The project:
I have a ccfl backlight from a portable dvd player and i want to use it for a keyboard light. I am tapping into my power supply in my computer, a free 12v connector.

The Problem:
I cannot get the voltage to drop even with the resistors in line i still get 12v. The resistors are good, I tested it with a multi-meter.

Here is what ive done so far from the 12v connector there is a 330 ohm resistor which should give me near the 3.3v. I need for the digital part of the circuit, and on the other 12v line I have 100 ohm resistor to give me about 9v. The 9v is going into a transformer to provide the high voltage for the ccfl. Anyone know why im not getting a voltage drop? are my calculations wrong?
Thanks for the help.

 

Well, don't forget that if you're using a spare drive power connector, it has both 12v and 5v on it.

Why don't you start with the 5v line, and simply use three forward-biased diodes to drop about 1.8v, getting you to 3.2v?

With the 9v, you should use a regulator such as a 7809 or the like. An LM317 would work, just calculate the resistors you'd need.

I would not recommend supplying a digital circuit via a resistive divider; you need a constant voltage for the varying current the digital circuit will require.

 

Well, don't forget that if you're using a spare drive power connector, it has both 12v and 5v on it.

Why don't you start with the 5v line, and simply use three forward-biased diodes to drop about 1.8v, getting you to 3.2v?

With the 9v, you should use a regulator such as a 7809 or the like. An LM317 would work, just calculate the resistors you'd need.

I would not recommend supplying a digital circuit via a resistive divider; you need a constant voltage for the varying current the digital circuit will require.

The connector im using only has 2 12v and 2 grounds, the square one dont know what its called not molex, all others are being used. The diode idea could work i didnt think of that. I had the resistors from another project so i was just going to use those.

Also there are 2 wires for the light, unfortunately I cut the white wire and forgot where it was connected, the pink is power in and it works ok with just the pink connected but im worried it needs the white hooked up as well. Admittedly, I know very little about these ccfl other than it stands for cold cathode flourescent(spelling?) light. Any help is appreciated

 

Well, if you cut a wire and don't know where it goes - LOOK HARDER!

You should be aware that CCFLs usually require a pretty high voltage to "start" them; in the neighborhood of 500v-750v, and then a lower voltage to keep them running; perhaps 150-400v depending on the manufacturer, size of the tube, et. al.

If you have damaged the wiring to the CCFL, you had better be careful. There isn't much power in the CCFL, but it's still enough to kill you if you provide a current path through your torso - so, keep one hand in your back pocket, wear rubber-soled shoes and sit on a wooden chair when you have power applied to it.

The CCFL supply is most likely a flyback HVPS. You will need to use a voltage regulator to supply it with a large and a small output cap to suppress transients, as the CCFL supply is an inductive load.

 

all the circuitry for the light was on its own part of the pc board so I just cut it off the board so I have the original equipment transformer and digital ic's here is a pic of the setup as it is right now. Its got the xformer an opamp and a few transistors.

And yea that transformer throws out a lot of voltage it arcs to my probe when i even get close to test it.

Got any idea where that white wire goes? I know I know i should have marked it. It works without it hooked up but I wont let it run for more than a couple of secs for fear of hurting something.
I can only make it work when attached to the dvd player it came out of because of the lack of proper input voltages. I think I will get the diodes tomorrow and try that. Which one was it?

 

Well, the red wire seems to be connected to the CCFL. However, there is no return path. That means that the CCFL supply will be continuously putting out it's maximum voltage as there is no return path.

Or did you deliberately disconnect the blue wire from the CCFL?

I really hate to guess at this.

The CCFL wiring connections must be insulated using high-voltage shrink tubing. The stuff they sell as "electrician's tape" I don't trust further than I can throw it. After a period of time, the adhesive on "electrician's tape" becomes "gummy" and it unwraps, leaving exposed wiring.

The photo does not show enough real detail to be certain of anything - except that the blue wire is disconnected, and it should be connected to something.

A tip: when you first start taking things apart, use that camera of yours to take detailed pictures. That way if anything goes awry, you will have a record of how things looked before you changed anything or wires/components broke off, etc.

 

I know it was a dumb mistake when i cut the wire I intentionally left a little of the wire with sheathing on the solder joint so I would know where it came from. The camera is my girlfriends so I didnt have it at the time. Somehow I lost track of it and have been kicking myself ever since.

I figured the initial charge needed would be higher like in a typical flouro bulb and there are 2 capacitors on the pc board as you can see in the attached photos, which are much closer and more detailed.

The blue wire was soldered on by me there was previously nothing attached there. From looking at the traces it seems like this is a ground point.

I agree about the electric tape I rarely use it. I have soldered and heat shrinked all connections.

so you think the white wire from the light is a return, I was thinking the same thing but I dont know where to connect to.

I work as electronics tech but mostly on older tube amps so some of the newer solid state stuff escapes me, like the small transformer. Im used to the big heavy ones.

What model number diode where you suggesting I use in your first response?

Maybe these pics will help you to see what is going on if you still cant see i have larger versions of the picture but I had to shrink them down to get the forum to host them. Thanks for your help so far.

 

Speaking of tape, it looks like that transformer is insulated using Kapton tape. Kapton tape is a great insulator, but if it gets damaged at all, you can wind up having a fire. The US Navy forbade the use of Kapton insulation in any aircraft wiring since the late 1980's. It was discovered that Kapton tends to revert to the shape it was originally moulded in. If nicked or scratched, it will tend to tear apart and curl up. If subjected to high temps, it will carbonize and act as a conductor. The next time the circuit is energized after that happens, a fire is a very likely result.

If the red lead is indeed connected to the HV output (which seems logical, since the CCFL does light up) then it needs a return path as well. Try connecting the blue lead to the cut-off white lead. Polarity to CCFL tubes doesn't usually matter.

For the diodes, you can use your basic garden variety silicon rectifiers. 1N4148 will work for low current. 1N4001-1N4007 will work for somewhat higher currents. If you wish to be able to "fine-tune" the voltage level, you could use some Schottkey diodes in the mix, like a 1N5817. At 100mA, a 1N5817 has a Vf of 0.34v, increasing to 0.45v @1A
Vf of a 1N4001-1N4007 also varies with current.
At 50mA, it's roughly 0.7v,
at 100mA, it's roughly 0.75v,
at 200mA it's roughly 0.8v. You'll have to measure yours to determine the actual Vf, as diodes vary somewhat.

 

wow did not know that about the kapton tape good to know.

I should have said this before but I tried connecting it and it wont work like that, I got a nice arc though. Is it going to hurt the light if I run it at full power? possible shorter lifespan I would think. Thanks again you have been a big help.

 

I don't know how it's even lighting in the first place - except by stray voltage, perhaps.

Basically, any circuit requires a full path - but you knew that already.

Similar to what I've already said - CCFL's require a high starting voltage to light off (ionize) and then reduced voltage to keep the ionization going and glowing and to keep the current down. Wired the way it is, there is no obvious return path, so I'm not sure how it's even working - unless the flyback supply is generating far above it's originally-designed voltage - and the supply won't last long like that.

It's likely that something which controls the voltage or current has gotten fried. Trace out the circuit, make a schematic, and figure out what all the SMD critters are.

Here's a page worth it's weight in gold for identifying SMD's:
http://www.educypedia.be/electronics/datacomponent.htm

 

good point ill make the schematic and see what I can figure out from there. Thanks again