Funderberg gold sputter test:

See Figure 24. The powdery looking material on the paper is gold meltballs.

Use a high-pot tester to apply 2x the rated anode potential. The grid is at ground potential. If the positive leakage current is much higher than the negative leakage current, gold-spluttering is indicated. To reduce gold meltball leakage between the anode and the grid, hold the tube upright and strike the top of the plate cap vertically with the flat side of a 4oz ball peen hammer. Tapping will dislodge gold meltballs, which will then drop into the tube base. The leakage current should decrease. When measured leakage current stops decreasing, stop tapping. Do not invert tube or the process will have to be repeated.

1, 8877 Notes (from parasitic suppressor retrofit kit)

8877s can be damaged by gold sputtering from the grid during an intermittent VHF/UHF parasitic oscillation. see:

http://www.somis.org/8877-gs2.JPEG

and

http://www.somis.org/Goldmeltballs.html

The main problem with loose gold meltballs is that they can cause anode to (grounded) grid flashovers - especially when a signal is present. Thus, decreasing the VHF-Q of the parasitic suppressor may not restore your kaput 8877 amplifier to good working order.

One problem related to anode-grid flashovers is a burned out heater. This comes about because when the HV-pos. at the anode arcs through the field of loose gold to the grounded grid, the anode goes to virtually ground potential and the cathode tries to spike at negative several kV. If the cathode of the 8877 is not connected to one side of the heater, a simultaneous arc can occur between the heater and the cathode -- which can burn open the heater wire. To avoid this potential problem, use a bifilar heater RF choke, connect 3, 3A diodes in series, and connect them from the cold end of the cathode's RFC to ground. . (cathode-end of diode string to 8877 cathode), wire the cathode to one side of the heater, and make sure that neither side of the heater supply is grounded. This prevents the normally positive cathode V from spiking to more than several V negative during a flashover. It is also a good idea to limit peak discharge-I from the filter C during a flashover. This can be accomplished by inserting a suitable "glitch" R is series with the positive HV lead feeding the bottom of the HV-RFC. A suitable glitch R is 2, 10-ohm, 10w glass coated (vitreous enamel) resistors in series. A Kanthal-Globar Type SP 886A 30w, 20 - 30 ohm resistor would also be suitable. (Kanthal-Globar order desk - 716 286 7610).

http://www.globar.com/ec/tubular.php.html

-- Rich, AG6k, 805-386-3734

end