give me schematic of everything in front of Superreg - R , L, C values , and I'll calculate for you secondary voltage .
I know Superreg schematic from yore times of VS site , where parts of Superreg were published independently , so I just solved the puzzle ; later I bought books from Allen ;
but - what's most important - I made few of my own regs , including Shunty , which operate on same basic principles :
-series Mosfet demands just few volts more added to basic Ugs voltage ; LM317 needs few volts plus voltage sag across programming resistor
- shunt part of reg needs just programmed voltage across it , to be happy
all that - as bare minimum of in-out differential voltage , in moment when mains voltage is lowered for 10%
that's approx. just 50% of 15V I mention in my last post ;
Allen claimed for 50V as value which will cover even worst case scenario , saving him endless correspondence with ambitious , but not exactly up to task amateurs
just an observation - using CCS as in Shunty is better solution than using cascoded crummy LM317 ; I'm not saying that you'll hear diference , but from engineering point of view .....
I know Superreg schematic from yore times of VS site , where parts of Superreg were published independently , so I just solved the puzzle ; later I bought books from Allen ;
but - what's most important - I made few of my own regs , including Shunty , which operate on same basic principles :
-series Mosfet demands just few volts more added to basic Ugs voltage ; LM317 needs few volts plus voltage sag across programming resistor
- shunt part of reg needs just programmed voltage across it , to be happy
all that - as bare minimum of in-out differential voltage , in moment when mains voltage is lowered for 10%
that's approx. just 50% of 15V I mention in my last post ;
Allen claimed for 50V as value which will cover even worst case scenario , saving him endless correspondence with ambitious , but not exactly up to task amateurs
just an observation - using CCS as in Shunty is better solution than using cascoded crummy LM317 ; I'm not saying that you'll hear diference , but from engineering point of view .....
ok
if your goal is to have 300Vdc after Superreg , as needed for RTP3D .....
you can decrease programmed current in Superreg to some 100mA
load is approx. 19+14+15 mA (per channel) and going to anything more than twice that for programmed current is just waste
correct that , simulate again and correct needed values of resistors in RC chain
edit: with 420Vac and 100mA as load , you're spot on
if your goal is to have 300Vdc after Superreg , as needed for RTP3D .....
you can decrease programmed current in Superreg to some 100mA
load is approx. 19+14+15 mA (per channel) and going to anything more than twice that for programmed current is just waste
correct that , simulate again and correct needed values of resistors in RC chain
edit: with 420Vac and 100mA as load , you're spot on
Attachments
Last edited:
yes
no side effects, you only raise the voltage of the cathode and something 7 or 8 or 9 doesn´t matter
and my superreg runs with 330V Vin since 10 years
no side effects, you only raise the voltage of the cathode and something 7 or 8 or 9 doesn´t matter
and my superreg runs with 330V Vin since 10 years
there is no way that Allen was using either 6DJ8 or Jfet at more than 10-12mA ... meaning that SLCF is drawing 20-24mA
when I wrote 15 , that was a typo , sorry .
find enclosed two sims - one for dual mono PSU ( one PSU and SR per channel ) , second one for common PSU for both channels (one PSU and SR for both channels)
of course that first version is preferred
when I wrote 15 , that was a typo , sorry .
find enclosed two sims - one for dual mono PSU ( one PSU and SR per channel ) , second one for common PSU for both channels (one PSU and SR for both channels)
of course that first version is preferred
Attachments
yeah
diminishing returns ;
but certainly less headache , regarding possible ground loops - when you completely control where grounds of L & R channel will meet
With separate shunt regulators, the HV+ are separated between the channels. On can add a capacitance multipliers, dropping say 10V, to the HV-, one for each channel. Thus the HV supply will be completely isolated between the 2 channels.
yup - positive and negative legs are certainly more than isolated through separate shunt regs (for positive ) and plain voltage regs (for negative ,even common for both channels ,due to isolation nature of CCS-es in cathodes) ;
but what I meant is strictly controlled galvanic separation of signal gnd between channels.
anyway , I'm speaking of that not as about sole possible way of doing it ; just trying to explain sort of purist approach , which Allen certainly had , when making of RTP3D is in question .
in any case - DIY-ing that sort of preamp is supposed to be a fun , when one if err - err on exaggerating side , not attempt in making something extraordinary for pocket money.
but what I meant is strictly controlled galvanic separation of signal gnd between channels.
anyway , I'm speaking of that not as about sole possible way of doing it ; just trying to explain sort of purist approach , which Allen certainly had , when making of RTP3D is in question .
in any case - DIY-ing that sort of preamp is supposed to be a fun , when one if err - err on exaggerating side , not attempt in making something extraordinary for pocket money.
As far as I can see, my solution enables complete galvanic isolation between the 2 channels, with very little added cost (over 1 PSU and 2 shunt regulators).
If that suits you , OK by me .
Can you see any possible benefit by having dual PSU?