Tube Tester Question (TV7D/U)

[k4kyv]

My mutual conductance tester TV7D/U has a gas test function.  The way the manual explains its function is that the tube is set up with a fixed plate voltage, and when you press the Gas 1 button, a negative bias is applied to the grid, and you manually use the bias pot to adjust this voltage until the plate current deflects the meter to the first large division (10 on the 0-120 scale).  Then, while holding down the Gas 1 button, you simultaneously press the Gas 2 button.  This inserts 80,000Ω in series with the grid. If the tube is gassy, it will draw a small amount of grid current which will cause the bias voltage to decrease, due to voltage drop in the series resistor.  This reduction in negative bias voltage causes the plate current to increase.  The manual recommends that a tube be discarded as gassy if the plate current increases by more than 10%, that is to say, if the meter increases more than one small division.

My question is, what if the pointer deflects downwards when the Gas 2 button is pressed? I see it deflect downwards far more often than upwards, and with some  tubes the decrease is several small divisions.  If the tube is free of gas, shouldn't it remain still?  Nothing in the manual mentions the possibility that the plate current might decrease when the resistor is inserted in the grid circuit.

There is no such thing as "negative gas" in a tube, but it would seem that if inserting resistance into the grid circuit causes the plate current to change in either direction, this could be a problem, since it is very common to have 80k or more resistance in the grid circuit of a tube type amplifier, particularly with R-C coupling.  What could be the cause of a decrease in plate current, other than a small contact potential bias, which should be thoroughly swamped out by the external fixed bias in this test circuit? 

I am aware that sometimes a triode amplifier that is designed to handle a very low signal level, such as a microphone pre-amp, may have no cathode resistor, and depends on contact potential to generate a fraction of a volt negative bias to operate the tube.  But in these circuit, there is usually several megohms of grid resistance. I always thought of this as a rather flaky way to generate the operating bias for a tube, but I am not even sure this has anything to do with the above described phenomenon.

[Detroit47]

Don

I don't know why it happens but my tv 7 does that on some tubes also. I have noticed it espically on pentodes.

N8QPC

[KM1H]

Has the tester been recapped, resistors checked and recalibrated?

I dont remember that problem on my TV-7C and it certainly doesnt happen on the regularly used 752A

Carl

[ke7trp]

If you find the tester is out of spec.  Dan Nelson is the man to see.  We have two Dan nelson calibrated TV7s.  He does a wonderfull job and has replacement parts and cables.  My Dan nelson unit does not go downward on Gas test.

C

[k4kyv]

I wonder if it could be the tester or the tube.  With some tubes it behaves normally.

I have an older version of the tester, probably of the BC-610 era.  Army olive drab wrinkle finish and no accommodation for miniature tubes; the transconductance meter is calibrated directly in micromhos, not the arbitrary scale of the TV-7.  I think I'll  compare some readings between the two units.

The TV-7 was crapped out when I got it, and I repaired and recalibrated it per the manual, but that was probably 20 years ago.  I can't see how a calibration problem could make the meter read backwards, but there may be a clue in the schematic.

[sndtubes]

Don,

When testing for Gas, are you keeping the bias setting the same or are you running the bias more negative so that the meter only reads about 10 before pressing Gas 2?  That step is sometimes overlooked.  If you leave the bias at the same point as where you tested the tube, there is too much plate current flowing to properly test the tube for gas current.

73,
Mike WB0SND

[k4kyv]

I run the bias to a reading of 10, per instructions in the book.

Two things can happen when there is resistance in the grid circuit of a tube.  Contact potential bias makes the grid go more negative. This is caused by a build-up of negative charge on the grid from the electron cloud that surrounds the hot cathode. This bias would cause the plate current to decrease but the phenomenon normally occurs with low plate current plus megohms, not kilohms of grid resistance.  The other, opposite effect is when positive ions of gas in the tube strike the grid, and the positive charge leaks back to ground through the grid resistor, reducing the negative bias on the tube and causing the plate current to increase. That is how the gas test works.

Maybe I  need to check the 80K resistor in the tube tester to make sure it hasn't opened up or drastically increased in value.

Some time ago I checked the 1st 12AX7 stage in the section of my speech amp used for the D-104, after I had increased the grid resistor to 10 megohms.  With the tube that was already in place, there was a slight but measurable negative voltage on the grid with respect to ground, 0.173 volts IIRC.  The stage is cathode biased, and I noted that the drop across the cathode resistor was 0.9 volts.  I tried another tube and re-took the measurements.  The new tube showed no negative voltage on the grid with respect to ground.  The cathode resistor voltage drop and voltage reading at the plate were exactly the same as with the first tube, so the extra negative grid potential with the first tube didn't appear to affect plate current at all. The plate current calculated out to be about a half milliamp. I don't remember whether or not I ran a gas test with those two tubes.

[sndtubes]

Don,

It sounds like you have a very deep understanding of how tubes work.  Better than I do and that's embarassing because I'm in the tube business.  I just know that many people overlook the step of increasing the bias before making the gas test. 

The other issue here may be with how Hickok testers test tubes.  They use the line voltage for the grid signal and in many cases, especially with the older designs like the TV7, they use way too high of a grid signal voltage on the tube (5 VAC).  This could have an effect on how the Gm behaves when making the gas test.  Hi Gm tubes do not test properly on the TV7 because of this huge grid signal.

Also, with my experience with Hickok testers, they check tubes at a fairly low plate / screen voltage.  Because of the low B+, not enough of the gas in the tube will ionize resulting in the gas test on Hickok's not being very sensitive.  Testers such as the Triplett 3444/A test at a far higher voltage which results in it being able to identify a gassy tube where a Hickok does not. 

[KM1H]

Last month I overhauled a NC-98 for a customer and while replacing the couplate and checking resistors in the 12AX7 1st audio I found a wiring error.  There was supposed to me a 10 meg from grid to ground. Instead it was connected grid to plate!  Worked fine.

Carl