The theory in that link is pretty far-fetched. The screen is an excellent shield for RF between the anode and grid. If the screen wasn't a nearly perfect RF shield, we would never be able to use the tetrode in a grid driven amp.
It's silly to think the screen shield the control grid for months or years, and RF feedback through the screen to the control grid is so low the amp is stable, and then all of a sudden some "mysterious oscillation" makes the screen's shielding properties vanish for 2 seconds and allow dozens of watts back into a radio! The most common fault that wipes things like radios out, and I have seen this dozens of times, are gassy tubes. If a tube arcs internally the grids take the brunt of the fault currents. This is a repeatable common problem that can be measured and diagnosed, and requires no "magical" screen shielding loss. What actually happens is the tube gasses, peak anode voltage exceeds the breakdown, and the gas ionizes and causes the anode to fault to the screen and control grids. This is incredibly common in tetrodes. In grounded grid amps, when something in the tube outgases or overheats and releases vapors, the arc is from the anode to the grounded grid. The cathode normally gets very little of the fault current, so the exciter is safe. In a tetrode, especially a minimal parts design that uses and untuned resistor loaded grid, the fault current that would normally be shunted to ground blows right back through to the exciter. I designed some plasma generators that used 4CX1500B's in the 70's or 80's that had this problem when the load went high SWR before plasma ignition. The peak anode voltage would go to 10 kV or more, and if the tube was not exceptionally healthy it would arc internally. The screen would take most of the fault, but it would pull up (watched it on a scope) to 3-4 kV. The control grid would pull up to 1-2 kV, and it would blow the transistors in the IPA stage. Anything you throw in line between the PA and the exciter will help divert some of the pulse energy when the tube faults, but I sure would not bet my K3 on a low pass filter rated at high power for reliably stopping a dc pulse. My suggestion for people not wanting to rely on snake oil and myths, and to actually protect the exciter, would be to use a monoband filter for each band that has low voltage caps, spark gaps, or voltage limiting gas tubes shunting the RF lines. The low pass should have a good dc path the chassis. A minimal inductance shunt coil with a resonating capacitor across it would be the best protection, and get inside the Acom and be sure it has an anode fault current limiting resistor of proper size (about 20-30 ohms) and design (high voltage type resistor) and be sure the screen has a solid high-current fault clamping circuit. This is not a new problem with grid driven tetrodes, it is really pretty common. I'd guess I've seen this happen about 100-200 times in the past 30 years, including the fits I had with the 4CX1500B plasma generator I designed. The liability of blowing up radios from common anode-to-grid faults is one of the reasons I never built a grid driven amp with an external driver stage. By the time the grid and screen are done right to protect the radio, the amp would cost more than a grounded grid amp. 73 Tom > Jens, you are right: K3 + ACOM 1000 = excellent combination. > > However, I highly recommend to anyone using a tetrode amp with a "swamped" > (i.e. untuned) input inserting of a lowpass filter BETWEEN the TX and the > amp. See for instance: > www.kk5dr.com/swampednetworks.htm ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html
