BD140 could possibly prove workable in your relatively low voltage application when you can't find the original but its not equivalent in spec. Although I did not have the need to try it so to say if and how it works in this circuit for sure. Yes the Q4 Q5 PNP cascode is calculated to divide the voltage at about half for even dissipation. As you have measured in your build. The 2SA(KSA)1381 is not only sure to work, its also a significantly better spec part than the BD series (Philips grey color old issue was faster than modern shiny black BD but still no comparison to 1381's linearity) so stick to it if possible if not only for practical concerns but for quality reasons too.
Ok., will use BD140 temporarily. Farnell has delivery and tax deserved for a wider cart. Enjoy Europe.
I've been optimizing the SSHV2 for my load. BSP129 as Q2 wants an R load of 122R. Slight higher value than 100R intended for DN2540. 100R is suitable also for IXTP01N100. R6 1,8R is turned to be very imporant to sound quality performance, I ended up replacing for a hand made NiCR resistance wire of value 1,7R which seems to be best value for me. R7 100R works, but on some IRF840 samples sound better 103R, which I find very disturbing. I have bought already a dedicated tester for transistors, but I am afraid no matter, further optimisations again on an empirical way.
Is been difficult setting the choke value on the rough DC side. Choke don't let variable adjustement like a pot. I have used a 10 Henry between the capacitors. Then I calculated trough formula 86V (B+) / 0,145mA (load consumption) = 593 / 1000 ( 50Hz AC mains) = 0,593H.
Being suspectingly low I picked my lower choke value is 2H, a resin potted Bartolucci, that proved going this low, removes bass and midrange frequencies. Otherwise going up above 10H. has proven to be very succesful, as two 10H in series (20H) turned to be much better balanced than 10H.
Only a bit less high frequencies extension than 10H, so I may guess the correct value must be between 15H and 20H.
Is there any other formula available, that prevent buying expensive chokes at glance?
Being suspectingly low I picked my lower choke value is 2H, a resin potted Bartolucci, that proved going this low, removes bass and midrange frequencies. Otherwise going up above 10H. has proven to be very succesful, as two 10H in series (20H) turned to be much better balanced than 10H.
Only a bit less high frequencies extension than 10H, so I may guess the correct value must be between 15H and 20H.
Is there any other formula available, that prevent buying expensive chokes at glance?
Last edited:
All formulas are included in what you can predict with graphs by software tools like Duncan's PSUD2 free simulator.
Why you doubt the information I gave you in private????
Also you don't like simulators because is a lot of work....
Use your methode to put & listen if you don't believe in formulas, there is no different formulas to acomodate your taste.
Felipe! Beg yur pardon!!!
You try help me in private, and in return sharing here! I notice my bad, please not hard feelings, my apologies.
Jordi
You try help me in private, and in return sharing here! I notice my bad, please not hard feelings, my apologies.
Jordi
Salas could bé used without modification enhancement mosfet in place of depletion mosfet at cascode CCS?
No, no way. Depletion bias is normally on. Enhancement bias is normally off.
In the two datasheets you posted both say normally on in "Features" to the right side page 1. Although not stating depletion in one's headlines they probably both are.
bought IXTA3N100D2 or IXTP3N100D2 from a cheap vendor. They appeared to be 'normally on' in name only.