A Simple, Valve Curve Plotter

[Alan Douglas]

Ferrite beads hadn't been invented when the I-177 was built, but luckily high-gain pentodes hadn't been either.

[Ed_Dinning]

Hi Nick, as you are only interested in early valves a useful feature would be to be able to read Ra directly instead of calculating it from other parameters.
Ra is important if you are sorting unknown valves as you can then determine if it is suitable as an RF, AF amp or detector.

Ed

[Nick]

Standard pin numbering for the more common bases:

   

[Yorkie]

I can conform that the pin layout and pin diameter of a B5 base is exactly the same as a B4, but with the added centre pin. Thus, a B4 valve will fit into a B5 base, but not the other way around.

Some time ago I made several valve base adaptors to extend the range of valves that can be tested on my Taylor 45D valve tester, so that these additional bases could be plugged into the Octal test socket on the 45D.

Originally such adaptors were available as accessories to the 45D but in the ensuing decades have become unobtanium.

I did post a thread on the forum and an article about the adaptors was published in the BVWS Bulletin, but a couple of pics are attached which show the B4 and B5 bases, among others.

[pwdrive]

Some common valve base dimensions:

http://www.tubecollector.org/documents/bases.htm

Lawrence.

[SurreyNick]

....a useful feature would be to be able to read Ra directly.....
Ed

Ra was one of the formulae in my spreadsheet Ed. Without complicating the design too much, how would I read it directly?
Thanks
Nick

[pwdrive]

Nick (Surrey) A simple way to determine Ra without plotting curves would be to apply a fixed bias voltage to the grid and measure the voltages at the anode and the anode currents for two different load resistances, divide the difference of the former by the difference of the latter.

There's a set up in the RCA Home Study Course here (experiment lesson 22):

http://mikeyancey.com/files/RCA%20Home%2...iments.pdf

It shows a log pot across the -ve supply with a shift resistor in series for setting the grid bias, the log pot can be replaced by a lin pot, the shift resistor wouldn't be need then, calibrate the pot and you wouldn't need a voltmeter to measure bias voltage.

The B + supply used is 100 volts, that should put the reference points well into the most linear part of the Va Ia slope for most valves.

Here's the link about valve characteristics that's relevant to the above experiment, it's well written and clear and uses a known valve (6J5) and it's published curves and specs for the calculations, well worth a read.

http://km5z.us/files/RCA%20Home%20Study/...stants.pdf

The link to most of the rest of the course:

http://mikeyancey.com/RCA_Radio_TV_Course_1958.php

Lawrence.

[Nick]

Ra (or Rp if you want to call the anode a "plate") is a funny one as it is completely dependent on the chosen operating point (Va & Vg), i.e. its a dynamic resistance - it is defined as the change in Va (delta Va or dVa) divided by a given small change in Ia (dIa) and can be visualised as being the "slope" (first differential) of the Ia (y-axis) vs. Va (x-axis) curves for a given Vg.

If you look at these graphs it becomes obvious that Ra can change fairly dramatically depending on the chosen operating point.

The relationships between Gm (transconductance), mu (gain) and Ra are pretty straightforward:

• Ra   = dVa / dIa
  
• Gm = dIa / dVg
  
• mu = dVa / dVg
  

Gm & mu are dynamic (AC) not static (DC) values... A bit of algebra and you get:

• mu = Ra * Gm or Ra = mu / Gm
  

The effective Ra, i.e. what the circuit sees,  is the dynamic Ra from above, in parallel with the anode load resistance and the load from any following stage (e.g. the following stage's grid resistance).

Most data sheets compute Ra by simply dividing mu by Gm... i.e. its a 1st order approximation only at a stated operating point.

[pwdrive]

It is and it does, say's so in the books.

If Surrey Nick designs his valve characteristic metering device ok he should be able to work out Ra no problem, it'll be a good unit to build, learning as he goes.

Lawrence.

[SurreyNick]

This thread is really useful, thanks.
I'll be out of touch for a couple of days as my wife and I are leaving the boys at home and making a flying trip to Wigton property hunting. We've finally had a confirmed offer on our house!
Nick