Dual 12" Cold UV Cathode

I wanted to make the +/- wires to the inverter longer on one of the Dual 12" Cold UV Cathode that I just bought. after soldering everything with longer wires and put it back into the case I noticed that it wasn't as bright as the other one. I used 22g wire. I was thinking that it need to be inslated better so I used some thick speaker wire that I had and the same thing. then I just put it all back together thinking that I would just deal with what I got. it's still not as bright as the untouched one. Is the one i missed with dead?
I'm still new to all this modding stuff.
any tricks out there to help me?

thanx

Are you changing the 12v wire which powers the inverter, or the high voltage wires to the light itself?

If it's the 12v wires and it got dimmer, I'd suspect a bad solder joint, bad connection to molex pins, or when moving stuff around you damaged the tube, or while soldering you may have damaged a component. 22 guage should have a low enough resistance that length won't really be a factor for lengths you will use inside a computer case.

If you did it to the high voltage wire, then all the same can apply, plus the possibility of leakage of current through the insulation (a problem which gets worse with longer wires), since ordinary wire is often not sufficiently insulated for ccfls.

The fact that it's still dimmer after putting everything back to original condition suggests to me that it's most probably a bad solder joint, or you damaged something.

dont try to run more lamps off of an inverter then its built for. buy seprate kits... you can get 2 CCFL lamps and and inverter for $1.99 (if you know where to look) or less then $10 at any computer store. this will come with everything necessary to hook up the 2 lamps.

let me elaborate. the high voltage needed to run CC's is not run on ordinary wire. the whole harness assembly is not meant to be tampered with. if you do, chances are you will not insulate it well enough and there will be current leakage... like the guy that wrote this:
http://www.casemodgod.com/ccfl_inverter_rant.htm
conveniently left out that he had tried to extend the high tension wires to the inverter.

right out of the box, everything will click in. you can mod all you want on the 12v side, but do not mess with the lamp-inverter wires. each lamp will have its own connector on the inverter. never wire more then one lamp per connector.

if you have done this, then you might as well throw out your inverter. it will never work at 100%

I'm changing the wires from the inverter to the tube. like I said it's a dual and I only messed with one. the other one is working fine. I think that is was my soldering. I'm gonna try resoldering the wires together. Is there some better well installed wire that I should use instead of the 22 gauge that I've got now?

this is my first project ...

I would recomend wire rated for at least 1000v between the inverter and tube. Offhand, ebay may have some. I know I got some 15kv wire there but that's pretty thick and stiff, which you probably don't want.

You can’t say I didn’t warn you… Cold Cathode inverters are built as cheaply as possible, and any voltage leak or high resistance path will cause it to die: The question is if it will simply stop working or if it will catch on fire.

So for the records, my position is: buy a new one, and throw this one out.

But if you must continue, here are some tips:

A CCFL lamp runs between 500-700vac. Find the lowest resistance wire rated to this voltage that you can. Take into account the impedance of your wire, and keep them as short as possible. I would not use a length longer that that which it came with. Get high voltage heat shrink and use it over every connection. Do not use Radio Shack quality heat shrink, or any electrical tape. And finally, make sure that all connections are neatly soldered.

Please be safe and use common sense!

In my experience, the brightness problem is usually/partially to do with how the cables are laid, not the resistance. If they are laid close to each other, or sleeved together, you are very likely to get loss of brightness, or loss of light entirely. See my post in this thread for a more detailed explanation

Harlequin said:

You're getting too much capacitance between the wires. Separate the wires as far apart from each other as you can. If you have the tube turned on while you do this, you will notice the tube going on and off and the brightness changing as you move the wires around. If you put each wire in it's own aquarium tube, and route them seperately, trying to keep them 3 inches apart, you'll be able to use significantly longer wires than normal.

CCFT work on oscillating power at high voltage, and putting a capacitor on the wires would buffer the oscillations, lowering the voltage peaks below what is needed to create the light. By putting wires close together, you are in effect making a simple capacitor, and by making the wires too long, the capacitance of that capacitor is increased until it is problematic.

I believe the capacitance problem is also exacerbated by routing the wires close to a metal plate, especially when the metal is connected to metal close to the other wire, such as if you routed the wires along the metal insides of your case. Put lots of air around them. Note that even trying to minimise capacitance like there, there will be a length limit, it's just greater than what you could get away with otherwise.

Click to expand...

Actually, your wrong.

The problem is not capacitance, it is inductance. Inductance is energy is stored in the form of an electromagnetic field which has the property of opposing any change in current. Capacitors, however, only pass high frequency alternating current. If you do not understand how this works, go get an electronics textbook and study up. This principle is used in audio crossovers, and filters of just about any type.

Any straight piece of wire exhibits inductance (probably a fraction of a milli-henrie) although not of any major significance until VHF frequencies. The thing is that the inverter outputs VHF.

The reason that the longer wires dim the lamp is more inductance over the length of the run. So by separating said wires, you can decrease the total uH, and allow them to pass current the HF AC to the bulb over a longer distance. A more effective solution would be to shield one of the wires and ground the shield. But aquarium tubing…works…

If you don’t believe me, take a ferrite core out of a radio and wrap your CCFL wires around it a few times.

So -- separating the wires will allow you to make longer runs, but I discourage it because most people will never use a proper dialectic. Taking into account the shoddy construction of most inverters, I try to dissuade people from modding these things.

And next time, do your homework before you post ^^ like that

//theshadow

Theshado27:
I just did some homework, since my original assessment was my own, but it looks like if I'm wrong, I stand in excellant company alongside the engineers that build CCFL units, (among others).

http://www.ergpower.com/pdf30/confer.pdf
http://www.spwg.org/spwg_newsletter_october_2004.pdf
http://www.microsemi.com/micnotes/1401.pdf
http://www.dansdata.com/danletters127.htm

A capacitor tries to maintain voltage much like how an inductor tries to maintain current. CCFL cables need to carry a high voltage, and so capacitance between them limits their length. As you note, capacitors are used in audio crossovers and filters, but you seem to have missed that the cable capacitance in this situation would create something akin to a lowpass filter, not a highpass filter. (And regarding the original question, the filter is made more effective by any added cable resistance, so avoiding capacitance doesn't mean crappy wire is suddenly ok, it's still got to be reasonable )

Perhap you were arguing that inductance also limits their length and is the greater limitation. If so, I would agree that inductance is certainly relevant, but from reading around, it seems the main limiting factor in cable length is capacitance.
The advice remains sound.

An inverter schematic off one of the links that you posted:

hmm… i wonder, why these are in there? I bet they want to make cable length shorter

edit: look what else i found. again from your PDF:

not to be picky, but nowhere in http://www.ergpower.com/pdf30/confer.pdf does it say that excess capacitance causes malfunction... just mismatched impedance or resistance.

theshadow27 said:

A CCFL lamp runs between 500-700vac. Find the lowest resistance wire rated to this voltage that you can. Take into account the impedance of your wire, and keep them as short as possible. I would not use a length longer that that which it came with. Get high voltage heat shrink and use it over every connection. Do not use Radio Shack quality heat shrink, or any electrical tape. And finally, make sure that all connections are neatly soldered.

Click to expand...

did i not say all of that before i saw your pdf?

EDIT2: on re-reading the first artical...

it is parallal capacitence they are talking about here, not series. that is why i was confused when you said that capacitors try to maintain voltage... as no series capacitor would do... capacitence between (keyword) the two wires is what i missed.

learn somethin new every day

i shall look thrugh the others when i get a chance

Like I said, you seem to have missed the difference between lowpass and highpass - the caps in inverters are in series with the cables, NOT across them. All of my inverters have these caps, I'm well aware of them.

Also, your note that confer.pdf does not warn against capacitence is flat out wrong. Relevant quotes from that document include "Any stray capacitance produced by the inverter-to-display wiring can reduce the voltage to the tube(s), and reduce the life and brightness of the tube(s).". That's pretty clear.

As I also agreed, inductance is certainly relevant, but the primary problem when lengthening cables is capacitance.

The capacitors circled are for current limiting much like the resistor used with an LED. When gas inside the tube ionises it gets a negative resistance, that it it draws more and more current until it destroyes itself or the transformer. The capacitor prevents this. Having two ensures the tubes share the current. The reason for the loss of brightness is capacitance betweent the leads.

OP: Try using coax cable, the same stuff as used for TV aerials. I haven't tried it but they are designed for high voltage and high frequency so may work.

Harlequin said:

As I also agreed, inductance is certainly relevant, but the primary problem when lengthening cables is capacitance.

Click to expand...

I kinda agree with both of you...kinda. I believe capacitance is playing an issue, just not in the manner Harlequin is describing. The act of lengthening the wires is not necessarily resulting in higher stray capacitance...its the fact that the lengthened wire has shitty shielding on it. This shitty shield/insulation/whatever-you-wanna-call-it is essentially creating a better dielectric between the wires, thus making it easier for charge to jump between them. It literally is like wiring a cap in parallel.

The physics behind the problem should only be important to the people designing the devices. Fact of the matter remains...modifying high voltage lines for CCFLs is dangerous and should be avoided.

joecool234 said:

the fact that the lengthened wire has shitty shielding on it. This shitty shield/insulation/whatever-you-wanna-call-it is essentially creating a better dielectric between the wires, thus making it easier for charge to jump between them. It literally is like wiring a cap in parallel.

The physics behind the problem should only be important to the people designing the devices. Fact of the matter remains...modifying high voltage lines for CCFLs is dangerous and should be avoided.

Click to expand...

Exactly! very well put... i think this is the real issue, inductance and cable length are just theory.

Well, let me jump in with my $.02. One of my hobbies is designing speakers, so I'm quite familiar with analog passive filter circiutry, specifically as it applies to speaker crossovers, but generally as it applies to electronics. What we're dealing with here is impedance, not inductance. Impedance is resistance that varies with frequency. In the diagram above, the series caps are used to block DC voltage, preventing overheating of the transformer and wasted power. They will act as a highpass filter (filter that allows high frequency signals through), but the resistance/capacitence is such that it should let through just about any AC frequency above a few Hertz.

Putting the wires closer together increases the capacitence between them, but not necessisarily the inductance (unshielded twisted pair is a good example, the series inductance is negligable, but the capacitence is measurable due to minimal separation and use of a good diaelectric- Teflon). What happens is that as the capacitence increases, the characteristic frequency at which the filter operates is reduced. Characteristic frequency is the frequency at which the signal is reduced to half original strength (-3 dB). Because the capacitor is in parallel with the load, it becomes a low pass filter - allowing low frequencies to pass, but attenuating higher frequencies. Basically the capacitor provides a path for the high frequency voltage to "short out" the circuit load. In the stock design, the capacitence is low and the characteristic frequency is well above the operating frequency of the inverter. As the capacitence increases, the characteristic frequency is reduced, eventually moving into the frequency range that the inverter operates in. This causes the signal to be attenuated and the brightness to be reduced. The more the capacitence, the lower the frequency, and the more the frequencies beyond the characteristic frequency are reduced.