SemiSouth power Jfets .......... FW J2

[Tea-Bag]

No worries. Q13 - 1.31 UgS @ 2A, Q14 1.42 UgS @ 2A

[Zen Mod]

OKtry to preset - in cold ( prior to powering up ) sum of R12+R35 to ~250Rthen try standard procedure with variac

[Tea-Bag]

OKtry to preset - in cold ( prior to powering up ) sum of R12+R35 to ~250Rthen try standard procedure with variac

250R, 310R, did not work. I remember after replacing Q14, but not fixing short around R27, that it biased up to .750, but after fixing it, I got this. I had de-soldered most of the upper right part of the schematic trying to figure that out, maybe something toasted there, but uncertain. Drilling that area, really just takes the R27 to C3 track out of the picture.Everything around 'front end Jfet's' look good.

[Zen Mod]

have you continuity between R32 (not jfet side but left side on schematic) and C of Q2also - check continuity between R32 and Q14

[generg]

hi Zen Mod, only for pleasure I simulated with multisim the two buffers...... at 5 v and 10kH your buffer had only a tenth of the distortion of the B1 Buffer (0,001/0,01) next week I will build up.......hope you had a nice weekend

[generg]

Oh I forgot,matching the two j-fets?at about 8mA Idss?

[Tea-Bag]

have you continuity between R32 (not jfet side but left side on schematic) and C of Q2also - check continuity between R32 and Q14

Yes, those are fine on both channels.

[Zen Mod]

:dubbio:point is that upper jfet is open but lower one isn't , so you are having same offset as in case that you have high value resistor instead lower jfet , while upper one is trying topump current through it .I can't figure how on earth - what's stealing current from R35&R12 combo ; is Q4 OK ?with positive offset on output - it will be logical that Q11&Q12 are sort of choked and that Q9&Q10 are conducting more current , so you should have greater amount of voltage across R35&R12 ........ but not - here is contrarycheck Q4 please

[Tea-Bag]

:dubbio:point is that upper jfet is open but lower one isn't , so you are having same offset as in case that you have high value resistor instead lower jfet , while upper one is trying topump current through it .I can't figure how on earth - what's stealing current from R35&R12 combo ; is Q4 OK ?with positive offset on output - it will be logical that Q11&Q12 are sort of choked and that Q9&Q10 are conducting more current , so you should have greater amount of voltage across R35&R12 ........ but not - here is contrarycheck Q4 please

Replaced Q4 - still same appoxiamate voltage .765Measured again Q6 - - ~ -12v like last time.Good channel Q4 - ~2V, Q6 around - 6V (I think). Not as high. No BF245's around to swap out.Thanks for looking.

[Zen Mod]

Replaced Q4 - still same appoxiamate voltage .765Measured again Q6 - - ~ -12v like last time.Good channel Q4 - ~2V, Q6 around - 6V (I think). Not as high. No BF245's around to swap out......

what's important regarding Q6 is just voltage at bases of cascodes - Q2 & Q3 ;they must be in range of minus 11-12V - exact what is voltage drop across leds+1N string

[Tea-Bag]

- exact what is voltage drop across leds+1N string

Is this a question? I think your just telling me role of Q6. Not related to issue...

[Zen Mod]

that was just pointing that I don't care and it's not imprtant what you are measuring at Q6 - if you have -11V to -12V at bases of Q2 & Q3 ; in that case LED string and Q6 CCS are serving their functionif channel is working - then they are functional , so you don't need to check them there - voltage check points at schmtc are good enough for purpose

[Tea-Bag]

swapped out Q13 and Q14 for a better matching pair, and still no bias. Hmm.I think my next step is another analysis of continuity and cold resistance figures on both channels. If that provides nothing. I will overhaul the motorcycle.

[sp300b]

Small package Babel J2 with most important Parts (PCB s, Semi..., CNY17), finally arrived yesterday.Hope I'll find some spare time and completed it (at least) during year 2011. p.s. Zen

[Zen Mod]

........p.s. Zen

our love and respect certainly isn't supported by frequency of our meetings:air_kiss:

[sp300b]

our love and respect certainly isn't supported by frequency of our meetings:air_kiss:

It's more by amplitude

[Zen Mod]

[sp300b]

Something is wrong...couldn't be...must be... couldn't be...let us check

BabelJ2_Part_List_www.zip

[sp300b]

Something is wrong...couldn't be...must be... couldn't be...let us check

and compare

[sp300b]

and leave the puzzle solution here

BabelJ2_Part_List_www.pdf

[Zen Mod]

if you look at parts nomenclature on your set of pcbs - that's Tibor's mistake , during multiplication of pcb pattern to accommodate his production frame/formatas said in e-mail - take care of latest addition - protecting zenners on output gatesadd them on copper side , or as suits you

[generg]

Hi to all builders of Babelfish J2 :Viannen_loungelizard:

the outputstage Babelfish is so good that I have a real problem with my pre´s!

best choice up to now was always Pumpkin, but this time combined with J2 and 22 Ohm gate stoppers, the instruments image got a little bit to small for my taste, going to 100 Ohm made the size correct.

Aleph P had no size problem, but had not the resolution of Pumpkin.

Buffer B1 is not bad, but hearing Gregorian chants you can hear voices got more unclear and less detailled.

Best solution up to now, but commonly not recommended, no pre at all, only shunt attenuator 10k with Charcroft foils in the signal path and short cables to J2.................. simply marvelous!

Most natural sound! May be bass is not totally precise, but all colors of sound are much better, which is more important to me.

I suppose the high input impedance of J2 (220k balanced) and the high gain (20db) and as Zen Mod wrote, the very small input capacitance of J2 makes this direct solution possible.

I recommend to try this possibility also ............... when you have your building of J2 finished. :headphone: