JG´s Nobrainer and Nobrainer Discrete

[FdW]

The zeners and the diode form the voltage reference.... the setup aims for lowish impedance and temp compensation.

The transistor and the cap before the reg form a cap multiplier.... it works, I have used this for +- 15v in my cdp..... powering the output opamps....

But it is no cap multiplier and I'm sure it will work as a 15V supply, the BC's change nothing about that!

 

[FdW]

You mean on the input of the reg ?

For the CRC filter? Yes

 

[FdW]

The zeners and the diode form the voltage reference.... the setup aims for lowish impedance and temp compensation.

The parallel zeners do not improve anything (they would if the matching was 100% (0% difference) but they are not and even then ... mmm (doubts expressed )) Anyway long term thermal stability is not a problem (over periods of minutes even seconds). I would guess that modern day 317/337's are even more stable then the zener(s)/diode.

 

[FdW]

Found some D44 D45 power trannies in my stock... good... now I have an old Meridian 101 PSU that I would like to use in this project.... the output is +-35v so it is too much for the opamp.... can I use LM317 337 to feed the opamp at 18volts ?

When powering down, the capacitor is emptied through the BC's base-emitter diode, that is not advisable. To be a capacity multiplier you need some voltage drop between the BC's collector and base. On sudden large current demands, the output power is supplied through the BC's base-emitter diode (not good).

 

[RCruz]

Ok, this was an old idea that I used 10 years ago.... you know better than me of course... I will use the standard 317 layout

 

[RCruz]

How much current can the opamp draw from the psu ?

I am really considering using the kmult to filter but it needs enough current to work best.

 

[Joachim Gerhard]

What opamp ?
Something like LME49710 draws around 8mA.

 

[Joachim Gerhard]

You can burn some current in a parallel resistor at the output of the cap-multiplier.

 

[RCruz]

Yes, otherwise the kmult will struggle, but now it seems to add too much complexity.

I will try the "normal" 317 reg first.

 

[Joachim Gerhard]

OK, good. The new National opamps have very good PSU rejection anyway.
I use a little trick i learned from Bonsai.
Use 22 Ohm resistors in series with PSU plus and minus and decouple with 220uF each.
I then use only one 0.1uF foil from plus to minus for HF decouple ( Self ).
That way you improve PSU rejection in the treble where PSSR of the opamp gets worse.

 

[RCruz]

I then use only one 0.1uF foil from plus to minus for HF decouple ( Self ).
That way you improve PSU rejection in the treble where PSSR of the opamp gets worse.

Where do these 0.1uF go ?

plus to minus near the opamp Voltage inputs ?

 

[FdW]

Where do these 0.1uF go ?

plus to minus near the opamp Voltage inputs ?

as used in my power amp

 

[Joachim Gerhard]

Thanks Frans.

 

[RCruz]

So I got it right

This will be a blast

 

[FdW]

So I got it right

This will be a blast

You did not got it wrong say no more...
Monty Python - Nudge Nudge Wink Wink Say No More - YouTube

 

[RCruz]

Lovely sense of humor Frans

 

[RCruz]

Here we have my first p2p draft.

Please comment on the component's layout and particularly on their pinout.

PS: Should the drivers be placed in contact with the heatsink ?

 

[Joachim Gerhard]

I have the drivers close to the heat sink but not in direct contact.

 

[RCruz]

Would you care to verify pinouts of the opamp and output trannies... this is the copper side view btw.... I would really appreciate that and maybe I will save a lot of trouble

 

[RCruz]

Now with the regs included.... not sure about zener and leds orientation on the negative side (and it is a 337)