Appropriate Wire For A Deuterium Bulb, 1kV 250kHz 3W Heraeus D2 Bulb

[WiegerTheFarmer]

I've been struggling with this problem for several weeks now and am getting no where. But I'm not giving up, because there's always a solution to every problem...at least, that's how it's always been for me.

I've been tasked with trying to run a Heraeus Fiberlight Deuterium Lamp "remotely". That is, off the circuit board it comes with. Here's the specs on the system. FiberLight Specs

The lamp ignites at 1kV 250kHz AC and uses 3W.

I've had extremely low success getting it working off the board. I started with standard 22 gauge solid core hookup wire. And I got it working at a wire length of about 8 inches. But nothing more. My ideal distance is about 3 feet. But, even at 8 inches, the intensity of the bulb is lower.

Nothing else can I get working. I've tried higher gauge, shielded, coax.... Nothing. Now. I'm not a technician, and although I'm an electrical engineer, I specialized in the computer side of things, not the power side. And my classes in high power conductors has long vanished.

The lamp still works (if I put it on the board it works great). And the tungsten bulb works fine at any distance with any type of wire. Just the D2.

So. What should I do? What wire/conductor should I get? Is there resources I can read? The manufacter has no idea (they've never done this off board before). Any help would be amazing. I have the manual for the system, but it doesn't really give any more details.

Thanks!!!

Aaron Visser

[Skeith]

you say the lamp ignites at 1kv, once its struck what is the operating voltage?


EDIT.
I just skimmed the link your supplied. These lamps are RF excited so it is a bit more than voltage and power at this point. By lengthening the leads, you may be introducing an impedance miss-match into the system.

its like a radio transmitter, introduction of a miss match will introduce RF losses in the transmission line resulting in a lower output power at the load end than at the feed point.

Why does the lamp need to be mounted remotely. I thought the light was designed to be routed to its desired location with fiber?

[WiegerTheFarmer]

Not sure what the operating voltage is... If I get my hands on a scope I'll measure it.

Why does the bulb have to be remote? We are making a product that needs the bulb in it (to prove a product design). I don't have room for the board, and if successful we'll miniturize the board or....use UV LEDS. Actually at this point I'm thinking of recommending we just pursue the LEDs.

[Sch3mat1c]

Probably not impedance mismatch per se that's doing it; 3' isn't anything, electrically speaking, at 250kHz. I think it's most likely the capacitance of the run did it, at least if it's not starting. Extra capacitance will make the starting circuit work harder. Once running, the arc has a very low dynamic impedance, and putting inductance in series increases it; this could dim the output (apparently!) or confuse the circuit.

That said, 3W isn't much, and if the bulb drops, say, 100V, 3W is only 33mA, so the average impedance isn't very high, and (putting things back in impedance matching terms..) the impedance of the cable shouldn't add much to it.

The product as shown drive a fiber... isn't that compact enough?

Is the intense wideband spectrum of a deuterium lamp really necessary? Would a (shielded or un-) mercury lamp be a cheaper UV-C source? If you can use LEDs, you don't need any of this dangerous UV at all.

Tim

[CWB]

i agree with tim ...
it may be the capacitance of the line that is killing it .
what type of cable/wires are you using and construction methods ?