Money is not the object at this point - I would opt for minimalism only to avoid an audible sound-quality penalty. Hopefully the soundstage would be reasonably adaptable to a variety of equipment changes over time.
Might as well chime in, after all it is the NEW YEAR, the SALAS Phono Amp is very fine sounding and continues to make me very happy.
Had taken a break to listen to my highly modified SOEKRIS DAC, beyond what anyone on the VENDOR site is doing to correct the Vref problem and thought it sounded good.
Returned to SALAS last night (with new ASC (cap) cans underneath the board and there is no contest. Makes me almost sure the analogue advantage is insurmountable. My digital setup is tweaky and a pain in the *** to deal with - not a typical setup by any means and there is just no comparison that favors digital.
Not saying the cap mod is the reason - without the cap mod it would be the same. Have not had a chance to listen long enough to really know if the cap mod is worth the trouble - not that it was that much trouble - but to say the SALAS Phono Amp is supremely satisfying and worthy of using good parts.
Such an elegant design and even better, such a pleasure to listen to music through.
THANKS, yet again, for this great gift, Salas.
Had taken a break to listen to my highly modified SOEKRIS DAC, beyond what anyone on the VENDOR site is doing to correct the Vref problem and thought it sounded good.
Returned to SALAS last night (with new ASC (cap) cans underneath the board and there is no contest. Makes me almost sure the analogue advantage is insurmountable. My digital setup is tweaky and a pain in the *** to deal with - not a typical setup by any means and there is just no comparison that favors digital.
Not saying the cap mod is the reason - without the cap mod it would be the same. Have not had a chance to listen long enough to really know if the cap mod is worth the trouble - not that it was that much trouble - but to say the SALAS Phono Amp is supremely satisfying and worthy of using good parts.
Such an elegant design and even better, such a pleasure to listen to music through.
THANKS, yet again, for this great gift, Salas.
I'll chime in too with the New Year wishes, and say that the FSP has given me great pleasure in the last year. I have been visited by old friends and family members this year and we spun some disks, and the visitors seemed uniformly shocked at the terrific sound from some old (and new) albums. My tt setup has continued to slowly evolve and gets closer to where I want it.
Now I just need to dig out my copy of "Hunky Dory" to say goodbye to one of the great talents of our time. BTW, my copy of that album is a very good Japanese pressing with a short but deep scratch on "Charges", but the FSP handles transient spikes so well that I can listen through it. Thanks again for a well designed phono stage!
Now I just need to dig out my copy of "Hunky Dory" to say goodbye to one of the great talents of our time. BTW, my copy of that album is a very good Japanese pressing with a short but deep scratch on "Charges", but the FSP handles transient spikes so well that I can listen through it. Thanks again for a well designed phono stage!
Happy new year everybody,
I had Teabag kit in my possession for the longest time but, alas, no time to put it together.
Finally in the past days I got a break from work and I assembled it...
Problems. I got the infamous selectronic r-core 18V 30VA transformers and hooked everything up BEFORE I read the issues some other fellows had with them and the high voltage they produce. Of course, DC In was very high (59V)...
On the other hand, everything turned on according to instruction and readings on one channel were normal but on the other channel I wasn´t able to bring rail+ to gnd higher than 29V using VR1.
This morning I have removed the selectronics and installed a pair of generic toroids from farnell. Still high but I believe within range (53,1V and 52,9V). These are 18V, I have some 15V that might work better.
Channel 1 is fine I think. Adjusting VR1 I got this values:
Rail+ / GND: 32,1V
TP1 / GND: 7,5V
TP2 / GND: 3.82V
TP1 / TP2: 3,6V
Channel 2 continues to be problematic. I got to the limit of VR1 and I cannot bring Rail+ / GND to range. These are the readings:
Rail+ / GND: 29,7V
TP1 / GND: 7,63V
TP2 / GND: 1,53V
TP1 / TP2: 6,11V
I appreciate in advance any help or hint that I could get from you guys as my electronics knowledge is very limited.
Can´t wait to hear her on my system....
Regards,
Luis
I had Teabag kit in my possession for the longest time but, alas, no time to put it together.
Finally in the past days I got a break from work and I assembled it...
Problems. I got the infamous selectronic r-core 18V 30VA transformers and hooked everything up BEFORE I read the issues some other fellows had with them and the high voltage they produce. Of course, DC In was very high (59V)...
On the other hand, everything turned on according to instruction and readings on one channel were normal but on the other channel I wasn´t able to bring rail+ to gnd higher than 29V using VR1.
