Let's discuss ZM's Good Gemini shunt

[dimkasta]

The shunt was shown as a brother of Iron pre and Iron pumpkin and was partnered with a capacitance multiplier.

http://www.diyaudio.com/forums/pass-labs/293169-whats-wrong-kiss-boy-19.html#post4883680

I would like some help discussing its Vout range and load current limits to see if it could fit any other projects like the B1, the BA-3 or even digital stuff.

 

[Ketje]

Output max 45mA adjustable between 14.2V and 18.5V , negative side same thing.
Input min 4volts extra.
Somewhat over the hill for my taste.
Mona

 

[Zen Mod]

dimkasta , say your needs (V, A) and you'll get it

 

[dimkasta]

Let's say for BA-3 for starters.

So it would be nice to have 25-30V range at 150-200mA to allow for some hotrodding and output biasing experimentation.

 

[Zen Mod]

confirm:

-Ba-3 , not balanced

-one Good Gemini per channel

- +/-24Vdc , ~55mA , V/A per channel consumption

 

[Ketje]

Let's say for BA-3 for starters.

So it would be nice to have 25-30V range at 150-200mA to allow for some hotrodding and output biasing experimentation.

Not very much to change but the BD's and IRF's must be better cooled.
Mona

 

[Zen Mod]

cooling - conditio sine qua non

200mA is way too much for 55mA A class stage load

matter of principle - 120mA is more than enough, when having (active) error amplifier

will post schm later tonight

 

[Ketje]

Yes I now, 200mA is a lot.But the TS asked for it.But now he knows what to change.
With 55mA the BD139/140 can stay.
Mona

 

[dimkasta]

To quote NP

Of course you can bias this circuit higher if you wish – 100 mA bias is perfectly OK as long as you properly heat sink Q3 and Q4, and if you are crazy (like me) you can experiment with higher bias, remembering that the parts are rated at 25 watts

Now where is my crazy hat....



Anyway, I did some reading.

If I understand this, R3||R4 sets the constant current output using

I = ~(Vbe on)/Rtot

right?

And the divider R7/R8/R9 sets the voltage?

 

[Zen Mod]

here it is - for BA3 FE , +/-24Vdc , programmed current 120mA

yup , you got it , for programming current and for setting output voltage

however - when changing output voltage , one must maintain optimal currents both through reference diode and through voltage divider for error amp (R7,R8,R9)

reg itself is simple ...... pretty much everything is covered in & Shunty CookBook , simply because GG is smaller bro of Shunty , but not less effective

consider one important thing , pretty much covering all active electronic stages , but especially supply regs - PCB design is of same importance as circuit itself

BD139/140 are good up to 5W , if cooled properly
above that , go to BD239/240

xformer - everything bellow 56Vct (2 x 28Vac) is on your own risk, considering regular mains voltage sagging of 10% or even more

15VA per channel , 30VA for both , as minimum

first elco caps after bridge can go up , rest of them no need or counterproductive

if you want 200mA for programmed current , just change those resistors to
5R1 // 9R1

everything else as is

diss on BD :

(input V - out V) x programmed I

diss on IRF :

out V x (programmed current - load current)

 

[Zen Mod]

and yes , I'll make more potent (current wise) pcbs , relatively soon , with CCS-es instead of R1,R2 and R24,R25

then you can use them in pumpupthevolume manner , as much you want ......

but ........that right after finishing special iron for Iron

 

[dimkasta]

Should I worry that I got the musical reference?
Am I getting... old?

https://www.youtube.com/watch?v=w9gOQgfPW4Y

 

[dimkasta]

consider one important thing , pretty much covering all active electronic stages , but especially supply regs - PCB design is of same importance as circuit itself

Yeah I noticed you love your return currents

 

[Zen Mod]

Should I worry that I got the musical reference?
Am I getting... old?

https://www.youtube.com/watch?v=w9gOQgfPW4Y

yup

and right after that , you can listen to https://youtu.be/PbbYn7d_W8s

 

[Zen Mod]

aha , forgot to say

you mentioned digital , as possible area of use

nope ....... due to construction , even with possibility of major re-work , I wouldn't go bellow 10V for output voltage

simply - matter of choice ...... for 5V and similar , we need different approach

 

[Ketje]

BD139/140 are good up to 5W , if cooled properly
above that , go to BD239/240

diss on BD :

(input V - out V) x programmed I

diss on IRF :

out V x (programmed current - load current)

Changing BD139/140 to BD239/240 is not a good idea.Then you have to get smaller R1,R2 (need more base current) too, making poor regulation.
That's why I put darlingtons for more current, the resistors can stay as is.
Mona

 

[Zen Mod]

you're right , I forgot to mention change in value of these resistors

however , 5W is more than enough with BD139/140

above that , I would consider something in range of BDX53/54 family

in any case , this is not exactly swiss army knife , every major change (meaning Vout/I out) demands necessary analysis , to maintain proper phase behavior of reg

 

[schultzsch]

What is better between the Gemini and a Salas V1.3?

 

[Zen Mod]

for Salas his, for me mine

 

[schultzsch]

And for me?