Choke primer

[Structo]

I believe it stands for DC resistance.

[roberto]

Rdc is resistance dc. What you see with a ohm meter.
You can ehnance this effect by a series resitor 100/220ohm.

[Jana]

In reference to the theory that high values of filtering lead to ghost notes.

I find that hard to believe. If the person who claims that did indeed experience that phenomenon, I suspect there was something else at play that just happened to disappear when lowering the capacitance.

Never in all the amps I have built has ghosting resulted from large amounts of filtering. I use 600uf of filtering in my favorite amp design and there is absolutely no ghosting. None. My results are repeatable from one amp to the next.

[iknowjohnny]

You can ehnance this effect by a series resitor 100/220ohm

Would that add to a more midrangy tone as you mentioned? I have tried a resistor IN PLACE of the choke and that doesn't sound more midrangy. But i guess i'll try both in series and see what it sounds like since it's a 2 second tweak. Thats one thing i always feel i would possibly like to change about the tone of this amp is to add some more response in the 500-700 Hz range or thereabouts.

Edit: series would be either a sag resistor or adding resistance to the screen supply. Hmmm......i guess i missed something there.

[roberto]

jana, really 600uF for first filter? It's a 100W guitar amp? Do you use high filtering also on screens?

iknowit, just disconnect a choke wire, and add a resistor in series.

I usually use 100 to 220u, 3 to 5H choke (100mA for 100W, half for 50W), 50 to 100u screens.

[Jana]

No, 600uf total in the B+ supply chain. 200uf on the first filter.

[iknowjohnny]

just disconnect a choke wire, and add a resistor in series

Got it, tried it, but it just basically sounded like it does when i remove the choke altogether and use a resistor. Thinner mainly. I prefer the choke by itself, but thanks for the suggestion.

[roberto]

Real chokes are ideal chokes in series with a resistor (and something else around). If you add a resistor, the bigger the resistor, the less ideal the choke. So yes, it will sound more like a resistor. What I was talking about, is the effect of different Rdc chokes (due to wire resistivity, non to added resistors).

Jana, now I'm into it. 600uF, something like 200u - choke - 100u - 50u PI - and 5x50u nodes (maybe loop, 4th stage, 2nd+3rd OD stages, 1st OD stage, Clean stages).

[memphis032]

Hi there,
I'm putting together a Dumble "clone" with a parallel SE 6L6 output, and had been planning on a pi filter with 5H 200mA choke before the plates. I think I've been dissuaded by his topic so far, and will probably go for a smaller choke after the plates, but the topic of standby came up earlier. I hadn't really thought about the choke's self inductance when switching standby! Is there a sensible way to do it, or do I just switch the grids of the power tubes to ground?

[Structo]

What I remember from class was that inductors oppose a change in current.

Capacitors pass AC and block DC.

So the choke opposes the ripple current smoothing it some, and the caps shunt the AC content to ground, leaving DC power.

I try not to second guess Alexander Dumble when building one of his amps.

If you want it to sound like a Dumble you have to build it like one.

[memphis032]

I think you missed my question...

Is there a sensible way to do standby switching when there is large choke in the supply, as just turning off the HT could have disastrous consequences due to the sudden change in current. Maybe a large cap across the switch?

As I mainly use standby for muting, the alternative is grounding the output tube grids when in "standby".

[Firestorm]

I don't think grounding the grids is the way to go. Obviously you can't do that with fixed bias, but even in cathode bias it seems like grounding the grids would be noisy at best. Why not lift the cathode ground; maybe put a large resistor in series with the cathode resistor and short it out for "play" mode. Or switch the screens between ground and B+2.

[Super_Reverb]

Most designs don't filter the output plate supply because the current is quite high compared to preamp + phase inverter current draw. You could use a choke to filter the plate supply, but due to current rating required, wire wrapped around iron core would have to large --> due to larger guage wire, choke would be physically large, heavy, and consequently expensive. Designers use LC filtered screen voltage, because current is low compared to plate current.

Inductors:
1) store energy in the form of an electromagnetic field
2) resist changes in current --> changes in inductor current result in changes in inductor voltage to counter change in current --> V=L*(di/dt)
3) frequency domain --> Z=2*pi*f*L --> shorted at DC --> open at high frequency

Capacitors:
1) store energy in the form of an electrostatic field
2) resist changes in voltage --> changes in capacitor voltage cause changes in capacitor current to counter original change in voltage --> I=C*(dV/dt)
3) frequency domain --> Z=1/(2*pi*f*C) --> shorted at high freq --> open at DC

cheers,

rob

[Roe]

> a choke input filter .... They do provide the best regulation, though.

I believe this is semi-true with vacuum rectifiers and costly caps, not at all true with sand rectifiers and big cheap caps.
...

In some cases it might be true even with sand rectifiers and cheap caps, since adding more capacitance will sometimes change the tone and feel too much. In this case, a choke at the mains can help remove ghost tones without getting a stiff feeling.

[rdjones]

The early Sunn amps used a CLC "pi" filter before the output plates.
This applied to the models with tube rectifier and EL34s as well as solid state diode/6550 versions.
The same is true of the Dynaco Mark III that the Sunn was based on.

The choke is fairly small considering, about the size of a Fender choke.
Small enough to fit inside the chassis.
I don't know the specs on the choke.

reddog Steve