The Lundahl that you mention at the recommended 15 mils dc would have a voltage drop of 36.

On most of our plate chokes we try to stay at or below 20 volt drop. Worse case that I can think of is our unit with a published dc current rating of 50 mils which has a dcr of approx 550 ohms which equates to a 27.5 volt drop and a dc copper loss (in watts) of 1.375.

Like Lundahl and many other manufacturers we do not publish Cw.

cheers,

msl

see if I can give you some of the "advantages" of a plate choke.

first...your right... a plate choke will not have the freq response of a resistor. the resistor wins hands down on that criteria.

where and when a plate choke may have some advantage is...

say you were going to use a 22K ohm plate resistor.... so that it presents the anode with a 22K load impedance... this is what the anode works into load wise (keeping it simple for the moment)....

now your going to have a voltage drop across that 10K to 22K ohm plate load resistor... of X amount.

now sub in a plate choke... depending on specs and etc... but suppose that the plate choke has a dcr of 985 ohms and so at any given current level you will drop Y amount of dc supply volts across this resistance.

So the plate choke will typically have a lot less resistance than the resistor has. Hence less supply volts are lost through the loading mechansim of the anode.

But... we have yet to account for the important work of the load resistance... that it supplies a load impedance and it's this magnitude of load that does the work...

the resistor basically provides the same amount of load impedance irrespective of frequency (especially in the audio band)...

the plate choke... because it is a reactive component... the load impedance that it presents to the tube is dependent on the amount of inductance that it produces and the frequency...

so if we have a 100 henry plate choke... then at 20 hertz it will provide or make an inductive reactance of 12,567 ohms. At 40 hertz the inductive load impedance seen by the anode of the tube will be approx 25,000 ohms and at 80 hertz 50,000 ohms and at 160 hertz 100,000 ohms and etc. So that as your double the freq and if L remains constant then the inductive reactance also doubles in magnitude. At some frequency the L will begin to taper off and fall... but still we have in the equation L times F ... so that even though L may be falling as we go up in freq... the freq itself multiplies the effect of L (if I am saying this the right way) and produces a load impedance of Y ohms.

so that a plate choke may offer some advantage in providing a larger load impedance to the anode of the tube while consuming less of the raw dc supply voltage in series with it....

that might be one of the advantages of a plate choke in simplified terms.

msl

I keep wanting to call you mercedes benz. Wonder why?

Perhaps this will (or maybe it won't :=)) be more intuitive....

say on a conventional series feed single ended output stage... you have an output trans with two windings... a primary and a secondary... the dc plate supply for the anode goes in the primary from one end and gets hooked to the anode of your 2A3 or 300B on the other end....

guess what you have.... you have in essence a plate choke feeding the anode of your tube... only now the ac signal is also in series with this same winding...

the primary of your output transformer produces a certain amount of L... it is this L (in para w\ the reflected load from the secondary) that governs or determines the magnitude of load impedance as seen by the tube.

so... in a sense your standard output transformer could be thought of as a plate choke with a secondary hung on it....

tell me if that helped or not...

msl

Yet the sonic benefits and power stage benefits to using a grid inductor, would sugget otherwise. So, for a given productin of grid charge, the inductor seems able to deal with it in a way that is less disruptive to the signal than simple Ohmic resistances. I have made some interesting measurements on an amp equipped with a well known multi-kHy grid choke that support this theory. AC grid current is treated resistively in the same way an equivalent resistive load would. So for a given omega*L the high L chokes begin to behave like resistors. In addition to this, the grid choke has substantial capacitance in parallel. So, use just enough L to get the job done is my current practice. Fortunately it is quite easy to custom specify coil production with Heyboer. those guys are brilliant! For me, it's indespensible. Can't make reasonable tube amps without them.

I will soon take delivery of some grid chokes specially wound for both balanced and minimized winding capacitance. The previous ones were simply balanced, and less attention paid to winding capacitance. They still smoked a few other devices I managed to acquire samples of. Even some allegedly well respected and well thought of designs.

I have a quote on amorphous C-cores in qty which makes them *VERY* attractive in comparison to laminated sheet steel. Hardly more than low qty M6 even. Just a question of absorbing that sort of volume....:)
cheers,
Douglas

somebody will, I guarantee it.
cheers,
Douglas

Or just sit back and wait for an interested customer to publish thier won measurements. This leaves the question: do you want the quantity in question, in this case winding capacitance, to be a matter of speculation and private measurement, or of public record from your own measurements?

The former will certainly occur, would you be happier with that or your own measurements?
cheers,
Douglas