Tube tester done in a different way?

[kostas_sv3ora]

As far as concern tube testers, I have read that the best test (even from transconductance) is to operate the tube in the actual circuit that needs to operate.

I was thinking, why not one build a power self-excited tube oscillator and measure it's RF output? The oscillator must be in the range of Longwave, so as even audio tubes will be able to generate RF.
By comparing the RF out of the bad tube, to that of a new one, one could have a direct indication of the specific type tube's amplifying state.

The thing is that with this idea (if I am not wrong) one can do comparative measurements, independent of the test points and not needing all of this metering and calculations. That is, just set a test point where a new tube oscillates and produces high RF power. Use this as a reference for this partucular tube type. Then insert a bad one and watch how much less RF is produced.
This thing is actually noticed inherently when operating RF tubes in old HAM radio equipment. Very used tubes have their output RF powers dropped.
I do not know if this works, but it seems reasonable.

It will not give you actual values to compare against the datasheet.
This idea will give you RF output percentage comparison, between a new tube and a used one. It will use the new tube as a reference instead of the datasheet.
To my thinking the result will be even better (practically) than measuring transconductance, because of the higher input signal (RF feedback) which shows the tube operation in RF and because of the fact that the tube is actually measured in a working circuit.

This will measure a direct percentage of the overal performance of the tube, not for each electrode alone (which is practically all that matters at the end).

The disadvantage is that one has to have a new tube type (and possibly brand?) for every old tube he wants to compare against.

How do you find this idea?

[Chrispy57]

Hi Kostas - nice to see someone "thinking outside the box". You have invited our thoughts on this proposal, so I'll offer a couple that occurred to me.

Firstly, when they say "operate the tube in the actual circuit that needs to operate" that is quite different from "operate the tube in a random design of oscillator circuit". I don't think your proposal would meet that requirement.

Also, if you have, or have to purchase, a new tube, then you may as well fit that into the equipment that is being repaired. You can then compare the new and suspect old tube in that actual circuit if you wish, but obtaining a "percentage of badness in one particular circuit" isn't really very useful in practice.

You either leave the new tube in the equipment as a repair and keep the old tube on the shelf as a "pull", not knowing how it will behave in a different circuit in the future, or throw it away.

It sounds like an interesting experiment for somebody without a tube tester, but would require multiple tube bases and high voltage test equipment to conduct the experiment each time you need to use it.

Cheers
Chris

[Smithdoor]

I did read about this in Popular Electronic back 1960's. I think was for sound work.
They used a signal generator and a VOM.

Dave

Originally Posted by kostas_sv3ora

As far as concern tube testers, I have read that the best test (even from transconductance) is to operate the tube in the actual circuit that needs to operate.

I was thinking, why not one build a power self-excited tube oscillator and measure it's RF output? The oscillator must be in the range of Longwave, so as even audio tubes will be able to generate RF.
By comparing the RF out of the bad tube, to that of a new one, one could have a direct indication of the specific type tube's amplifying state.

The thing is that with this idea (if I am not wrong) one can do comparative measurements, independent of the test points and not needing all of this metering and calculations. That is, just set a test point where a new tube oscillates and produces high RF power. Use this as a reference for this partucular tube type. Then insert a bad one and watch how much less RF is produced.
This thing is actually noticed inherently when operating RF tubes in old HAM radio equipment. Very used tubes have their output RF powers dropped.
I do not know if this works, but it seems reasonable.

It will not give you actual values to compare against the datasheet.
This idea will give you RF output percentage comparison, between a new tube and a used one. It will use the new tube as a reference instead of the datasheet.
To my thinking the result will be even better (practically) than measuring transconductance, because of the higher input signal (RF feedback) which shows the tube operation in RF and because of the fact that the tube is actually measured in a working circuit.

This will measure a direct percentage of the overal performance of the tube, not for each electrode alone (which is practically all that matters at the end).

The disadvantage is that one has to have a new tube type (and possibly brand?) for every old tube he wants to compare against.

How do you find this idea?