Step 1, used another tranny (18-0-18):
I have now 49.5VDC on C105
But cannot get more than 39.5VDC output
I change R103 to a 10k type, right?
Franz
I have now 49.5VDC on C105
But cannot get more than 39.5VDC output
I change R103 to a 10k type, right?
Franz
Its the Vref then, no doubt. With now Q103 IDSS you chanced on, is a bit short of steam. If you will up the R103 value it will increase. What voltage drop you got over R103 and what is its now value?
2.97mA. Means it sits at the left of midpoint 4mA that the examples are given for. Make R103 10K. Should suffice. Will give ~9V more range.
First quick and dirty test (dont do the same, it is a dangerous arrangement):
No remote sensing yet.
Recently, I purchased the latest Tom Waits album on vinyl (Bad As Me). And it contained also the CD.
Now, I am switching for and back between vinyl and my squeezebox, as I did it some weeks ago, using a LM317 psu for the phono pre.
You may ask yourself: is Tom Waits a good choice to compare?
Yes, it is. It contains very clean and sublte details.
First minutes, I was thinking: not a big difference to the LM317 psu.
But, now listening on the second side and switching for and back:
This is it. A big step towards real analog audio for me.
The vinyl reproduction is beating the digital reproduction in some sublte details now.
I simply like it.
Thanks for all Salas!
I wish a merry christmas to all readers
Franz
An externally hosted image should be here but it was not working when we last tested it.
No remote sensing yet.
Recently, I purchased the latest Tom Waits album on vinyl (Bad As Me). And it contained also the CD.
Now, I am switching for and back between vinyl and my squeezebox, as I did it some weeks ago, using a LM317 psu for the phono pre.
You may ask yourself: is Tom Waits a good choice to compare?
Yes, it is. It contains very clean and sublte details.
First minutes, I was thinking: not a big difference to the LM317 psu.
But, now listening on the second side and switching for and back:
This is it. A big step towards real analog audio for me.
The vinyl reproduction is beating the digital reproduction in some sublte details now.
I simply like it.
Thanks for all Salas!
I wish a merry christmas to all readers
Franz
A Salas Shunt requires at least 5V differential and preferably >7V from input DC voltage to output DC voltage.
If you require 45Vdc then the input DC when mains voltage drops to lowest supply range must >50Vdc and preferably >52Vdc.
When mains comes back up to nominal supply voltage expect the dc input to output differential to be around 10V, i.e. your existing 49.5Vdc should be ~ 55Vdc
If you require 45Vdc then the input DC when mains voltage drops to lowest supply range must >50Vdc and preferably >52Vdc.
When mains comes back up to nominal supply voltage expect the dc input to output differential to be around 10V, i.e. your existing 49.5Vdc should be ~ 55Vdc
No 10V in this case, its at the upper range edge. It will threaten Q102. 7V max Vin-Vout. If the mains are steady and 5V is always available, OK.
if the mains supply is not steady then your advice leads to trouble.
I know of no mains supply that has a +-0% tolerance on nominal mains voltage.
If he needs 45Vdc at the output then show him which components need changing to allow for +60Vdc on the input (mains supply at maximum of range).
Q102 is a 50V Jfet. Can't get any low noise compatible with more Vds max that I know of. He should see his DC at C105 not dropping below now value for a period and be conservative.
Yes, this would be very welcome.
Better for +62Vcd, then I could use the the 22-0-22 V tranny I have.
Franz
Then change the circuit (or maximum output voltage advice) to tolerate/survive over voltage due to mains supply voltage variations.
Franz,
22+22Vac is far too high.
mains +-6%.
regulation 10%
diode drop @ full current 700mV
diode drop @ near zero current 500mV
minimum DC voltage at smoothing caps ~= [44*0.94*sqrt(2)]-1.4 ~57Vdc +-half the ripple voltage when supplying full current.
maximum DC voltage at the smoothing caps ~= [44*1.06*1.1*sqrt(2)]-1 ~71.5Vdc +-half the ripple when near zero current.
That 22+22Vac gives a range of 57Vdc to 71.5Vdc.
You would have to design the Regulator to survive all those voltage variations at the input.
You are asking too much.
Franz,
22+22Vac is far too high.
mains +-6%.
regulation 10%
diode drop @ full current 700mV
diode drop @ near zero current 500mV
minimum DC voltage at smoothing caps ~= [44*0.94*sqrt(2)]-1.4 ~57Vdc +-half the ripple voltage when supplying full current.
maximum DC voltage at the smoothing caps ~= [44*1.06*1.1*sqrt(2)]-1 ~71.5Vdc +-half the ripple when near zero current.
That 22+22Vac gives a range of 57Vdc to 71.5Vdc.
You would have to design the Regulator to survive all those voltage variations at the input.
You are asking too much.
Last edited:
Andrew
I understand, thanks.
I could insert an LM317 regulator circuit before C105, adjusted at 52VDC.
Franz
I understand, thanks.
I could insert an LM317 regulator circuit before C105, adjusted at 52VDC.
Franz