P3A amp 100 watts Help

[j a y k e z]

updated layout

-added base stopper for drivers and output transistors to avoid oscillation
-added signal ground lifter
-updated BOM

use suggested components in BOM to get optimum musical performance 

+1 sa effort sis

[jlester87]

No prob sis

sure na stable na yan, lower thd rin dahil sa ground lifter... 

Important yung component choice sa vas at drivers, pati yung C5 cap malaki daw epekto sa sound...

yung mga overlap na component sa layout sa ilalim dapat ng pcb nakalagay...

[abeTIR]

ahehe..nag-etch na pamandin ako neto.. Me update pa pala..
Pa-send rin uli niyang latest pdf files mo sis.

TY!

[jlester87]

aw! ok lang yan kung hindi nag ooscillate (manipis ang tunog ng p3a pag may problem)

nilagyan mo ba ng ground lifter?

[abeTIR]

Eto sis yung revision na ginawa ko, pacencia na nawala yung name mo kase merong Zobel ito..may mga piyesa na itong naikabit sa board needs completion na lang..upon seeing some P3A boards medio nagbago yung isip ko dun sa dating lay-out.

[jlester87]

nice! oks lang kahit wala name ko haha..

pwede mong kaskasin yung link sa power ground - signal ground para malagyan ng ground lifter...

saka mas ok kung iba ng tranny sa vas, saka yung ltp try mo gamitan ng bc546 as suggested by sakis

update mo kami sa sound

[fantom_mayonaise]

Laser toner ba yan sir Abe? anlinis ng pagkakatuklap.

[abeTIR]

^oo sis, toner transfer yan.

To preserve yung dating orientation ng diff input (pang 2N5551-EBC) MPSA18 ang gagamitin ko dito almost same specs din naman ng High gain-more linear BCs trannie. BD139/140 muna ang the rest. Then test ko muna..

[jlester87]

kahit yung vas lang sir yung palitan mo... pangit daw kasi ang bd's as vas

at sa c5, get the best capacitor... pwede na yung panasonic fc series...

[gorex]

Plus 1 jlester87 sa effort mo at sa tips. nabasa ko din sa diyaudio yung mga post mo. ok na ok pcb design.

[jlester87]

Update: added optional C-E cap on VBE multiplier
http://www.4shared.com/archive/WVklWkHx/P3A-PDF.html

[Johannah Aleksandria]

Na test mo na ba effect nitong cap sis?

[jlester87]

hindi pa sir, pero allegedly mas magiging ok ang bass, still googling for the theory behind this

[JojoD818]

hindi pa sir, pero allegedly mas magiging ok ang bass, still googling for the theory behind this

the bass will not improve but your biasing will in terms of stability but at the price of response speed.

[tony]

hindi pa sir, pero allegedly mas magiging ok ang bass, still googling for the theory behind this

a capacitor has a frequency response inversely proportional to frequency, so that at very high frequencies it is likened to a short.....therefore used to shunt the Vbe multiplier, transient response is enhanced since at AC the collector emitter junction of the Vbe trannie appears to be a short....values of 0.1ufd are typical although higher values can be used....

[Johannah Aleksandria]

hindi pa sir, pero allegedly mas magiging ok ang bass, still googling for the theory behind this

It would better the ac operation of the amp.
Capacitor is a short for AC signals, while the Vbe multiplier would not completely be a short for AC signals, although there have been no comment on this. If there is a capacitor, ac signals are intact while the Vbe multiplier is effective only on the dc biasing of the drivers and output.
Vbe multiplier will be ignored for ac signal, the capacitor handles it.
The capacitor have no negative effect on the dc operation of the Vbe multiplier, even then stabilizes it.

[abeTIR]

I'm done...



For 2 hours play (at -/+24vdc), quite stable naman malinis at walang sign ng oscillation problems.



My final patchwork dun sa orig lay-out ni sis Lester...yung power ground traces kinapalan ko lang ng konti using pentel prior to etching...

Zobel is wired to power ground ng PCB.
I used MJE15032/15033 for drivers, B649A for VAS (Ic 1.5A, Pc 1W, ft 140mhz, Hfe 200max, Cob 14pf) Input cap Bennic film 9uf. Differential trannies MPSA42. Dale 0.22 5W precision resistor.

Mukhang a little overkill for the job ang ginamit kong mga piyesa.. Anyway yan kase ang available na meron na ako. Later-on baka palitan ko rin uli yung akma sa power rail around -/+42vdc. At power-on gumana kagad all components ay normal naman walang overheating na nadetect. A little over the edge ata yung DC offset at 37mv (pero super quiet naman), quiescent current at around 70ma.

My comments...Great dynamics with the presence of ambiance. Forget the harmonics it has to measured at hindi naman ako audiophile with a golden ear... Anyway adjective praises to sonic quality is subjective to one's listener ears.

Thumbs-up! to sis Lester for the lay-out...plus uli sayo...

