[Amps] AL-80B reversing its grid current

[R.Measures]

>November 7, 2003
>AL-80B reversing its grid current
>de VO1TV
>Our club station, VO1CRC, uses the AL-80B amp with the 3-500Z tube. Quite
>suddenly, the amp stopped transmitting and the Ig meter actually reversed
>itself and pinned on the bottom of the plastic case. I shut off the amp and
>then turned it on again. All OK... Sure enough, after a short time the amp
>stopped transmitting and the Ig meter reverse-pinned its little self.

***  The likely cause is a bent filament helix.  (see Sep.- Oct., 1990 
*QST*) When things get hot and expand, the bent filament touches the grid 
cage and a filament-to-grid  short occurs.  When the tube cools, the 
short may disappear..  //   The hot (1820ºK) tungsten filament/cathode is 
apparently bent by the perpindicular EM force which accompanies the large 
burst in cathode to grid electron flow during an intermittent osc. at c. 
155MHz.  If you have a dipmeter, you measure the freq. of the resonance 
that supports the oscillation condition by coupling the dipmeter to 
either side of the DC blocking cap.
--    [note:  this resonance is formed by the anode-C (5pF), in series 
with the anode lead L, in series with the DC-blocker cap., in series with 
the Tune-C, in series with the chassis-L.   The total L is c. 600nH and 
the net C is a bit under 5pF]
-  The feedback path that supports the oscillation condition is the 
3-500Z's anode-cathode C of 0.15pF (c. 7000-ohms of XC),   One fix is to 
reduce tube amplification c. 150MHz and/or provide a series-resonant 
lossy path to ground at the cathode-input to attenuate the150MHz feedback 
signal.
>
>Now, the amp will do the reverse-pin routine even before we transmit. The
>problem is getting worse. Because the amp stops transmitting, I conclude the
>Ig meter is accurately telling us that there is a problem. I donít think ití
>s just a metering problem.

**  I agree.

>Any idea what the problem might be?? Iíd like to get some collective wisdom
>before opening the case and fiddling with the innards...
>
**  Some people will tell you thatVHF  parasitic oscillation in a MF or 
HF amplifier is absolutely an impossibility.  I am not one of them.  Mr. 
Murphy was right.  

>I do have an old tube tester, the Heathkit IT-17. Unfortunately, the
>Heathkit tube manual I purchased in the 1980's was written in January 1978
>and does not list the 3-500Z tube, so I donít know how to test it. 

***  I test 'em initially on the bench with a high-potential tester.  A 
3-500Z with an OK vacuum and a straight filament helix will withstand at 
least 5kV between the filament and the grid (2uA point).  New tubes test 
at c. 8kV.