B1 Buffer Preamp

[Salas]

Originally posted by Zen Mod
in 10 min. I'll post re-drawed circ for you

edit :

little more than 10 min

here it is : you can mount it in same case with chip/drek amp , using any type of IC or discrete voltage reg , to have nice and stable +/-9V , from same PSU ( of chipamp ) ;

dunno - probably LM317/LM337 will do the job for you , or nice discrete zenner followers

Hello Zen Mod

I prefer the symmetric supply because it loses the input capacitor.
I guess that if the NJFETs are chosen very close in IDSS we may lose even the output capacitor, yes?

At least I don't see any DC if simulating with theoretical devices (totally the same). Can't we put a null trimmer in real life?

[Zen Mod]

Originally posted by salas



Hello Zen Mod

I prefer the symmetric supply because it loses the input capacitor.
I guess that if the NJFETs are chosen very close in IDSS we may lose even the output capacitor, yes?

At least I don't see any DC if simulating with theoretical devices (totally the same). Can't we put a null trimmer in real life?

I can't see why we can't .........

feel free to put trimmer , but I'll not loose my sleep with proper 10uF in output ......... as Papa probably would say .

without servo - these critters are wandering around all the time ........

it's your choice what you'll use

[Zen Mod]

Originally posted by BURNER
Yes Choky, offset with 10uF

However, offset drops to 1-2 mV after few minutes.
supposed to be problem during charging period of MKP 10uF but I think it takes too long in my case

I'll try with another cap type though...

Originally posted by stefanobilliani



I have seen something like that too in my F5. Being directly coupled the F5 is pretty sensitive to little dc voltages at its input .
The cure for that is to connect the B1 to the F5 when the power of both is off . Then Power on the B1, wait a minute , and power on the F5 .

smaller resistor from output side of cap to ground - will speed up charging ; decrease it from present 221K value to , say , 100K 0r even less

[stefanobilliani]

Originally posted by Zen Mod





smaller resistor from output side of cap to ground - will speed up charging ; decrease it from present 221K value to , say , 100K 0r even less

That is R105 and R205 of the B1's circuit .

[AndrewT]

Originally posted by stefanobilliani



That is R105 and R205 of the B1's circuit .

no, the output resistor dissipates stored charge when the amp is dormant.
A similar resistor for the input cap will bring both input and output of the respective units to DC zero volts and allows connection between units without a transfer of charge between the caps.

At switch on the 100k, or 1M0 if you prefer, has almost no effect on the way the caps charge up and the way they pass signals once partially/fully charged.

[BURNER]

Thanks Choky and Stefano, I'll try to reduce R105 - 205 and see how it will go.

[float]

Originally posted by Zen Mod



without servo - these critters are wandering around all the time ........

Hi Zen, could you elaborate on what you mean by this?
I'd be very interested in getting rid of both input and output caps if possible, as they are probably the biggest influence on the transparency of this circuit.
Do you mean dc offset would still be a problem with a dual rail supply?

Also wondering why Nelson built this as a single rail supply, as the dual rail alternative seems to need fewer parts?
No disrespect to the maestro intended.

[Diomedian]

Originally posted by float

Also wondering why Nelson built this as a single rail supply, as the dual rail alternative seems to need fewer parts?
No disrespect to the maestro intended.

Just speculation here, but it seems to me that a dual rail supply only uses fewer parts if you happen to have +/- 9V DC (or so) already. A single rail allows you the convenience of using any old wall wart. Further, I'd imagine that the big man would encourage you to modify to your heart's content. :-)

-d

[MRupp]

Have not built this buffer yet (hopefully soon) but I am intending to use a bipolar PS and no caps. I do have 2SK389 but from what I understand so far you just might get away with 2SK170s, preferably matched for Idss as closely as you can and off course thermally coupled (e.g. glued together). If you have an offset after warmup you might try and tweak the + or - voltage ever so slightly.

[Zen Mod]

Originally posted by float



Hi Zen, could you elaborate on what you mean by this?
I'd be very interested in getting rid of both input and output caps if possible, as they are probably the biggest influence on the transparency of this circuit.
Do you mean dc offset would still be a problem with a dual rail supply?

Also wondering why Nelson built this as a single rail supply, as the dual rail alternative seems to need fewer parts?
No disrespect to the maestro intended.

simple jfet circuits have tendency of changing current in time domain ,strictly dependable of jfet temperature change ;

if you can secure steady temperature conditions for them ( closed case , or case with controlled amount of air circulation) you'll have so-so stable offset on output , but not in time domain .

point is that we don't have DC feedback path , which can secure offset stability

Originally posted by Diomedian


Just speculation here, but it seems to me that a dual rail supply only uses fewer parts if you happen to have +/- 9V DC (or so) already. A single rail allows you the convenience of using any old wall wart. Further, I'd imagine that the big man would encourage you to modify to your heart's content. :-)

-d

agree with all that ;

as said numerous time ( at least from humble me ) Papa's main intention and probably satisfaction is to see that some of his spoiled (veeeeeeeeerrrry spoiled ) kids use gray cells sometimes .

in same time , you can learn (and see - in first place ) few things from his gadgets - humor and simplicity . nothing bigger than life itself .... as in some other constructions circulating around

hehe - I can't find smiley rotten enough for that last sentence