CCFL mod

W

WhiskeyBreath

n00b
Joined
Oct 26, 2004
Messages
2
How would I go about wiring my CC's so they flash along with the hd led? And furthermore would it be possible to make a three-way switch that can turn the lights on, off, and to the flashing mode?
 
this is one way you could do it.

ccathodethin2g.gif


HDD LED is meant to point to the positive terminal of the HDD led thing on your mobo. Not the ground otherwise death!


WhiskeyBreath said:
How would I go about wiring my CC's so they flash along with the hd led? And furthermore would it be possible to make a three-way switch that can turn the lights on, off, and to the flashing mode?
Click to expand...
 
that circuit won't work unless the HDD LED puts out 12V... :D

The HDD LED implementation on motherboards varies a lot based on manufacturer; some manufacturers directly drive the HDD+ line through a resistor, others have +5V connected to HDD+ through a resistor and drive HDD- through an open collector output. So it's hard to make a single transistor circuit that works on all boards...

For safety's sake, I'd drive a 4N28 or similar transistor-output optoisolator directly with the HDD header, and connect it so that its output pulls the base of a bigger NPN transistor (TIP29A or something) to +5V or +12V with an appropriately sized resistor. Ground the big NPN's emitter and connect its collector to the - of your cold cathode. + on the CC goes to +12V. I can sketch it later.
 
thats why i have the transistor.....or do i have it diagrammed improperly? The HDD led is supposed to switch the transistor on and allow the 12v to flow to the CC

gee said:
that circuit won't work unless the HDD LED puts out 12V... :D

The HDD LED implementation on motherboards varies a lot based on manufacturer; some manufacturers directly drive the HDD+ line through a resistor, others have +5V connected to HDD+ through a resistor and drive HDD- through an open collector output. So it's hard to make a single transistor circuit that works on all boards...

For safety's sake, I'd drive a 4N28 or similar transistor-output optoisolator directly with the HDD header, and connect it so that its output pulls the base of a bigger NPN transistor (TIP29A or something) to +5V or +12V with an appropriately sized resistor. Ground the big NPN's emitter and connect its collector to the - of your cold cathode. + on the CC goes to +12V. I can sketch it later.
Click to expand...
 
Whatsisname said:
thats why i have the transistor.....or do i have it diagrammed improperly? The HDD led is supposed to switch the transistor on and allow the 12v to flow to the CC
Click to expand...
The NPN right now is configured as an emitter follower. If the HD LED is off, the transistor will be off and won't conduct. When the HD LED turns on, the CCFL voltage will be the HD LED voltage minus whatever the Vbe of the transistor is.

If the NPN had a grounded emitter and its collector pulled the - side of the CCFL to ground, that might work. But as mentioned before, the implementation of the HD LED on motherboards varies from board to board.
 
wouldn't it be easier to wire in a relay vs. the transistor? the hd LED should be putting out 5volts, which would trip, and allow the CC to be running from the main 12V line. No?

-Jeff
 
The problem with relays is they require more current to turn on than a transistor does usually, plus they are usually bigger and more expensive.
 
what matt said

Also, you'd hear the relay making a terrible CLICKITACLICKITACLICKITACLICKITA sound non-stop during any hard drive access, and the relay probably won't last all that long.
 
Hmm, from my limited experience, I would think to use a FET, what's everyone's take on that?
 
xonik said:
Hmm, from my limited experience, I would think to use a FET, what's everyone's take on that?
Click to expand...
If the HDD output was a totem pole output (can both source/sink current), then connecting it to the gate of a logic level FET would work wonderfully... ground the source, connect the drain of the FET to the CC's - and the CC's + to +12V.

But if the HDD output isn't totem pole, you might end up with the FET being stuck on or off.
 
gee said:
that circuit won't work unless the HDD LED puts out 12V... :D

The HDD LED implementation on motherboards varies a lot based on manufacturer; some manufacturers directly drive the HDD+ line through a resistor, others have +5V connected to HDD+ through a resistor and drive HDD- through an open collector output. So it's hard to make a single transistor circuit that works on all boards...
Click to expand...
The IDE spec says pin#39 shall be an open-collector NPN with a 10k pull-up resistor, with HD activity the transistor Q1 is switched on, the pin goes low. I've no idea on SATA, etc.

Between each IDE mobo socket and the HDDA header is a steering diode (D1 & D2), so activity from either channel switches the single led. Last 3 boards I've checked have been like this:

pin39.gif


R2 varies, 2 boards (TMC AI5VG, MSI Turbo) had a 330R, Asus A7N8X-E has 220R.

Safe (and universal) way to interface is with an optoisolator in place of the activity LED, another way is a PNP transistor switch with base resistor to the LED's cathode pin (bottom one). With an NPN it's more tricky, D1/Q1 drop more than the switch-on voltage.
 
You must log in or register to reply here.
Back
Top