of HD files is improved too - but not improved as much as the redbook. The extent to which "YMMV" in this transducer tweaking technology has to be *extreme* - maybe not unlike carving a cello, to my mind. The background science is sketchy and might be more related to auditory system neuroscience than EE. So be it.
A good point to stop on is this: with B2/Legato to supply the experimental reference signal, one would have the best possible shot for success in tweaking transducers. OK. Back to your scheduled programming.
Frank
I've never heard any digital source sound like this. I'd like to sit you where I'm sitting, hand you a lovely beer, and watch your face when some wicked transients or typically muddy, turgid signal comes along in the music. Hearing harmonics in the piano's *hammers* as well as the strings? No sweat.
Overture amps w/ 24 V transformers put out, what, 100 W @ 8 ohms? Pfffff... Mono Opus -> Overture -> goofy painted JBL Pro 15" drivers sound cleaner and tighter than amps rated at 4-5X the Overture output.
I entertained friends and played (among other things) some 96/24 Minnesota Orch. recordings made locally by RR. The guests knew the concert hall well. We all agreed from experience that B2/Legato2 -> (different) Overture* -> goofy painted JBL Pro midrange and tweeter sounded better than sitting in a live audience where the recording was made. *[25 V supply and 1/2" solid silver inter-board connections.]
I used to rant (and my evidence supported the idea) that redbook engineering generally sucked. That's why I was all gung-ho to get into HD source files. And they were/are sweeter, of course. But I was deceived about the information in 44/16 music files. All those tunes recently rose from the dead. I didn't realize that the real life behind the fidelity is, like, ~ -30 db below all the detailed 'up front' sound which we are constantly striving to improve.
B2/Legato2 puts out *amazingly* realistic background ambiance if you can transduce it. The low noise balanced amps aren't hurting either... It's just that the redbook I/O through B2 and the balanced amps was causing ambiance-canceling and dynamic-canceling "surface distortion" in MY transducer cones (which, unmodified, were some of the best transient reproducers I had previously heard!). Reproduction of HD files is improved too - but not improved as much as the redbook. The extent to which "YMMV" in this transducer tweaking technology has to be *extreme* - maybe not unlike carving a cello, to my mind. The background science is sketchy and might be more related to auditory system neuroscience than EE. So be it.
A good point to stop on is this: with B2/Legato to supply the experimental reference signal, one would have the best possible shot for success in tweaking transducers. OK. Back to your scheduled programming.
Frank
Well Frank,
I am sure I would be glad to take you up if I were close by. I love to visit with people of like mind. Beer is just a nice bonus.
Many people know that I have a few friends in the recording/production industry here in Nashville. I also have couple of dear musician friends. I lean on them heavily for two things: High quality recorded material, and unbiased and trained criticism. I think those two things have given me a unique perspective on the things I design.
One thing we (the collective we meaning me and all of my friends) noticed right off with B2 + Legato was how absolutely stunning the dynamic range could really be. Black silence to incredible attack. If you have a power amplifier and transducers capable of reproducing it.
You can put you ear right up to some of my 100db speakers and still hear nothing but silence when it should be so. This with no preamp. Nothing to attenuate any noise. There simply is none to worry over.
Once again, thanks for reminding me why all of this is worth it.
Cheers!
Russ
I am sure I would be glad to take you up if I were close by. I love to visit with people of like mind. Beer is just a nice bonus.
Many people know that I have a few friends in the recording/production industry here in Nashville. I also have couple of dear musician friends. I lean on them heavily for two things: High quality recorded material, and unbiased and trained criticism. I think those two things have given me a unique perspective on the things I design.
One thing we (the collective we meaning me and all of my friends) noticed right off with B2 + Legato was how absolutely stunning the dynamic range could really be. Black silence to incredible attack. If you have a power amplifier and transducers capable of reproducing it.
You can put you ear right up to some of my 100db speakers and still hear nothing but silence when it should be so.
This with no preamp. Nothing to attenuate any noise. There simply is none to worry over.Once again, thanks for reminding me why all of this is worth it.
Cheers!
Russ
Last edited:
One thing we (the collective we meaning me and all of my friends) noticed right off with B2 + Legato was how absolutely stunning the dynamic range could really be. Black silence to incredible attack. If you have a power amplifier and transducers capable of reproducing it.
Yes sir. It's such a piece of work and now no output caps!
