Avo Valve Tester

[Sparks]

Lawrence,
Thanks, that is what I wanted to know. I was pretty sure that I was using it wrong, and I had not seen the link on using it that you have just supplied. I have downloaded it, and will look at more testing tomorrow.
I have always had problems with indecipherable instructions, loaded with jargon that meant nothing to me, and your concise explanations have been a great help. I have apparently been checking them with Replace/good scale only, and wanted to know how to check the gm. I suspected that it could be done because of the mA/Volt on the scale, but just did not know how to do it!
Bob

[pwdrive]

No worries, keep at it, if I can I'll try and do a simplified schematic as to what's going on, not sure when though.

Lawrence.

[pwdrive]

Simplified schematic for the Avo two panel valve tester, shown with a triode as the DUT, circuit shown with the mA/V - C Ins. switch contacts omitted except for the grid circuit contacts for that switch, circuit shown as it would be connected with the mA/V-C Ins. switch in the neutral position.

Any errors then guilty as charged.

   

Lawrence

[Sparks]

Lawrence,
Thanks for simplified circuit that gave me the necessary clue for using it correctly (That everyone except me probably knew ) The circuit revealed that the mA/V control had two resistors in one end, and that after setting the zero with it on "100 set," it needed to be advanced about 1/8 of an inch to the mA/V. Then when the joystick was pushed to the left, the needle leapt to the GM of the valve. (I had it fully anticlockwise on the 100 set position), and when I pushed the joystick, the needle hardly moved.
I am sure I have got the hang of it now.
Many thanks.
Bob

[Sparks]

I have now tested a number of valves, with realistic results. I have seven or eight HL2 triodes, the various conditions which have been known to me for some time because of their individual performance. The test results match my observations. I note that when testing pentodes, they don't bother with the supressor grid at all - there is no connection for it. I did try linking it to the cathode, but that just sent the test readings way out, so I suppose G3 can just be ignored when testing! I found that my dirtiest, disreputable looking EF39, testing up round 90% with correct gm, where several nice clean red ones showed 50% or less! I am very pleased with this ancient 79-year-old device now that it is working properly! It probably was from the start, but I didn't know how to use it!
Bob

[Mike Watterson]

No, g3 should be to cathode or 0V or f- always on valve testers. On power types often internally done. It's NOT connected by default on RF types etc because on common grid mode where signal is fed to cathode, g3 goes to 0V or g1 instead, if it was tied to cathode, it would oscillate.

[Sparks]

As I said, I linked it to the cathode, and it threw the readings miles off!
Bob

[pwdrive]

What valve was it?

Lawrence.

[Sparks]

An EF39.
Bob

[Mike Watterson]

As I said, I linked it to the cathode, and it threw the readings miles off!    
Bob

Double check.
It should only be correct with G2 to HT and G3 to Cathode/OV.
I've tested EF39 "properly" wired as a Pentode.

If you are testing in triode mode (g2 tied to anode), then g3 connection should be cathode or anode depending on valve construction (often Anode, but Rod Pentodes* definitely and Beam Tetrode type Pentodes probably want g3 to 0V/Cathode even in Triode mode).

Hexodes are more complicated.
Heptodes and Octodes: Read data sheets, not short form as the same numbered grid can be HT or 0V depending on design. Example is 1L6 and DK92, looking at pin names they seem incompatible bases, but actually they are different designs, and strangely are actually interchangeable on at least LW/MW (tuning will be badly out on higher frequency short wave bands on many radios).

(* Rod Pentodes are not really Pentodes at all, thus for DF97 pentode, triodised instead of a DC90 triode, you tie g2, g3, anode together, but using 1j24b or 1j29b you tie g2 and anode but g3 goes to f- as it's really a kind of beam device, no real grids at all)