hi all,
how necessary is giving each chanel (right and left) its own final RC filter on B+supply if using cascode CCS (dn2540) as anode loads.
any views?
Cheers
how necessary is giving each chanel (right and left) its own final RC filter on B+supply if using cascode CCS (dn2540) as anode loads.
any views?
Cheers
Last edited:
Well, maybe I am missing something, but I wouldn't attempt to use a CCS anode load for more than one valve stage, unless I was paralleling both sections of 12ax7, or another dual triode, for example.
The HT supplying the CCS should be filtered in the same way, I guess, unless some Aikido hum cancelling PSRR improvement scheme would be compromised in doing so.
The HT supplying the CCS should be filtered in the same way, I guess, unless some Aikido hum cancelling PSRR improvement scheme would be compromised in doing so.
sorry, single stage but Left and right channels fed from the same b+.
my terribly worded question (sorry about that) was if CCS loading (and high impedance, cascode), would do you even need both have a Capacitor and resistor on each channel B+, or just fork after the HT supply last Cap, and go to both CCS's
my terribly worded question (sorry about that) was if CCS loading (and high impedance, cascode), would do you even need both have a Capacitor and resistor on each channel B+, or just fork after the HT supply last Cap, and go to both CCS's
I'd still put an RC in, though granted it acts more as a reservoir for the CCS to draw from.
The ripple reduction of the typical DN2540 cascode, or the one I used, was giving maybe a 10:ratio CCS Z to anode Z, so though PSRR is improved, it can be improved as much with decent filtering before hand
Depending on the current requirements for both channels, you might be ok with 1kOhm and 33uF from the main supply, though typically I try to run 220-470uF and appropriate resistance to remove ripple, and let the CCS draw on that supply.
The ripple reduction of the typical DN2540 cascode, or the one I used, was giving maybe a 10:ratio CCS Z to anode Z, so though PSRR is improved, it can be improved as much with decent filtering before hand
Depending on the current requirements for both channels, you might be ok with 1kOhm and 33uF from the main supply, though typically I try to run 220-470uF and appropriate resistance to remove ripple, and let the CCS draw on that supply.
Last edited:
A cheap chip (10M45) CCS will have an approximate equivalent impedance of at least 1 megohm across the entire audio band with more at LF. The average 12AX7 has a plate resistance in the 60K range, and with loading from the next stage, the impedance is probably less than 50K. This gives you at least 26 dB of isolation from B+ ripple and noise. Better tube and CCS choices will have more.
Both my SSE and TSE designs are two stage amps, and use ZERO decoupling on the B+. Both CCS chips, and both OPT red wires are all wired directly in parallel. They are very stable designs that have been produced for over a dozen years.
To avoid possible low frequency instability, I would add decoupling in a 3 stage amp since the input and output stages are operating in phase, or in a high powered push pull amp for the same reason.
Both my SSE and TSE designs are two stage amps, and use ZERO decoupling on the B+. Both CCS chips, and both OPT red wires are all wired directly in parallel. They are very stable designs that have been produced for over a dozen years.
To avoid possible low frequency instability, I would add decoupling in a 3 stage amp since the input and output stages are operating in phase, or in a high powered push pull amp for the same reason.
Exactly, my input stage manages nearly 2M at low current, with the DN2540 cascode, but the output stage, same cascode, higher current, more like 200k.
Heaps of PSRR for VAS, but not so much for power output stage (does that really matter?)
Heaps of PSRR for VAS, but not so much for power output stage (does that really matter?)
I totally trust what Tubelab says here, if no decoupling is required in his experience, then I believe.
I just put decoupling in forces of habit, as long as it's not causing harm, and improving something then I'm happy