It's wonderful when the reproduced range of subtle detail references program dynamics to the ambiance of the space where recordings are made, rather than mere 'blackness' in one's room... In summary
Seems to me that I did read everything... I have my Legato II in basic configuration with SE installed and Buffalo 24 and before changing something propose to get mods in summary. Let's say thats way (schematic attached) :
1) Change R1-R4 resistors value from 150R to 187R to get ~2.0v RMS output. R9-R16 remains, so we have 187R on pos. side and 340,5R on neg. ( 681Rx2 in parallel ).
Question: Is changing PS to +15/-13 v gave the same effect? If NOT then --->
2) Change +/- 15v supply to +15/-13 v - what is the main benefit? What is the optimum for power supplies and resistors values in case of "musicallity" ? Also what we are talking about: PS to Left/Right balanced outputs only or SE too. In other words J1-J4 jumpers installed?
3) Omitting C1-C4, as has been told here, give us "openess" and something else like "glassy". It's understandable, cos' in this case all harmonics are welcomed to our next stage. Don't think it's necessary and in anyway strongly depends from the next stage ( i.e. tubes, opamps & etc.).
Question: Replacing with 1n2 or whatever else still recomended or not?
4) Coupling caps C5-C8 in case of next stage BALANCED load not less than 10kOm can be replaced with 4,7-10uF so we can use film capacitors here.
Question: In which case we have to leave 100uF in place - if load less than 10kOm or more than xx ? For example, I have 10kOm balanced input in my Genelec HT208 active monitors and SE stage in Legato II installed, so if I'm right there will be ~ 15kOm load totally. I'm still OK with 4.7-10uF for C5-C8?
5) Replace FET transistors with Supertex TP0604 in reverse. This is really did the trick as I'm understood.
Question: So finally how many FETs gave better results? As in original scheme?
6) Something else missed?
Thank you,
Valery.
P.S. Russ & Brian, thank you for your exellent job! Just one kind entreaty. When you'll finally made Legato X & Bugffalo VIII versions and there will be nothing to do else, please, make one with synchronous/masterclock input...
Seems to me that I did read everything... I have my Legato II in basic configuration with SE installed and Buffalo 24 and before changing something propose to get mods in summary. Let's say thats way (schematic attached) :
1) Change R1-R4 resistors value from 150R to 187R to get ~2.0v RMS output. R9-R16 remains, so we have 187R on pos. side and 340,5R on neg. ( 681Rx2 in parallel ).
Question: Is changing PS to +15/-13 v gave the same effect? If NOT then --->
2) Change +/- 15v supply to +15/-13 v - what is the main benefit? What is the optimum for power supplies and resistors values in case of "musicallity" ? Also what we are talking about: PS to Left/Right balanced outputs only or SE too. In other words J1-J4 jumpers installed?
3) Omitting C1-C4, as has been told here, give us "openess" and something else like "glassy". It's understandable, cos' in this case all harmonics are welcomed to our next stage. Don't think it's necessary and in anyway strongly depends from the next stage ( i.e. tubes, opamps & etc.).
Question: Replacing with 1n2 or whatever else still recomended or not?
4) Coupling caps C5-C8 in case of next stage BALANCED load not less than 10kOm can be replaced with 4,7-10uF so we can use film capacitors here.
Question: In which case we have to leave 100uF in place - if load less than 10kOm or more than xx ? For example, I have 10kOm balanced input in my Genelec HT208 active monitors and SE stage in Legato II installed, so if I'm right there will be ~ 15kOm load totally. I'm still OK with 4.7-10uF for C5-C8?
5) Replace FET transistors with Supertex TP0604 in reverse. This is really did the trick as I'm understood.
Question: So finally how many FETs gave better results? As in original scheme?
6) Something else missed?
Thank you,
Valery.
P.S. Russ & Brian, thank you for your exellent job! Just one kind entreaty. When you'll finally made Legato X & Bugffalo VIII versions and there will be nothing to do else, please, make one with synchronous/masterclock input...
Last edited:
OK, Russ, I understand. Well, here is my integration of Buffalo24/Legato into Adcom GDA-700 DAC. As Ultraanalog chip inside Adcom was broken I'm decided to make this exchange. This is not the final version, I'm also tried transformers as I/V but Legato works very well to me, just have to increase output level up to 2 v. So that's why I'm posted "summary' to make all changes before final intergration.
Attachments
PCM1702 and Legato 2.
Hi,
I have a Kenwood DP-7090 CD player, and are now using the Legato 2 I/V stage.
The DP-7090 has 2x4 PCM1702 (change to K-Grade) so 4 DAC for each channel.
