High priced
So are servo circuits which sound better with high priced caps also. There are no free lunches guys. Things could be alot worse than one cap per output phase per channel. Use a reasonabley priced cap we have discussed this topic before as I remember.
H.H.
So are servo circuits which sound better with high priced caps also. There are no free lunches guys. Things could be alot worse than one cap per output phase per channel. Use a reasonabley priced cap we have discussed this topic before as I remember.
H.H.
good points everyone
if necessary, i can build protection circuitry to detect DC fault condition of a servo and mute (short) the outputs to protect the rest of the system. yes it is added complexity but it is easy to implement in a non-intrusive manner (just stick a JFET opamp at the output and feed a comparator trigger, should not degrade sound quality at all) and it would be a really trick setup when combined with the microprocessor control. when the servo blows up, i get a big message on my LCD panel saying "DC FAULT CONDITION, CHANNEL 1" and the channel gets muted, saving the day and looking slick in the process. it's so cool i'd almost want my servos to blow up just so i could see the protection circuit kick in! (i can hear the groans of agony from jocko already...)
the question is whether the servo itself will degrade the sound quality. my MSB Link DAC uses half of a dual opamp for the output stage DC servo and i think it sounds fine given that it an IC design, but i don't know how the servo influences the sound maybe
if necessary, i can build protection circuitry to detect DC fault condition of a servo and mute (short) the outputs to protect the rest of the system. yes it is added complexity but it is easy to implement in a non-intrusive manner (just stick a JFET opamp at the output and feed a comparator trigger, should not degrade sound quality at all) and it would be a really trick setup when combined with the microprocessor control. when the servo blows up, i get a big message on my LCD panel saying "DC FAULT CONDITION, CHANNEL 1" and the channel gets muted, saving the day and looking slick in the process. it's so cool i'd almost want my servos to blow up just so i could see the protection circuit kick in! (i can hear the groans of agony from jocko already...)
the question is whether the servo itself will degrade the sound quality. my MSB Link DAC uses half of a dual opamp for the output stage DC servo and i think it sounds fine given that it an IC design, but i don't know how the servo influences the sound maybe
For what is worth I find that servos imposes a sound on the amplifier which can be minimised by designing it with a low a time constant as possible, but what if drift is a problem?
I have tried servos before and it is not worth the time or effort. I notice that Nelson Pass (notice the capitals in his name ) does not use servos in his designs. Unecessary complexity and time which could be spent elsewhere.
Jam
I have tried servos before and it is not worth the time or effort. I notice that Nelson Pass (notice the capitals in his name ) does not use servos in his designs. Unecessary complexity and time which could be spent elsewhere.
Jam
aurora, i will take care of the control module. as you say it will be separate from the gain circuits. it should be completely abstracted so in the end we should be able to come up with whatever gain schemes we want and still use the same control module. we will need to define a control bus of course but that can be taken care of later. as i mentioned earlier, it should be perfectly possible for me to build 6 preamp modules, then send 2 of them to, oh, i dunno, harry, to try out with his control module. or more likely harry will build his own preamp modules and i'll send him my control module. or something like that.
back to the gain circuit! 🙂
back to the gain circuit! 🙂
jam said:
that is exactly what i wanted to hear. thanks for the insight jam.
how would you characterize the coloration of a servo? i'm curious how it compares to the coloration of a good cap (and i believe every cap does have a "sound").
aren't there tricks to minimize drift? good point that Pass does not use servos though. are all the X preamps cap-coupled? i notice a -3dB point of 2Hz on the X0.2. that would have to be a pretty big cap...
marc
p.s. i've been capitalizing Nelson's name lately! i swear!
I suppose the sound of the servo is that of the op-amp used. This is also true of power supply regulators that use op-amps for comparators. No kidding! You should try it. Different op-amps used in this application sound different.
Jam
Jam
wow, that is pretty cool jam. i believe you too. i always suspected that power supplies had a "sound" that was just as relevant as the sound of the signal circuit itself. well ok, in theory PSRR makes that a bit of an exaggeration but it does not take into account the dynamic performance of the supply. i've noticed some op-amp based regulators use the AD797 for its super low noise figure, but perhaps the sound of the opamp will impose a bit (i've heard the 797 tends to sound cold and sterile)...
i guess that is why some people prefer passive regulation... maybe the negatives associated with feedback in signal circuits also apply somewhat to active regulators?
i guess that is why some people prefer passive regulation... maybe the negatives associated with feedback in signal circuits also apply somewhat to active regulators?
jam said:
Yeahhrrr look at a Quad34 ... i own one of these and i cannot sell it!!!!!!!!!!!!!!
