A choke input supply has advantages over a capacitor input supply. It draws almost constant current from the transformer rather than high current pulses. It has also lower impedance. Also, using a couple of chokes in the power supply allows you to use capacitors with modest capacitance. Then you can use film capacitors rather than electrolytic.
Yes, completely agree. (One cannot argue with physics! Well, one can, but it never ends well.)
What I find interesting is that the choke supply was originally the ONLY way to build a power supply, because electrolytics were small and expensive until about the 1970s. Yet with the introduction of large capacitance electrolytics in small packages and at relatively low cost the choke fell by the wayside. Yet it has so many advantages, as you noted.
I have not seen good hard data on whether or not a large capacitor bank, say 1,000uF, is as good as a small dual (smoothing and swinging chokes) choke supply with 1 or 1.5H chokes and 100uF per stage. I've run the filter calculations which show a point of diminishing returns. But that doesn't address the real world under transient load. I must learn more about analog simulation to model this.
I think your explanation could be easily applied to making power supply chokes.
Why would AC heaters pull less continues current than DC heaters? I have experimented with different heater supplies. In my amplifiers, which have indirectly heated cathodes, I have found that heating the tubes with AC gives better sound. If you are careful when doing the heater wiring, you will get no problems with hum. Directly heated cathodes may be different, but I have no experience from it.
Sorry, that was poorly worded from brevity.
What I was trying to say is that from what I've read once a DC heater is hot the current demand is relatively constant, but the current load varies considerably during initial heating. This is why DC heater regulators are often Constant Current Sources (CCS) to reduce the startup load on the power supply (transformer, rectifier, and caps). When cold the current demand is very high, of course, but this falls off as the heater warms up. So CCS doesn't stress the power transformer as much.
Now as far as AC, I don't know exactly what AC heaters do in terms of current demand once hot, because I've read that it slightly varies with the AC cycle. The initial load of an AC heater can, as with the DC heater, be quite high, of course, and since it is not generally regulated (one could use a non-linear thermistor to gradually ramp up the current, of course) this can stress the transformer on startup.
So I was disclaiming knowledge of the effect of AC heaters on the transformer magnetic flux available for other windings on an ongoing basis.
What I have read is that having B+ on the same transformer can siphon off heater current during transients. So a DC heater supply would, if it had a bank of caps and maybe a choke to stabilize, be reduce that leeching problem.
As far as hum, I have read the trick of using twisted pair with solid-wire. I am surprised that DC didn't deliver lower noise in your experiments.
I have also read that keeping the heater windings on a separate transformer is the best. On my pre-amplifier, I did that. A matter of cost, effort and chassis space...
Ahhh, yes, chassis area.
When I discovered how and why chokes were used I thought, great, why not start adding big ones, like 10H, 20H, etc. to amps. Should be great power even under transients. And then I found out just how physically large and heavy they were, and how much they cost, and thought, hmmm, now I understand why smaller ones are used on most amps.
Sorry, didn't mean to divert your thread into the merits of chokes.
I'm curious about the difference between the various E/I core materials for power transformers. Did you find any difference between them?
I saw you haven't (yet?) wound an output transformer, but I wonder would it be of any quality? From what I've read, the core materials and shapes matter, as does the winding technique, which is why some output transformers are reported to sound better than others.
