DIY tube tester

Hi folks,

I am currently (trying to) design a computer controlled tube tester. My first goals is to do a bunch of tests on double triode tubes like the 12A*7 family. I only recently started with this, and so far only have my transconductance test up and running. I can now read out the values for gm for both triodes.

Now I would like to implement an emission tester, but I am having trouble understanding how this test is best performed. In the case of the 12AX7 tube, I would have to connect the grid to the plate and connect the cathode to the ground to make the tube function as a diode. That I understand. So now I am wondering: 1) which voltage should be applied to the plate+grid? 2)When I then measure the plate current, which value do I use to compare it with?

Also, as I am programming the user interface in a computer program that is able to perform frequency analysis on the measured data, I was thinking of doing some sort of test for example to find relevant disturbances within the audio range, decreased gain response at certain frequencies etc… Does anyone have experience with frequency related phenomena in old/bad tubes?

Please excuse me if these are stupid questions.. I am relatively new to the vacuum tube world :

Any help, advice, reference to a good book,… would be very much appreciated! If you would like to know more about my project, don’t hesitate to ask me!

Lebowski

Thank you for your kind reply!

About the emission tester: I have read about the emission measuring method you mentioned. I have also found a small table with the combined anode-grid voltages that should be used, and the current value to compare your own measurement with. For an 12AX7 -according to this table- the combined anode-grid voltage should be 25V and the ideal anode current that results from this would be 50mA. They also mention that this test should be only performed for a maximum of 3 seconds, to not damage the tube. Honestly, I find it difficult to believe that a 12AX7 tube would not be damaged by a 3 second emission of 50mA. I would think that the cathode would boiled by this test. Any thoughts on this?

Would it be possible to do this test by connecting the grid and the cathode to ground, so that the bias is 0V, and then applying for example 100V to the anode, and measure the anode current? Because then one could easily compare the measured current to the 'ideal' current on the Va-Ia curve @ OV bias. Do you think this value could be representative?

About the reduced filament voltage test: I have indeed heard 'stories' about this, but have found no references at all to this test method. I decided to test a couple of new and old 12AX7 tubes for transconductance at reduced heater voltages (8 values between 4.8V and 6.9V). The result for every tube were surprisingly linear (so no sudden drops in transconductance at lowered voltages). All I know about the old tubes is that they were at least 40 years old, and have been used and replaced. I will look into this with greater detail. It would be great if anyone could provide me with a reference, advice or experience about this test method!!