Two winding
To carlosfm : Two winding and two bridges is nonsens in case, when you have only ONE transformer and moreover DC drop is twice higher than with using only one bridge. My PCB is correctly designed and amp have SNR in rank near 120 dB - can you more ? Stabilised DC rails : yes, it will be probably better, 'cos LM's have unsymetrical PSR, but I was designed simply amp, not any monster
, 'cos in my opinion it was basic thought by developing this IC ( with this circuit you can't to do " new version of ML's 20.5 " ). If somebody don't can to use output filters, have here this possibility too : look at output resistors : 10*0.6 W = 6 W, that is the same, as you recomended . And at last : shematic is without component value, 'cos it is preliminary schematic, used by designing of PCB for control layouts. Schematic with values come ASAP .
To carlosfm : Two winding and two bridges is nonsens in case, when you have only ONE transformer and moreover DC drop is twice higher than with using only one bridge. My PCB is correctly designed and amp have SNR in rank near 120 dB - can you more ? Stabilised DC rails : yes, it will be probably better, 'cos LM's have unsymetrical PSR, but I was designed simply amp, not any monster
, 'cos in my opinion it was basic thought by developing this IC ( with this circuit you can't to do " new version of ML's 20.5 " ). If somebody don't can to use output filters, have here this possibility too : look at output resistors : 10*0.6 W = 6 W, that is the same, as you recomended . And at last : shematic is without component value, 'cos it is preliminary schematic, used by designing of PCB for control layouts. Schematic with values come ASAP .
Upupa Epops,
The dual secondaries and bridges are MANDATORY, I was talking about the LM338, a POSITIVE reg for both rails.
No negative reg as good as this, so...
Each one to his own, I was giving suggestions to improve the SOUND of the amp, take it or leave it.
I see a zobel there and I suggested you to take all that thing out and just use the resistors in series with the output.
The value I gave you IS the optimal (after many HOURS of testing zobels and series resistors WITH difficult cables and speakers), and I'm not talking wattage (nonsense!) but the resistors' value.
No resistors/caps from the output to ground, please!
I think I'll change attitude and keep my secrets to my own.
Thanks.
The dual secondaries and bridges are MANDATORY, I was talking about the LM338, a POSITIVE reg for both rails.
No negative reg as good as this, so...
Each one to his own, I was giving suggestions to improve the SOUND of the amp, take it or leave it.
I see a zobel there and I suggested you to take all that thing out and just use the resistors in series with the output.
The value I gave you IS the optimal (after many HOURS of testing zobels and series resistors WITH difficult cables and speakers), and I'm not talking wattage (nonsense!) but the resistors' value.
No resistors/caps from the output to ground, please!
I think I'll change attitude and keep my secrets to my own.
Thanks.
Dual secondaries
To carlosfm : I'm sorry Carlos, I don't read your words attentively - yes, if you use for stabilisation positive regulators for both rails, you must to use separate windings and two bridges. Two 0.47R in parallel give 0.235 R, which ( in my opinion ) too much reduce damping factor ( only 17 ref. to 4 Ohm ) and amp's sound approach to the tube ones. This is maybe reason of your prefering of this solution.
To carlosfm : I'm sorry Carlos, I don't read your words attentively - yes, if you use for stabilisation positive regulators for both rails, you must to use separate windings and two bridges. Two 0.47R in parallel give 0.235 R, which ( in my opinion ) too much reduce damping factor ( only 17 ref. to 4 Ohm ) and amp's sound approach to the tube ones. This is maybe reason of your prefering of this solution.
Carlos, do that!carlosfm said:I think I'll change attitude and keep my secrets to my own.
Thanks.
.... if you can.... Don't forget that a design is sometimes like a painting. We all have different tastes. If you don't like Pavel's ideas about the output resistors, just skip them. It's not harder than that.
I have lot's of options in my Gainclone and I let people choose themselves. They can do whatever they pleases but I support only 6 versions.
Is it a fry yet?Re: Dual secondaries
Actually it's the only way that this amp will work with any speaker cable without problems AND without mucking the sound.
It's the only way to use my 5-meter pair of Kimber 4TC with my difficult to drive Epos11 speakers.
You all know the GC doesn't work well with many cables.
Why does 47 Labs use that special thin cable?
This way it works, it sounds very good, no compromize at all, and your amp will not be picky with cables.
Again, I'm talking with the knowledge of who had a problem. tried plenty of combinations and picked the best one.
A suggestion for who whants to use it, nothing more.
Have in mind that I just made some suggestions.
It's up to you.
Upupa Epops said:
Actually it's the only way that this amp will work with any speaker cable without problems AND without mucking the sound.
It's the only way to use my 5-meter pair of Kimber 4TC with my difficult to drive Epos11 speakers.
You all know the GC doesn't work well with many cables.
Why does 47 Labs use that special thin cable?
This way it works, it sounds very good, no compromize at all, and your amp will not be picky with cables.
Again, I'm talking with the knowledge of who had a problem. tried plenty of combinations and picked the best one.
A suggestion for who whants to use it, nothing more.
Have in mind that I just made some suggestions.
It's up to you.
To carlosfm : Certainly, you can to have less problems with cable, which have lower summary diameter of copper, than with " fat " ones. Problem with cables indicate low stability of an amp, but cure low stability by series resistor is not, in my opinion, thrue best way. I had try this amp with RC load ( from cca 100 pF up to several microfarad in all decades ) and was all the time stabil. About speaker cables : after many listening tests, which we do here, absolutely best results give hf power coaxials, such as Andrew and similars, belive or not .
.Kimber cables are mid to high capacitance.
A series resistor is used on any op-amp's output to couple with this.
Why do some speaker cables work better than others on a given amp/speaker combo?
Without the output resistors a pair or cheap Supra 2.5 sounded better.
With the output resistors the Kimber sounds much better, as it should, as it costs 10 times more!
I found this sollution better than zobels.
Anyway, check post #62 here:
http://www.diyaudio.com/forums/showthread.php?s=&postid=187335&highlight=#post187335
A series resistor is used on any op-amp's output to couple with this.
Why do some speaker cables work better than others on a given amp/speaker combo?
Without the output resistors a pair or cheap Supra 2.5 sounded better.
With the output resistors the Kimber sounds much better, as it should, as it costs 10 times more!
I found this sollution better than zobels.
Anyway, check post #62 here:
http://www.diyaudio.com/forums/showthread.php?s=&postid=187335&highlight=#post187335
carlosfm said:
roibm said:
??? why ???
It seems like designer here made his mind already about all the choices and it's pretty much useless trying to convince him otherwise. It also represents certain design style and audio approach philosophy. Nothing wrong with that, as long as it works for him. It is very conservative approach, with a big concern on safety of operation and proper interfacing. It seems like sonic perspective is a little overshadowed by that, but at least it gives everybody a peace of mind