This is not one of my favorites!
Sonny
op-amps for comparators
Error amplifiers actually..... I gave up on op amp based regulators years ago.
Error amplifiers actually..... I gave up on op amp based regulators years ago.
The DC coupled line stage is probably 15 years old. I would stick with SE balanced line stage with is quite a new creation from Borberly.
i was just looking at that borberly line stage...
i went hunting for something DC coupled and was just about to ask you guys about that one. it is complementary however, in fact isn't it a discrete op amp? i have nothing against either but i thought you guys were die-hard SE zero-feedback fanatics... 😉 i bet it sound pretty good though... better than any IC at least.
ok, so i guess we will NOT be using opamp-based regulators for this design... passive regulation it is... i just wish it was more efficient... 🙄
p.s. i notice borberly calls his SE stage "very musical" but he doesn't use my preferred word, "accurate." not that the word "accurate" means much (strictly speaking, neither does "musical") although with preamps i think we are in a better position to judge accuracy by comparing with passive pre's or even straight wires.
i went hunting for something DC coupled and was just about to ask you guys about that one. it is complementary however, in fact isn't it a discrete op amp? i have nothing against either but i thought you guys were die-hard SE zero-feedback fanatics... 😉 i bet it sound pretty good though... better than any IC at least.
ok, so i guess we will NOT be using opamp-based regulators for this design... passive regulation it is... i just wish it was more efficient... 🙄
p.s. i notice borberly calls his SE stage "very musical" but he doesn't use my preferred word, "accurate." not that the word "accurate" means much (strictly speaking, neither does "musical") although with preamps i think we are in a better position to judge accuracy by comparing with passive pre's or even straight wires.
Oh really???!!!!
The schematic came out of May 2002 issue of Audio Electronics! I have been following Mr. Borbely's work for longer than 15 years.
THIS IS NOT A 15 YEAR OLD DESIGN..... maybe we need to sign you up for GRollins sensitivity training course. I posted it to show how rapidly a design's complexity could increase when you want everything included. It actual is not a bad design and Borbely is right on up there with the lastest parts and semiconductors. I watch him just as closely as Mr. Pass.
The schematic came out of May 2002 issue of Audio Electronics! I have been following Mr. Borbely's work for longer than 15 years.
THIS IS NOT A 15 YEAR OLD DESIGN..... maybe we need to sign you up for GRollins sensitivity training course. I posted it to show how rapidly a design's complexity could increase when you want everything included. It actual is not a bad design and Borbely is right on up there with the lastest parts and semiconductors. I watch him just as closely as Mr. Pass.
He's using "outstanding sound quality". In my book it may also qualify for accurate. If you guys could set some guidelines as to final topology I could build it over the weekend and say something about the sound. I would suggest PS rails around 30V.
just wanted to say...
...that you guys are doing a great job in this thread, and i appreciate it very much. i think we've got the S/N ratio much higher now and the creative juices are flowing more freely. although the active participants seem limited to a few "regulars" at the moment, i'm sure there are plenty of silent observers out there following the thread with interest. keep up the good work!
i think i will make a web page for this discussion soon - it would be nice to archive the comments on DC servos, active regulation, etc. they may seem a little off-topic at times but they are still very relevant to the project as a whole. i suspect it will all be enormously helpful information to countless DIYers out there.
thanks,
marc
...that you guys are doing a great job in this thread, and i appreciate it very much. i think we've got the S/N ratio much higher now and the creative juices are flowing more freely. although the active participants seem limited to a few "regulars" at the moment, i'm sure there are plenty of silent observers out there following the thread with interest. keep up the good work!
i think i will make a web page for this discussion soon - it would be nice to archive the comments on DC servos, active regulation, etc. they may seem a little off-topic at times but they are still very relevant to the project as a whole. i suspect it will all be enormously helpful information to countless DIYers out there.
thanks,
marc
Harry,
Well error amplifier then. I am with you and have not used them in years either.
We sometimes forget the power supply is part of the amplifier and is just as important as gain stage and sometimes probably more important.
Marc.
If direct coupling is so important to you it is possible to design a complementary version of the circuit that does not require a servo but you will have to use a fair amount of feedback to ensure low offset and drift. Matching devices also becomes critical then.
Jam
Well error amplifier then. I am with you and have not used them in years either.
We sometimes forget the power supply is part of the amplifier and is just as important as gain stage and sometimes probably more important.
Marc.
If direct coupling is so important to you it is possible to design a complementary version of the circuit that does not require a servo but you will have to use a fair amount of feedback to ensure low offset and drift. Matching devices also becomes critical then.
Jam
- Status
- Not open for further replies.