• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Low voltage high current rectification

Jump to Latest
#42
It is all engineering.

I have built brute force DC supplies for a 2A3 filament.
6.3VAC secondary, Schottky diode bridge, 22,000uF, Series resistor to get 2.5V out (as well as to get lots of low pass filtering given the 2nd capacitor, another 22,000 uF cap across the 2A3 filament).

Power-on . . . is like an Olympic Weight Lifter . . . UGH!
But it works.

Quality parts, with high enough ratings, and it will outlast the very best quality 2A3 tubes.

Oh, and it is an Intrinsic Soft Start Circuit.
The very high wattage series resistor, is much higher resistance than a 2A3's Cold Filament resistance.
Wastes Power, puts out heat, yes.
Simple, but effective.

Not the only engineering solution, just one of the working solutions.
Pick and design your own simple or complex solution.
 
#44
I was wondering more specifically about the non-linear switch-mode rectifiers and regulators. Of course, a Coleman type regulator solves all signal performance issues, but has a thermal overhead penalty. No biggie for 2A3s and such, but am currently interested in 304TLs and parallel 4E27As (delayed mid-life crisis) where a different approach seems important (responsible?). Your D M-L C project has some similarities.

All good fortune,
Chris
 
#46
Slow start can be achieved by using a super capacitor in the after-choke position. Something like 1-2 F. 2.7 V rating is just enough for 2A3, 45, or 1624. For 300B more complicated, need 2 in series and protection circuit.

Yes, low voltage Ge power diodes need protection from the initial voltage spike at start-up. A small capacitor or a varistor, or both.

What I like about choke input at high current low voltage supply is small critical value of the choke, a few millihenries. Ten turns of thick wire on an inch-size iron powder core, no sweat winding it.
 
#47
trobbins said:
Commutation noise from a choke-input filter can be managed by deploying a small capacitor before the choke ... The part value of such a capacitor would normally be low enough not to noticeably raise the output dc voltage .... [and] also likely slows the commutation dV/dt of the diode's node.
Click to expand...
Yep, I agree with this. I use the smaller cap and a Quasimodo-derived snubber and results are good.
 
#50
Check Micrometals or Amidon catalogs, they give Al for different cores. Some materials have relatively high permeability while allowing substantial DC current. Below is a link to one of the inductors on eBay. It is 1 mH 10 A. Double the turns on this core, and it will be 4 mH 5A. Critical inductance for a pair of 2A3 filaments is about 0.4 mH.
https://www.ebay.com/itm/3139308674...CZDsfwONaTLlZQdfeFqNDhWw==|tkp:Bk9SR7zU6_PeYw
 
#54
Bucks, I agree that the inductor in the post 50 is likely exaggeration. But here are Micrometals catalog data for a couple of their cores:
T106-45 (1" OD) Al=125 nH/N2, 60 turns give 0.44 mH,
MP-133205-2 (1.3" OD) Al=267 nH/N2, 50 turns give 0.675 mH.
CM chokes in post 52 are not for input choke, but for filament isolation, as in post 28.
What's the difference between choke and inductor?
 
#55
sser2 said:
Check Micrometals or Amidon catalogs, they give Al for different cores. Some materials have relatively high permeability while allowing substantial DC current. Below is a link to one of the inductors on eBay. It is 1 mH 10 A. Double the turns on this core, and it will be 4 mH 5A. Critical inductance for a pair of 2A3 filaments is about 0.4 mH.
https://www.ebay.com/itm/313930867404?itmmeta=01HVVKTX8FPBD9VR8N1H9NPZTG&hash=item4917bce2cc:g:T6AAAOSwuYFlWBx9&itmprp=enc:AQAJAAAAwFz8FhdzgF5oVktXNIrDWvotKKr1uDbhIXNFezGar2/eSaUFYMD3xxzxLHf5VmSaZtNZ6eZQMATwc5dFv1z+KWV0V4tfb+GwT9NwqYTx9e3/l90tIILC4FG+SrZSM34wNdEfo6oT31S29rwb7Z0b8czLnfPj3ZpHs3eo5+0r8D6OE1viK6eCJFFWZbzqg6Qbt+rW4klJgy+XLhw2qoyMuax7nnQfJwWqZVvmtg+y0MCZDsfwONaTLlZQdfeFqNDhWw==|tkp:Bk9SR7zU6_PeYw
Click to expand...
Based on the number of turns, and assuming a peak induction of 1.25T, the effective core area should be 6.66cm². I didn't see a spec for the dimensions, but if the area is indeed 6.66cm², the wire gauge has to be very substantial if the scale of the pic is to be respected
 
#57
sser2 said:
Bucks, I agree that the inductor in the post 50 is likely exaggeration. But here are Micrometals catalog data for a couple of their cores:
T106-45 (1" OD) Al=125 nH/N2, 60 turns give 0.44 mH,
MP-133205-2 (1.3" OD) Al=267 nH/N2, 50 turns give 0.675 mH.
CM chokes in post 52 are not for input choke, but for filament isolation, as in post 28.
What's the difference between choke and inductor?
Click to expand...
Precisely a common-mode choke vs a storage inductor. The common mode choke always is fitted with 2 identical windings t hat are wired in a way that current-flow induced magnetic field inside the core is cancelled. This explains very hi common mode inductance and saturation current, several decades bigger than storage chokes of same dimensions. Hence storage chokes of high inductance with high saturation current are really bulky - there is no free lunch.
 
#60
There are formulas in the catalog to calculate effect of DC current on permeability coupled with mu-H curves. According to this information, the T106-45 60-turn inductor will have approximately half of its no-DC inductance at 4A DC, or about 220 mH. However, high AC magnetization will somewhat compensate for the loss of permeability, so it will be in the ballpark of 300-350 mH. Same is true for the MP inductor, for whose material permeability drops faster with DC magnetization, but it has longer magnetic path and fewer turns.
 
  • Like
Reactions: Rod Coleman
You must log in or register to reply here.
Top Bottom