The DAC´s in each channel are in a balanced configuration, 2 DAC´s goes positive and 2 DAC´s go negative.
Kenwood uses 32x oversampling, so each DAC is shifted 1/32 x FS, but the current output from the DAC´s are summed after the I/V conversion in the original I/V opamp design.
For long time I have been using AD797 for the I/V conversion in this DAC, and I think the AD797 made a great clean sound.
Then I read about the Legato 2 circuit, and decided to give it a try.
I notice that the Legato 2 was designed for a DAC with much higher output current than the PCM1702´s +/- 1.2mA (twice in dual config)
But I first tried it with the original resistor configuration, but it gave to little voltage out.
Then I modified the Bal/SE Opamp resistors, so I got 2Vrms out. This sounded really great !
Well read some more on the DIY forum, and decided to change the resistors in the I/V stage. Here is the values I ended up with:
V+= 12V 4.5AH Battery
V-=12V 4.5AH Battery
R1 = 680 ohm
R5 = 1K (unchanged)
R9//R10 = single 1.5K
Vdc at I-input 8mV DC.
These resistors gives the same DC voltage in the stage, as the orignal resistors.
These larger resistors reduces the current to about 8.2mA in each stage.
Now I have about 1.6 Vrms signal at R1, which is perfect. I change the BAL/SE opamp resistors so I get 2.3 Vrms out.
I have measured the SE/BAL output THD @1kHz 0dBFS , and the this is pretty awesone, 2.H is below -111dB 3.H = -107dB
I use the ASUS Xonar STX as my analyser card (Input THD+N at 1kHz:0.0002% /-113dB for Line-in)
The Legato 2 in this configuration sounds very dynamic and open. I particular like the highfrequency better than from the AD797!
I was wondering if I should change the R5 resistor ? Would I gain anything to make this bigger/smaller ?
PS.
In my Kenwood DP7090 I have also changes to the Slow rolloff filter option on the NPC SM5843 8x Oversampling filter.
This gives NO pre or post ringing in a square wave, only a small under/overshoot can be seen on the scope.
It is also possible to turn off the 32x oversampling and DRIVE circuit, and only run it 8X Oversampling, but I think Kenwoods 32FS sounds better.
Hi,
I have a Kenwood DP-7090 CD player, and are now using the Legato 2 I/V stage.
The DP-7090 has 2x4 PCM1702 (change to K-Grade) so 4 DAC for each channel.
The DAC´s in each channel are in a balanced configuration, 2 DAC´s goes positive and 2 DAC´s go negative.
Kenwood uses 32x oversampling, so each DAC is shifted 1/32 x FS, but the current output from the DAC´s are summed after the I/V conversion in the original I/V opamp design.
For long time I have been using AD797 for the I/V conversion in this DAC, and I think the AD797 made a great clean sound.
Then I read about the Legato 2 circuit, and decided to give it a try.
I notice that the Legato 2 was designed for a DAC with much higher output current than the PCM1702´s +/- 1.2mA (twice in dual config)
But I first tried it with the original resistor configuration, but it gave to little voltage out.
Then I modified the Bal/SE Opamp resistors, so I got 2Vrms out. This sounded really great !
Well read some more on the DIY forum, and decided to change the resistors in the I/V stage. Here is the values I ended up with:
V+= 12V 4.5AH Battery
V-=12V 4.5AH Battery
R1 = 680 ohm
R5 = 1K (unchanged)
R9//R10 = single 1.5K
Vdc at I-input 8mV DC.
These resistors gives the same DC voltage in the stage, as the orignal resistors.
These larger resistors reduces the current to about 8.2mA in each stage.
Now I have about 1.6 Vrms signal at R1, which is perfect. I change the BAL/SE opamp resistors so I get 2.3 Vrms out.
I have measured the SE/BAL output THD @1kHz 0dBFS , and the this is pretty awesone, 2.H is below -111dB 3.H = -107dB
I use the ASUS Xonar STX as my analyser card (Input THD+N at 1kHz:0.0002% /-113dB for Line-in)
The Legato 2 in this configuration sounds very dynamic and open. I particular like the highfrequency better than from the AD797!
I was wondering if I should change the R5 resistor ? Would I gain anything to make this bigger/smaller ?
PS.
In my Kenwood DP7090 I have also changes to the Slow rolloff filter option on the NPC SM5843 8x Oversampling filter.
This gives NO pre or post ringing in a square wave, only a small under/overshoot can be seen on the scope.
It is also possible to turn off the 32x oversampling and DRIVE circuit, and only run it 8X Oversampling, but I think Kenwoods 32FS sounds better.
