Khom
02-12-2015, 13:41
My phonostage with balanced input based on schematic of Konstantin Musatoff described here: http://musatoffcv.narod.ru/Projects/RIAA5.htm , see also attached picture for convinience. The main features are:
- no common feedback loop
- passive correction
- relative high (+/-24 V) power supply.
- relative powerful output stage, so it should be non-sensitive to the interconnect
Unfortunately, there is no EN translation for device, so I will describe it in short.
R* - additional HF-correction for some cartridge with lack of HF
R7 - tau1 (75 us)
R26 - tau2 (318 us)
R29 - tau3 (3180 us)
R15, R21 - 7950 us
R12 - bias for output stage (Q9)
SW1 - switches between balanced and asymmetrical input
DA1 (low-noise opamp with MOSFET input, for example LF353) and components around it works as integrator and maintain DC zero output. As you see, all correction RC-circuits are passive, i.e. there is no frequency correction in feedback. Of course, output stage Q9 works in Class A.
So, it's basic schematic. I make some changes:
1. C3R7 excluded, I make tau1 correction (75 us) on input by reducing R3 and R4 to 3.1 kOhm each. Calculated value is (R3+R4)=6.2 kOhm for my AT-440MLa (moving magnet, approx. 480 mH inductance). I used this solution for tau1 with different schematic a long time ago since 80-th and it works well.
2. Integrator DA1 with it components removed, RIFA polypropylene 10 uF capacitor installed before R35. My opinion - one good PP capacitor is better than integrator circuit with a lot of active and passive details plus additional power supply for it.
3. 2SK184 replaced with 2SK170 (or 2SK117). Reason is enough simple - I didn't find here supplier for genuine 2SK184 :). Of course, all transistors in differential stages and in current mirror should be matched in pairs.
All capacitors in "sound" way are polystyrene ones, C7 polypropylene, other big caps C4, C6 are PTFE.
Phonostage powered by Two-transistor Shunt Regulator (http://www.tnt-audio.com/clinica/regulators_noise4_e.html), it has low noise at the wide frequency range and good performance regarding 50*n Hz. Regulator recalculated to 24 V, I used two identical ones with two separated Shottky rectifiers.
Sound. Unfortunately I'm rather technician than poet. It sounds light and airy, with soft deep bass. By the way, I don't see significand difference in sound with balanced and unbalances modes, maybe result will different with other interconnect from TT to PhS.
Used with upgraded Technics SL-1200Mk2, MM AT-440MLa. Tonearm rewired by litz, stock interconnect replaced with DIY Klotz MC5000 (twisted pair in common shield each lead, it's very suitable for this FS) and XLR connectors, external PSU, internal onboard regulator replaced with LM317, control board fully recapped, main bearing upgraded in first iteration... I see a long way ahead!)
15786
- no common feedback loop
- passive correction
- relative high (+/-24 V) power supply.
- relative powerful output stage, so it should be non-sensitive to the interconnect
Unfortunately, there is no EN translation for device, so I will describe it in short.
R* - additional HF-correction for some cartridge with lack of HF
R7 - tau1 (75 us)
R26 - tau2 (318 us)
R29 - tau3 (3180 us)
R15, R21 - 7950 us
R12 - bias for output stage (Q9)
SW1 - switches between balanced and asymmetrical input
DA1 (low-noise opamp with MOSFET input, for example LF353) and components around it works as integrator and maintain DC zero output. As you see, all correction RC-circuits are passive, i.e. there is no frequency correction in feedback. Of course, output stage Q9 works in Class A.
So, it's basic schematic. I make some changes:
1. C3R7 excluded, I make tau1 correction (75 us) on input by reducing R3 and R4 to 3.1 kOhm each. Calculated value is (R3+R4)=6.2 kOhm for my AT-440MLa (moving magnet, approx. 480 mH inductance). I used this solution for tau1 with different schematic a long time ago since 80-th and it works well.
2. Integrator DA1 with it components removed, RIFA polypropylene 10 uF capacitor installed before R35. My opinion - one good PP capacitor is better than integrator circuit with a lot of active and passive details plus additional power supply for it.
3. 2SK184 replaced with 2SK170 (or 2SK117). Reason is enough simple - I didn't find here supplier for genuine 2SK184 :). Of course, all transistors in differential stages and in current mirror should be matched in pairs.
All capacitors in "sound" way are polystyrene ones, C7 polypropylene, other big caps C4, C6 are PTFE.
Phonostage powered by Two-transistor Shunt Regulator (http://www.tnt-audio.com/clinica/regulators_noise4_e.html), it has low noise at the wide frequency range and good performance regarding 50*n Hz. Regulator recalculated to 24 V, I used two identical ones with two separated Shottky rectifiers.
Sound. Unfortunately I'm rather technician than poet. It sounds light and airy, with soft deep bass. By the way, I don't see significand difference in sound with balanced and unbalances modes, maybe result will different with other interconnect from TT to PhS.
Used with upgraded Technics SL-1200Mk2, MM AT-440MLa. Tonearm rewired by litz, stock interconnect replaced with DIY Klotz MC5000 (twisted pair in common shield each lead, it's very suitable for this FS) and XLR connectors, external PSU, internal onboard regulator replaced with LM317, control board fully recapped, main bearing upgraded in first iteration... I see a long way ahead!)
15786