Connecting two CCFL lamps in parallel

Discussion in 'General Electronics Chat' started by timescope, Apr 1, 2013.

  1. timescope

    Thread Starter Member

    Dec 14, 2011
    Hi Everyone,

    After hours of troubleshooting the backlight inverter of an LG 42" LCD tv model 42LK450, I have finally found the problem to be a defective CCFL lamp after removing the fragile liquid crystal glass screen.

    Replacements are not sold here so I obtained two 42" broken screens but these use 20 thinner lamps (about 3.3mm diameter) and the LG uses 10 lamps 4mm diameter.

    The difference in characteristics of the thinner lamp is detected by the inverter circuit and it shuts down.

    I am looking for a method to connect two of the thinner lamps in parallel. I tried a direct parallel connection but the first lamp to strike took all the voltage.

    Would ballast capacitors work ?

    Any ideas would be appreciated.

  2. R!f@@

    AAC Fanatic!

    Apr 2, 2009
    If the lamps are OK and if the size is almost same it will light.

    I know cause I always check inverters with a replacement lamps that are of the same length. Regardless of thickness. My theory always worked for me.

    You might be having an inverter issue.

    Does the inverter shut down after lighting the tubes for a few milisec or two ?

    Try to go via this thread
  3. timescope

    Thread Starter Member

    Dec 14, 2011
    Thanks R!f@@,

    The lamps are soldered to a grounded copper strip at one end and plugged into a socket connected to the HV winding of the inverter transformers at the other end so it was easy to verify that the lamp itself was defective by exchanging the defective lamp with one of the others.

    The thinner lamp lights up but because the characteristics are not the same, the lamp voltage detection circuit (capacitive divider 3pF 6kV and a smd capacitor) and lamp current detection circuit (510Ω // 1kΩ in series with the grounded end of the HV winding) cause to inverter to shut down. I discovered this by measuring the voltages at the respective test points.

    I will do some more experiments and post the results later.