This morning I have removed the selectronics and installed a pair of generic toroids from farnell. Still high but I believe within range (53,1V and 52,9V). These are 18V, I have some 15V that might work better.
Channel 1 is fine I think. Adjusting VR1 I got this values:
Rail+ / GND: 32,1V
TP1 / GND: 7,5V
TP2 / GND: 3.82V
TP1 / TP2: 3,6V
Channel 2 continues to be problematic. I got to the limit of VR1 and I cannot bring Rail+ / GND to range. These are the readings:
Rail+ / GND: 29,7V
TP1 / GND: 7,63V
TP2 / GND: 1,53V
TP1 / TP2: 6,11V
I appreciate in advance any help or hint that I could get from you guys as my electronics knowledge is very limited.
Can´t wait to hear her on my system....
Regards,
Luis
If you are using a CCS/Shunt style voltage regulator, then the load on the transformer and PSU is constant.
That suits changing the PSU from an rC that nearly everyone uses, to a rCRC, or even to rCRCRC.
The R is an added resistance between the capacitor stages.
At a constant load these R will drop a constant voltage.
Say you want 50Vdc from a supply that is currently 59Vdc and you have set the CCS of the regulator to 300mA.
To drop 9V @ 300mA you need an added resistance of 9V/0.3A = 30r
You could use an rCRC of 1mF + 30r + 1mF, or an rCRCRC of 1mF +15r + 1mF +15r +1mF
This second will have slightly lower ripple at the input to the regulator. This remaining ripple when viewed on a scope will be more rounded than that from a basic rC type PSU, i.e. the remaining ripple has less of the harmonics.
BTW,
three 1mF 63V capacitors will give less ripple and much less harmonics than a single 4700uF 63V and will probably cost less than a big single. And maybe better than using a 10mF 63.
That suits changing the PSU from an rC that nearly everyone uses, to a rCRC, or even to rCRCRC.
The R is an added resistance between the capacitor stages.
At a constant load these R will drop a constant voltage.
Say you want 50Vdc from a supply that is currently 59Vdc and you have set the CCS of the regulator to 300mA.
To drop 9V @ 300mA you need an added resistance of 9V/0.3A = 30r
You could use an rCRC of 1mF + 30r + 1mF, or an rCRCRC of 1mF +15r + 1mF +15r +1mF
This second will have slightly lower ripple at the input to the regulator. This remaining ripple when viewed on a scope will be more rounded than that from a basic rC type PSU, i.e. the remaining ripple has less of the harmonics.
BTW,
three 1mF 63V capacitors will give less ripple and much less harmonics than a single 4700uF 63V and will probably cost less than a big single. And maybe better than using a 10mF 63.
Last edited:
Andrew, thank you for the tips on the power supply. Much appreciated. My space now is limited but I might try the big caps with the selectronics trannies in a bigger case on the future...
For now the biggest problem seems to be in the other case, where one of the channels is not biasing properly...
For now the biggest problem seems to be in the other case, where one of the channels is not biasing properly...
Yes use the alternative 9.1K. It looks like that the specific involved JFET Q3x is weaker than in the other channel (not a problem its just they vary for IDSS by wide enough production tolerances endemic in a JFET manufacturing process). More resistance counterbalances the I to V conversion level in the Vref for relatively deficient Rail+ max (Norton type resistive low noise Vref used here)
Done and working! Now letting it warm up for fine tuning. Thank you Salas, and thank you of course for your contributions to the DIY world.
Sound report to follow when everything is installed.
This might sound silly but does anyone have a mouser part number for a nice chassis ground screw? I´m searching but mouser is vast and I can´t figure what to look for...
Thank you!
Sound report to follow when everything is installed.
This might sound silly but does anyone have a mouser part number for a nice chassis ground screw? I´m searching but mouser is vast and I can´t figure what to look for...
Thank you!
The relative problem is mainly the excess heat on the CCS MOSFET sink this voltage level is going to bring. Not a no go area but not well allocated heat for boxing up and long term reliability or high voltage hours mains provider tolerance. I suggest you put a 20 Ohm 5W resistor between your bridge rectifier + and main filter capacitor's + pin. That is going to absorb some excess voltage and heat as well as contributing an EMI filter. Its a Win Win situation. You could even utilize your R-Cores by experimenting with even higher value resistor, maybe 33 Ohm 5W. Not sure on the value prediction because of the ? regulation % in the particular Tx models, but have in mind that anything 43-48V Raw DC makes good Vin-Out margin for your 32-33V Rail+. For the time being if not having such power resistors utilize your 15V+15V transformers.