Don't use impulse power supplies in final version of amplifier, only in test version.
TIP2955/TIP3055 TO-220 from ST - so badly BJT. If you want to use low cost BJT in output - I recommending "to look" on KEC 2SD718+2SB688.
Yes, sometimes it have a little problems with DC out, but it's not fatal, you can to change trimmer 100 om on 220 om or 470 om, but Ic will be smaller, you can play with risistors in collectors of output (1 om in stock) and resistors in small BJT collectors (1k in stock).
If DC out 100 mV or less - this is not problem.
Thanks for the tip. The transistor choice still confuses me, there have been many posts with different schematics and also modern version le monstre circuit diagrams in this post. Are these based on (Spice) simulations or on actual amplifier builds? Are the input or output transistors more critical for the sound? Is there a overview table comparing different le monstre transistors on the web somewhere?
Last edited:
Yes!
I am considering the toroidal transformer -> CRC solution (would 12V ac transformer be ok? ac 12V*1.41- 1.4 V(rectifier) =15.4 V before CRC, after CRC voltage drop ok at 12 V?).
But then again some post here from Munich say that a regulated power supply is better, has anyone heard both power supplies?
For the moment the switched mode power supply is sounding already good (just a bit noisy).
About using transistors:
Small signal BJT: On Semi BC550C+BC560C, Toshiba 2sc2240+2sa970
JFET: 2sk246Y+2sj103Y, 2P303A+2P103A,B (gold) (made in USSR for military).
Drivers BJT: in the theory: Sanyo 2sc2910+2sa1208, On Semi BC550C+BC560C
Output BJT: Kec 2sd718+2sb688 (low cost), ON Semi MJL21194+93, MJL21196+95, MJL3281+MJL1302 (a little danger, because 30 MHz), MJ15024+25.
Buffer operating mode of the accumulator 12V is 13,5...13,8V (as a rule).
I don't like integrated stabilizers (lm317+337, lm338, lt1083 and other), but maybe the sound with them will be better for you.
Not true at all. I went from a Blue Velvet Alps pot to wire wound pots. The difference in detail was significant. A plastic film is not an ideal resistor and is in no way comparable to a single fixed resistor.
Shoog
I read on the board "max 10.5Volt", is that because of the super capa's?
--> I run my 0,1F 10 V super capa's at 13,4 volt (pike Jean Hiraga did too) , and as long as it is stabilized (no ripple) this is OK.
albert
It's old first test board.
I want to make PCB with own noise level: -140 db on 50 Hz and 100 Hz.
No I decided not to go with the original ridiculous power supply. i have implemented a cap - capacitance multiplier - cap power supply. The main amp section hasn't been built yet but it will be on perf board.
Shoog
Aa. I want to rebuild power supply from 0,8F to 1,2F (36x33000 uF) and to make original pcb (for the sake of interest) with buffer's ACB 12V 4,5A*h .
Any Class A amplifier needs a substantial power supply. Cutting corners will only lead to disappointment.
It is a substantial power supply I have built and I am quite experienced in valve Class A designs knowing full well that they need lots of filtering.
Still I am not fanatical about having huge reserves of capacitance which represents the brute force approach. A capacitance multiplier offers a finessed approach to the same end.
But mine is a technical question which deserves a technical answer. Lets repeat it - is there any PSRR in the amp or will the full 15mV make it straight to the output ?
Has anyone measured their residual power supply ripple and can report its impact on the output.
Shoog
Member
Joined 2009
Paid Member
I believe the cap multiplier has been used by the 'Greats' around here. But I do remember making note that a) Hugh Dean found they didn't sound 'good' to him b) and that Nelson Pass said you need plenty of capacitance AFTER the cap multiplier.
I think the issue is keeping the output of the power supply at a low impedance over the audio range, having return currents flow through rail caps instead of modulating the emitter of the cap multiplier device. Some additional local feedback (in addition to it being an emitter follower) would help - I've looked at the CFP as the obvious candidate, but there are others.
I measured supply ripple on my SET amplifier and it did not have much impact on the hum at the output because the hum was magnetically coupled into the OPT !
I think the issue is keeping the output of the power supply at a low impedance over the audio range, having return currents flow through rail caps instead of modulating the emitter of the cap multiplier device. Some additional local feedback (in addition to it being an emitter follower) would help - I've looked at the CFP as the obvious candidate, but there are others.
I measured supply ripple on my SET amplifier and it did not have much impact on the hum at the output because the hum was magnetically coupled into the OPT !
My first amp was a ZENv3 and I had to resort to a cap multiplier. It did have an impact on the sound - which I wasn't that keen on either.
However I had already read Nelson Passes advice and have built the power supply with the bigger cap after the capacitance Multiplier.
Tthe configuration is 5,000uf>cap Multiplier>64,000uF. The ripple at the 500ouf is about 0.6V and at the 64000uF its down to 15mV.
64000uF at the output should dominate the response all the way down to a few 10's of hertz so I am anticipating that the cap multipler will largely be out of the picture sonics wise. The power transformer is something like 240VA so I do not anticipate it sagging significantly even under sever peaks.
Shoog