Q VR tube regulator in Gary Pimm designs

[zigzagflux]

I have always used a 100 ohm resistor, both for the above reason, and also a very simple way of measuring shunt current. 0.068uF cap has never been problematic; I could try increasing to 0.1 but never found a reason to. Noise has been effectively immeasurable w.r.t. the output signal, and regulation characteristic has minimal requirements when used in PP amp. It simply helps set the operating point of the rest of the circuitry. Wouldn't use it in a SET.

[anatech]

Hi GoatGuy,
Yes, a VR tube is like a zener diode. You actually do want to run some current through them and they tend to last a long, long time.

If you get the right capacitance across the VR tube, you can get it to strike, extinguish and strike again ... on and on. Probably rough on the tube. A CCS feeding the VR tube really expands the trouble area with capacitors since there isn't any resistance to damp anything from the charge side.

If you want to experiment, grab an NE-2 neon lamp (it's just as much a VR tube as the 5651 and others mentioned). Make sure your DC supply is above the strike voltage (~90 V) and use a resistor in series. Then place various capacitors in parallel and see what happens. I used to do this as a kid.

The biggest difference between an NE-2 and VR tube is that the NE-2 runs at much lower current and isn't shielded against ambient light (photons reduce the strike voltage). Aside from the uncontrolled dimensions and element placement, it is the same. Probably not the best idea to use one that way (as a VR tube). I may be wrong, but I think the minimum recommended current is around 5 mA and up to 25 mA or so for most VR tubes.

-Chris

[PRR]

The gas tube (or a Zener) isn't about a near-constant current. If it did that, we would not need it. It takes the -difference- between power source current and load current. Either or both may vary (our "problem"). The gas/Zener will take more or less current, trying to keep its Voltage constant.

The NE-2 is a reasonably useful reference. It is NOT tightly specified so you don't know quite what voltage you will get. A "jumpy" plasma does not hurt the light function but may give erratic regulation. And they are light sensitive, but we have duct-tape.

All the plasma "arc" processes WILL oscillate with some value of parallel capacitor, supply voltage, supply impedance. Negative Resistance isn't an easy date but an oscillator waiting for provocation.

NE-2s were reliable enough as semi-constant sources and as oscillators that H-P used one (or two!) as the built-in calibration source for some of their oscilloscopes.

I had never heard of a plasma being "inductive". If you plot an inductance on that same graph, it rises with frequency, but much less than a pure inductance. This makes sense. It is not "loose" charges sloshing around, but relatively heavy and lossy ions limiting the slosh.