CCFL backlight problems - device fall down (first time diagnose & repair)

[ed1973]

I have a Mega Macs 50 diagnostics device for cars, which fall hard on the ground. After that the LCD panel (NEC NL3224BC35-20) is working, but no light is coming from the 2 backlight tubes.

From the datasheet of the LCD panel : http://www.farnell.com/datasheets/68777.pdf?_ga=2.159499829.984870615.1543519062-2138990575.1540219252&_gac=1.204169764.1540219252.EAIaIQobChMI_-SX5qOa3gIVB-J3Ch37zwolEAAYASAAEgKxSfD_BwE  I have got the following info:

lamp max. voltage 1500V
T=25C starting 550V Typical 350V
T=-10C starting 780V Typical 350V
Frequency 39-47 kHz

Replaceable lamp for backlight :
Edge light type: 2 cold cathode fluorescent lamps
Replaceable part :
 Lamp holder set: Type No. 55LHS11

Recommended inverter (Option) :
 Inverter: Type No. 55PW131

The actual inverter in this box (MACS50_CCFL version 1.1), is different then the 55PW131, see the attached picture. On the inverter board there is a LT1768 controller, and I have gone through the datasheet : https://www.analog.com/media/en/technical-documentation/data-sheets/lt1768fs.pdf

Checked the transformer on the inverter board and all coils are in good condition.
On all input lines of the transformer (6) , I measured 14V DC (so there is no voltage difference) and I am thinking of measuring on the output lines, but do not know if that would blow up my Fluke 175 (despite it could measure up to 1000V AC and DC).

From the datasheet of the LT1768 controller, I do see 2 input lines which are measuring the current going though the tubes.

Actually (and I have never repaired an inverter) my biggest question is:

Is the problem related to the tubes, or is the problem related to the inverter ?

Can the tubes be measured ? I do not have the feeling that I can just measure them with my Fluke 175 multimeter.
Can I measure the output lines of the transformer, when the 2 tubes are not connected, with my Fluke 175 ?  I do have a feeling that there is no current flow, because of 14V on all lines ? I hold a CCFL light bulb next to the backlight connections, and nothing was seen in total darkness.
Can it be, that when the 2 tubes have been broken, that LT1768 does not measure anything on DI01 and DI02, and that LT1768 is not doing anything ?

Can somebody help me, where to start ? Your help will be appreciated very much, thanks in advance :-)

Kind regards,

Ed

[ed1973]

First I am going to buy myself a CCFL tester, unluckily they only come from China and I am located in Europe (The Netherlands).
Does somebody know where to buy a CCFL tester in Europe ?

[jiarui]

Do you have an oscilloscope? If you do, you should be able to see if there is any voltage on the input and output of the transformer and the amplitude and frequency of the input/output.  If you do have signal of the proper voltage and frequency on the output, the tube is likely dead.  If not, the problem is somewhere else.  I have had a similar problem with a digital mixer.  No back light. Ordered a new tube. Did not help. It turned out that the local regulator chip for the inverter circuit was dead. Replaced it and everything worked.

[Rasz]

go to your local recycle center and pull old lcd monitor out of the trash
I wouldnt even bother fixing this, instead look into LED backlight mods
like this

http://spritesmods.com/?art=ledbl&page=2

[Buriedcode]

Without reviewing any of the links I would say the CCFL is more likely to suffer damage than the inverter.

You were right not to measure the output with your multimeter, even if the nominal voltage is <500V it can peak far higher than 1kV when striking the tube, although there isn't much power behind this spike.. almost any loading should snub this. 

I had a similar problem with some old 5.7" LCD's, the only real way to test a tube is with the appropriate inverter, and only way to test the inverter is with the appropriate tube.  I ended up making a harf-arsed 20:1 voltage divider with cheap 100k resistors so I could use my multimeter to check the inverter output.  Note this will likely greatly reduce the initial strike voltage, so it won't reach the 1.5kv/20 at the start. This was with the (broken) tubes attached, but also it was a fairly simple inverter (royer two transistor inverter).  Yours is a proper controlled current fed royer (the LT1768), so requires feedback to control the current - and therefore the tube voltage, to function.  Like you, I am unsure if this would even function without a working tube attached.

You could check if the inverter is oscillating by scoping the bases of the switching transistors (that large 8-pin device is probably a two-transistor package). If you only have a multimeter with frequency measurement you could try their collectors.  These may reach twice the input voltage, but certainly shouldn't reach anywhere near the meters input ratings.  The LT1768 is the buck converter running at ~350kHz, but the royer converter will be at 39-47kHz (according to your specs).  I would have thought if the inverter isn't functioning, it won't be switching at all, so really you just need to see if there is AC there, or DC.

Hope that helps in some way..

The fact both backlight tubes aren't working does point to the inverter - but the shock could easily damage both tubes at the same time. 

[ed1973]

Thanks a lot for all your info, for sure I can continue now.

When I have news, I will post it here, thanks ! (Going to watch the youtube video first, very interesting)

[wraper]

It's more likely that CCFL transformer got damaged rather than lamps. It's a heavy part holding on it's terminals only. Check resistance of windings, some might fail open because of broken wire near terminal.

[Rasz]

The fact both backlight tubes aren't working does point to the inverter - but the shock could easily damage both tubes at the same time.

even just one bad lamp will trigger protection and shut whole thing down
previously (when I still bothered with ccfls) I would use old junk monitor/laptop to test lamps/inverter. Afaik those ebay ccfl testers are just repackaged old stock/recycled laptop inverters.

[ed1973]

Eventually I could buy 2 new CCFL tubes and I do have a CCFL tester.

Tested 2 tubes inside the LCD panel, both not working
Tested the new 2 tubes, both working

Installed 2 new tubes, connected everything, and still the tubes are not working. So there must be something defective in the backlight inverter PCB.

Now it comes to measuring high voltages, I do not have an oscilloscope, so I ordered resistors, which I will put in series, in order to measure high voltages with my Fluke 175 multimeter.

Should this be enough to find the root cause of the problem ? Or do you prefer a high frequency scope ?

Sorry for not updating you earlier.

[wraper]

Did you check transformer? Measuring high voltage is completely unnecessary.

Or do you prefer a high frequency scope ?

Very likely you will just kill the scope.

[ed1973]

I checked the transformer before, and all the coils were good.

I desoldered the transformer, and all looked good, all the values of the coils were good. So there seem to be nothing wrong with it.