[TinTopHack]

hindi pa sir, pero allegedly mas magiging ok ang bass, still googling for the theory behind this

Cap across Vbe multiplier

- The purpose of the Vbe multiplier is to supply a constant voltage (with temp compensation) across the bases of the upper and lower pre-driver transistors. Under normal conditions (without threat of thermal runaway) this voltage must be smooth and constant.
- You can imagine an ideal Vbe multiplier as a battery whose voltage is riding on the audio signal at the output of the VAS. Other way of looking at it is - a battery that slides up and down depending on the output of the VAS.
- Any sudden variations in the voltage of the Vbe multiplier (if not smooth or constant) will introduce distortion into the audio signal. Adjustments to the voltage of the Vbe multiplier as a result of temp compensation will not create noticeable distortion because such adjustments are NOT sudden; instead the adjustments occur gradually overtime depending on temp variations.

Why add a cap across it?

- Main reason to smooth the voltage across the Vbe multiplier - making the bias to the pre-driver transistors very stable. Thus the typical value is 47uF.
- The cap is usually added in circuits that use a constant current source as load to the VAS. This because at  signal amplitudes approaching the voltage compliance of the current source, the voltage across the Vbe multiplier could drop - resulting in a reduction in base drive to the pre-driver transistors - resulting in pre-mature clipping - thus distortion. By adding a cap across the Vbe multiplier, the cap will keep the voltage across the Vbe multiplier reasonably constant even when the voltage compliance of the current source is breached. In other words, the cap across the Vbe multiplier ALSO acts as a bootstrap capacitor.
- Rod Elliot's P3A, does not use a constant current source as load to the VAS. Instead it uses a bootstrap cap circuit to act as load (which presents a constant current load to the VAS). This bootstrap cap serves a similar purpose as a cap across the Vbe multiplier but doing it in a slightly different way.

*** You could add a cap across the Vbe multiplier of P3A but it may not add much in terms of performance.


 

[Johannah Aleksandria]

I'm done...
For 2 hours play (at -/+24vdc), quite stable naman malinis at walang sign ng oscillation problems.
My final patchwork dun sa orig lay-out ni sis Lester...yung power ground traces kinapalan ko lang ng konti using pentel prior to etching...

Zobel is wired to power ground ng PCB.
I used MJE15032/15033 for drivers, B649A for VAS (Ic 1.5A, Pc 1W, ft 140mhz, Hfe 200max, Cob 14pf) Input cap Bennic film 9uf. Differential trannies MPSA42. Dale 0.22 5W precision resistor.

Mukhang a little overkill for the job ang ginamit kong mga piyesa.. Anyway yan kase ang available na meron na ako. Later-on baka palitan ko rin uli yung akma sa power rail around -/+42vdc. At power-on gumana kagad all components ay normal naman walang overheating na nadetect. A little over the edge ata yung DC offset at 37mv (pero super quiet naman), quiescent current at around 70ma.

My comments...Great dynamics with the presence of ambiance. Forget the harmonics it has to measured at hindi naman ako audiophile with a golden ear... Anyway adjective praises to sonic quality is subjective to one's listener ears.

Thumbs-up! to sis Lester for the lay-out...plus uli sayo...

Thanks sa update sis, overkill? tama lang yan sis.
We assure ourselves of best performance, so we choose the ones we know will meet to expectation.
At least wala ka ng sisisihin pa., (mumong parts) 
Mganda ang pagkagawa. Trying to give the most presentable also(me).

[jlester87]

I'm done...

For 2 hours play (at -/+24vdc), quite stable naman malinis at walang sign ng oscillation problems.

My final patchwork dun sa orig lay-out ni sis Lester...yung power ground traces kinapalan ko lang ng konti using pentel prior to etching...

Zobel is wired to power ground ng PCB.
I used MJE15032/15033 for drivers, B649A for VAS (Ic 1.5A, Pc 1W, ft 140mhz, Hfe 200max, Cob 14pf) Input cap Bennic film 9uf. Differential trannies MPSA42. Dale 0.22 5W precision resistor.

Mukhang a little overkill for the job ang ginamit kong mga piyesa.. Anyway yan kase ang available na meron na ako. Later-on baka palitan ko rin uli yung akma sa power rail around -/+42vdc. At power-on gumana kagad all components ay normal naman walang overheating na nadetect. A little over the edge ata yung DC offset at 37mv (pero super quiet naman), quiescent current at around 70ma.

My comments...Great dynamics with the presence of ambiance. Forget the harmonics it has to measured at hindi naman ako audiophile with a golden ear... Anyway adjective praises to sonic quality is subjective to one's listener ears.

Thumbs-up! to sis Lester for the lay-out...plus uli sayo...

Maganda ba yung vocals? yan kasi yung una kong napansin...

Tinignan ko yung p3a ko sa oscilloscope and it's picking up high frequency signals, probably dahil sa input wiring ko...

@TinTopHack
Thanks sa info sir, so ok lang pala kahit walang cap...

Anyway optional lang din siya sa pcb layout, pati yung base stoppers pwede lagyan ng wirelink kung stable naman...