FFT plot of my Kenwood + Modified Legato2
Here is a FFT plot of a 0dBFS 1KHz sine from my Kenwood DP7090 with 2x4 PCM1702K and a modified Legato 2 I/V stage, run from a +/-12V battery supply.
Pretty amazing THD of 0.00067 %
And it sounds amazing too !!
Here is a FFT plot of a 0dBFS 1KHz sine from my Kenwood DP7090 with 2x4 PCM1702K and a modified Legato 2 I/V stage, run from a +/-12V battery supply.
Pretty amazing THD of 0.00067 %
And it sounds amazing too !!
Attachments
Thats not bad at all.
It not quite as good as the standard (ESS) config because there is not enough current through the devices. But those are still very nice numbers.
Thanks for posting the success!
You are actually running the current pretty low. If you are not using the BAL/SE section of the Legato you can increase both it's CCS current and its working voltage (say up to +/- 24V) . This will get you somewhat better THD in the config you are using. You actually only need to increase the positive rail, the negative rail could stay the same, just adjust the current to bring the static bias voltage at about half the positive rail.
You can also increase R5 about double because you are running so little current.
Good work!
Cheers!
Russ
It not quite as good as the standard (ESS) config because there is not enough current through the devices. But those are still very nice numbers.
Thanks for posting the success!
You are actually running the current pretty low. If you are not using the BAL/SE section of the Legato you can increase both it's CCS current and its working voltage (say up to +/- 24V) . This will get you somewhat better THD in the config you are using. You actually only need to increase the positive rail, the negative rail could stay the same, just adjust the current to bring the static bias voltage at about half the positive rail.
You can also increase R5 about double because you are running so little current.
Good work!
Cheers!
Russ
Hi Russ,
Please, follow:
In standart configutation R1=150, R9=340, PS (BachAudioDK) = +/-12v
Were changed to R1=680, R9=1,5k, PS (Russ) = +/-24v
In this case the ratio between R9/R1 remains and = ~ 2.2. As the power supply doubled, currents also remains or I'm not right? What is the main benefit?
Last edited:
Hello,
Just in case that next to Legato stage also play game in low pass filter.
BR,
Valery
I'd like to get a little clarification on this thread for those of us who are not so knowledgeable regarding the best possible configuration of a Legato2 running Dual Mono Buff 2's.
I'm planning to use only the balanced output of the Legato2 into my PrePro which is an Integra DHC80.2. The XLR input on the Integra I don't believe is a fully balanced circuit from end to end so perhaps I'd be better off just using the SE/ballsie stage on the Legato2 but will take opinions on the matter.
Second I've installed R187 resistors for R1-4 ... I was trying to achieve 2V output however that was before I understood the implications of running a single Legato2 with Dual Mono Buff 2's ... so from this thread I understand that R1-4 should be 75R or perhaps even lower such as 68ohm may be best...if I go this route in Legato 2 would I replace R21-24 with 2.2 times the value of R1-4? Such that if I use 68ohm in 1-4 I'd use 150ohm in R21-24...or does this apply only to Legato 1?
Last I've managed to have some left over .1uF Mundorf Silver Oil caps from another project. Would it be advisable / advantageous to replace any caps in the Legato 2 circuit with these?
This is a repost of one I made in the dual buff 1 legato thread so apologies for that but thought I'd post it here as well as more folks seem involved in this thread.
Best Regards,
Theo
I'm planning to use only the balanced output of the Legato2 into my PrePro which is an Integra DHC80.2. The XLR input on the Integra I don't believe is a fully balanced circuit from end to end so perhaps I'd be better off just using the SE/ballsie stage on the Legato2 but will take opinions on the matter.
Second I've installed R187 resistors for R1-4 ... I was trying to achieve 2V output however that was before I understood the implications of running a single Legato2 with Dual Mono Buff 2's ... so from this thread I understand that R1-4 should be 75R or perhaps even lower such as 68ohm may be best...if I go this route in Legato 2 would I replace R21-24 with 2.2 times the value of R1-4? Such that if I use 68ohm in 1-4 I'd use 150ohm in R21-24...or does this apply only to Legato 1?
Last I've managed to have some left over .1uF Mundorf Silver Oil caps from another project. Would it be advisable / advantageous to replace any caps in the Legato 2 circuit with these?
This is a repost of one I made in the dual buff 1 legato thread so apologies for that but thought I'd post it here as well as more folks seem involved in this thread.
Best Regards,
Theo
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.