Ok, I’m currently installing 6 cold cathode tubes in my car and I’ve hit unexplained difficulties. I had to extend the wires between the inverter and the tubes for two of them by about 2 feet. Just tested them and the two tubes on extended wires glow very dimly, and only about half the length illuminates atall. I tried various combinations of inverters and tubes and in desperation removed the wire extensions and resoldered the wires to their original length This solved the problem immeadiately.
So, why is two feet of reasonable quality wire, rated way above the orginal wire used, causing such a voltage drop? Is it just that the voltages are so small that the slightest resistance will soak up most of the power? I could reposition the inverters so the wires don’t need to be extended, but I’d rather not.
I don’t know what a cold cathode is. If the signal you are sending to this device is a high frequency ac, sometimes line length may be important. A shot in the dark really, yet you may have a standing wave causing high impedance. I am just taking a guess. What is a cold cathode?
Essentially a neon tube, a light source which doesn’t rely on thermionic emission. That’s a good point about the ac wave effect in a transmission line, unfortunately I have no idea what frequency is in use, else i could possibly tune the line length.
thats weird, i extended the wires on some cheap cold cathodes i used and it was fine. ALthough i didnt go as far as you did. only about 1foot. Maybe try extending the length of the wire before the inverter and not after?
It is almost surely to work. As an experimenter, trying a different length of line is more appealing. I think it is likely you have hit at an odd resonant point, it would be fun if just a longer piece of wire solved your problem. I confess ignorance, I am just curious how this will work.
Is there a chance that the wire going to the neon is shielded (coax), that is a sign that normal wire may not work so well.
Yes of course extending the wire before the inverter will work, for various reasons there is about 16 extra feet of cable between the inveter and the battery, unfortunately this is useless as the point of the exercise was to be able to hide the inverter away and place the tubes some way away from it. I will try a couple of different lengths tomorrow to see if it has any effect. The cable certainly isn’t co-ax but might possibly have some other sort of shielding. While the metal core of the original wires is thinner than the extension wire I am using, the insulation is thicker I think. I shall experiment and report back.
The basic rule is…Don’t extend the wires between the transformer and the tubes. I don’t know why exactly. You can change the length between the power source and transformer though. I ran into the same issue when I was putting cathodes in my xB. You just have to be able to install the transformer somewhere close. I have somewhere between 20 and 25 in my car.
Very wierd, I’m just going to have to shorten them again and resite the inverter. I’d still love to know why they can’t be extended, just for my own interest as I’m an engineer.
To a cold cathode FAQ. It mentions that extending the wire is undesirable, but only sorta infers that the reason is voltage drop. Apparently the white wires from the inverter are 650 v ac.
Cold cathode fluorescent lamps (CCFL’s) require a high voltage to strike an arc in a low pressure gas path. High voltage, low current supplies are generally the high frequency switching type (10’s of kHz, usually about 40 kHz) which can be made compact, lightweight, and efficient. At that frequency the wavelength is enormous (of order 5 km) and the connecting wires exhibit no transmission line effects. The frequency is high enough, however, for varying length wire to exhibit bulk impedance effects. Longer lengths produce larger series inductances and larger shunt capacitances. The wire lengths are optimized for their bulk impedance characteristics to deliver maximum power to a matched load. This is important for CCFL’s where the driving current is only a few mA. At the given wavelength increasing the wire length only increases power loss. There are no Vmax or Imax points at 1/4 or 1/2 wavelength intervals as resistive losses have become dominant in the many kms of wire. The wire will also be dispersive at that length if it is not high quality coax.