I'm in the process of finalizing aspects of my computer desk amp project, chassis making, layout, bread boarding, etc. I'm now entertaining the thought of using DC for the pre-amp stage filaments (6J6) and have extra current to experiment/listen with trying the power tubes (6BX7) as DC vs AC.
I would like some advise on when to bother with regulating and when its not necessary.
I understand that with DH filaments there is a great advantage of using a regulator like the Coleman regulator. Because the filament IS the cathode and all that that implies.
But for tubes with separate cathode and filament would you put a premium on voltage regulation for the DC supply? Considering the extra space you need for an external heat sink and the extra current/voltage you'd need for the regulator cutoff, etc.
For non DH tubes, is a 1950's style RC filtered supply, except using the large uF capacitors now available (that were not available in 1950) good enough? Filaments prefer current regulation correct? Yet the many circuits I see use a simple voltage regulator, few and far between are current regulated as I peruse schematics.
In laying out my chassis (actually a wood/aluminum box, think 1960's tube table radio-like). I have an area reserved (2 inch by 4 inch) for the DC supply components/board inside. And I have an area on the back where I can hang a flat heat sink up to 4 x 5 inches. Which is quite a lot of heat sink! This is good because I want to use this same form factor for other projects, when I get out to the table saw I will probably make 4 of these cases for my next 4 projects.
So my dilemma is do I go regulated or un-regulated? Is the payoff for going regulated really worth it over a really good low ripple RC supply?
I can obtain a middle ground in DC current regulation (without using a regulator) by upping my filament voltage to maybe 9V DC and inserting a large resistor to bring it back down to the 6 volts. (that cost is heat not space). I have total flexibility on voltage because I have a space for a dedicated filament torroid, so I can buy any voltage of filament torroid I may need and just plop it down.
Thanks for weighing this, I want to learn how to do things not in absolutes, rules, or biases like "I always do this, or I always do that, just because".
I would like some advise on when to bother with regulating and when its not necessary.
I understand that with DH filaments there is a great advantage of using a regulator like the Coleman regulator. Because the filament IS the cathode and all that that implies.
But for tubes with separate cathode and filament would you put a premium on voltage regulation for the DC supply? Considering the extra space you need for an external heat sink and the extra current/voltage you'd need for the regulator cutoff, etc.
For non DH tubes, is a 1950's style RC filtered supply, except using the large uF capacitors now available (that were not available in 1950) good enough? Filaments prefer current regulation correct? Yet the many circuits I see use a simple voltage regulator, few and far between are current regulated as I peruse schematics.
In laying out my chassis (actually a wood/aluminum box, think 1960's tube table radio-like). I have an area reserved (2 inch by 4 inch) for the DC supply components/board inside. And I have an area on the back where I can hang a flat heat sink up to 4 x 5 inches. Which is quite a lot of heat sink! This is good because I want to use this same form factor for other projects, when I get out to the table saw I will probably make 4 of these cases for my next 4 projects.
So my dilemma is do I go regulated or un-regulated? Is the payoff for going regulated really worth it over a really good low ripple RC supply?
I can obtain a middle ground in DC current regulation (without using a regulator) by upping my filament voltage to maybe 9V DC and inserting a large resistor to bring it back down to the 6 volts. (that cost is heat not space). I have total flexibility on voltage because I have a space for a dedicated filament torroid, so I can buy any voltage of filament torroid I may need and just plop it down.
Thanks for weighing this, I want to learn how to do things not in absolutes, rules, or biases like "I always do this, or I always do that, just because".
Why do you believe that you need DC heating? DC heating for a tube like the 6BX7 will add quite some current draw from your power transformer, and you may not be able to get enough raw DC voltage to regulate the DC supply in the first place.
You may very well find that things are adequately quiet with AC heating.
You may very well find that things are adequately quiet with AC heating.
In the old days, we built amplifiers, preamplifiers and other electronics using directly heated valves and as long as the AC is balanced the AC cancels itself out, if one side of the cathode has the positive peak the other side has the negative peak that is reversed at mains frequency, the ripple is cancelled and we never used DC for heaters as there was no need.
No, just no.
There are limits to how quiet you can get a 300B heated with a 5V AC transformer winding. If you add a directly heated driver that's also AC heated, you can really make a mess. I would never expect an AC heated 300B on a 3K output transformer to have much less than 3mV of hum at the speaker posts. I would never expect a DC heated 300B on a 3K output transformer to have more than 500uV of hum.
In the old days, directly heated tubes were often heated by batteries, and that was nice and quiet.
I understand, and honestly I dont know if I believe DC is better yet here, the breadboard and my headphones will, as always, answer that for me. Everything I've done so far has been done by A/B comparison on the breadboard (no sim). (I just finished using Quasimodo ringer to tune the B+ xformer ringing snubber values.) This entertaining of DC for the 6J6 is just me wanting to explore both AC and DC.)
Running the 6BX7 on DC was an afterthought, to maybe overbuild the filament supply as an experiment. I probably shouldn't have entertained that thought for this discussion. So, disregarding the 6BX7 power tubes... Many amplifiers do DC heat the pre-amp tubes. Maybe that's not necessary on many of them as well, probably?
So considering just the 6J6 input tubes (.49 A x 2). The torroid I have now is 6.3V @8A (with dual secondaries in parallel). It will run plenty cool that way for an all-AC setup. Running the 6J6 (only) on DC I could separate the paralled secondaries, using one winding for the 6BX7 (AC) and the other into a simple bridge RCRC filter to get back to 6.3V DC, with amps to spare as that winding is rated 4A and one 6J6 is .49A.
But going to regulated DC, all that is out the window, a 6.3V winding wont get me over the cutoff. So my original question... "to regulate or not to regulate?".