Long-winded capacitor question

[newguy]

What can I say. I continue to be in awe of the generosity of the members of this forum.

PRR: Thank you so much for the lucid and extensive explanation! I really appreciate that you took the time to write that up, and it helped me a lot.

I understand what you’re saying about how to do the calculations, but I am not totally sure that I “get” the logic behind the calculation of cathode impedance.

Let me take a stab at one part of it … is it at all fair to say that if cathode impedance = 1/gm=(change in grid voltage) /(change in plate current) then that’s almost analogous to r=v/I?

Thanks also for the pointer on Radiotron Designer’s Handbook v3 … at $7, I figured I couldn’t go too wrong, so I snagged that copy off of abebooks and it should be here next week.

Brett: Thanks for the pointer on MJ v3, and for the encouraging words! For some reason Amazon showed it as “not yet released”, but B&N had it for same day delivery, so it is sitting with my doorman at the moment. Hopefully I get out of the office before dawn so I can read it. Since Radiotron v.3 was so inexpensive, I figured I would get a start on that and then graduate to v.4 when I have some clue what’s going on.

Actually, based on your recommendation from another thread (to someone else), I’ve started to read through the tubecard article “The Grounded-Cathode Amplifier”. Hopefully together with the Morgan Jones book I’ll be able to understand a bit more of what’s going on here…

I tried to run the PS of the waarde amplifier through PSUD II, but wasn’t sure how to determine the impedance that the power supply would “see” (almost back to the question that started this thread).

For what it is worth, the price of Radiotron v.4 on CDROM seems to have fallen to about $30…

Wizdeluxe: Thanks for the pointer to the Solen Fast Caps. Given where I am in my understanding, the (apparent) simplicity of the van Waarde amp appeals to me. I’m sure it’s not the best out there, but hopefully it’ll be good enough. In any event, I’m not planning on touching solder to iron for a little bit.

Sorry if this message is rambling, it’s 3:30 a.m. where I am and it’s been a long day. Thanks again.

[Brett]

Hi,

For $7 RDH3 is a great buy. It has most of the info you'll need for quite a while yet, and having v3 means you can watch out for a bargain v4 at market, hamfests or even garage sales.

The Tubecad article is really excellent, and the way he's written it makes it very easy to understand the principles. My advice would be to concentrate on that and MJ until it all begins to 'click' and use RDH to flesh out the technical details.

I tried to run the PS of the waarde amplifier through PSUD II, but wasn’t sure how to determine the impedance that the power supply would “see”

A tube (triode) is basically a voltage controlled variable resistor so the voltage across the tube itself, divided by the cathode current gives you a resistor value for the tube at that operating point (add in the cathode R too). You can also right click on the resistor, select change, and choose the current source option, so then you just need to tell it the cathode current. There are pro's and con's to both methods, but both should be close enough. My experience with PSUD for tube ccts is that it will be very close to correct if you put in all the correct values for the components, but is usually about 3% low which is not a problem.

I also model it in two differnet ways; I check performance in the first say 3-5 secs to see the start-up, and then choose a 200ms window about 10s+ (use the reporting delay option) in to get the final stable voltages.

[PRR]

> wasn’t sure how to determine the impedance that the power supply would “see”

For a first approximation: supply voltage divided by supply current. I forget Waarde's numbers, but say 250V/0.15A= 1,667Ω. As Brett says, with triodes this usually does not vary much with power voltage (if the voltage is lower, the current is lower, so the impedance stays about the same).

For a Class B design, you also need to check the full-load current, see if voltage sags too much or if ripple gets huge.