LCD backlight not working

[kiltro]

Hello everyone!

I'm trying to repair a pcb for an lcd display, the display works but the backlight doesn't turn on.

On the pcb there's a label TG12864H-18, I can't find any datasheet but there is TG12864H-17 and TG12864H-19 here.

http://www.tinsharp.com/products_list/pmcId=47&pageNo_FrontProducts_list01-1329189735664=2&pageSize_FrontProducts_list01-1329189735664=15.html

The 3,3V for the backlight supply are on the pcb connector, I suspect the problem can be the diode on the backlight power supply.
On the image below in red is where I have removed the diode.

Using continuity test in my multimeter it bips in both directions, it this enough to say that the diode is dead?
What kind of diode can this be? Zener for overvoltage protection?

Thanks

[DrG]

Assuming that you are confident that the backlight is 3.3V....Before going too much further effort-wise, I would test the backlight. I can't tell from your picture if the two backlight leads are now isolated, but if they are (or you make them isolated), what happens if you connect 3.3v through a current limiting resistor, and to gnd. The backlight LED should light up. (assuming you have +/- correct - and you can test reversing them without damage because it is a diode and you are limiting the current), if it doesn't, the backlight led is fried and I don't know what you can do other then replace the display.

That's what I would do anyways.

[kiltro]

Thanks for helping DrG!

No the two leads are still soldered now.
Measuring on the two backlight wires at the board input (I had unsoldered them previously) I got 3,3V indeed.
As it's shown in the datasheet of the TG12864H-17 and TG12864H-19 in my link posted before.

I'll try what you suggest. Can I assume 15mA to choose the limiting resistor?

[DrG]

Thanks for helping DrG!

No the two leads are still soldered now.
Measuring on the two backlight wires at the board input (I had unsoldered them previously) I got 3,3V indeed.
As it's shown in the datasheet of the TG12864H-17 and TG12864H-19 in my link posted before.

I'll try what you suggest. Can I assume 15mA to choose the limiting resistor?

Yeah, 15 mA was what was stated in one of those datasheets, but you only need to detect whether it comes on or not. I just happen to have a backlight and stuck a 330 on it with 3.3V and it was easy to see it working.

[kiltro]

Yeah, 15 mA was what was stated in one of those datasheets, but you only need to detect whether it comes on or not. I just happen to have a backlight and stuck a 330 on it with 3.3V and it was easy to see it working.

The backlight lights up, so the problem is not there

[DrG]

Yeah, 15 mA was what was stated in one of those datasheets, but you only need to detect whether it comes on or not. I just happen to have a backlight and stuck a 330 on it with 3.3V and it was easy to see it working.

The backlight lights up, so the problem is not there

The "easy" check would then be to make sure that the LED+ and LED- cable leads coming off the board were not accidentally reversed. You now know the backlight +/- leads so it is easy to check.

If that doesn't do it, I would wait for someone smarter to answer 

[kiltro]

No, unfortunately they are not...

But now the problem should be in this part of the circuit.



Without the diode I have 3.3V on the backlight pins (on the right) so... I think it is fair to think it's the dead one

but what kind of diode can this be, a 3.3V zener?

[DrG]

No, unfortunately they are not...

But now the problem should be in this part of the circuit.

(Attachment Link)

Without the diode I have 3.3V on the backlight pins (on the right) so... I think it is fair to think it's the dead one

but what kind of diode can this be, a 3.3V zener?

I don't know without the schematic or at least the removed SMT, but I would be tempted to try a generic 914 or a 1n4001 because you did say that there was 3.3 coming in - but, again, that's me.

[kiltro]

I can be wrong but judging from my picture I'd say it something like this?

[DrG]

I can be wrong but judging from my picture I'd say it something like this?

(Attachment Link)

Do you have a clear pic of the component that was removed?

[kiltro]

I belive it's SOD80 so 3.5mm x 1.6Ø

[DrG]

(Attachment Link)

I belive it's SOD80 so 3.5mm x 1.6Ø

I thought it would be a surface mount. I don't know why it would be a zener but, again, I don't have a schematic. I would use a 914. Others here likely know much more and might be able to offer other suggestions.

[kiltro]

Yes it's surface mount indeed

[DrG]

Yes it's surface mount indeed

ahhh ok, you're right, I see that now.

[Till Matte]

I suggest you power on the lcd module, then check the voltage of backlight input, you could use the multi-meter to check the voltage between the A,K pin, if the +3.3V was input inside the backlight unit, if the power was input to backlight, but the backlight still not work, the problem would be from the backlight burned in, otherwise, it would be from the PCB circuit, then check the input voltage step by step on the PCB components, then you find which component was burned, or maybe from the surface mounting problem.

[kiltro]

I suggest you power on the lcd module, then check the voltage of backlight input, you could use the multi-meter to check the voltage between the A,K pin, if the +3.3V was input inside the backlight unit, if the power was input to backlight, but the backlight still not work, the problem would be from the backlight burned in, otherwise, it would be from the PCB circuit, then check the input voltage step by step on the PCB components, then you find which component was burned, or maybe from the surface mounting problem.

Thanks, but as I said before, by removing the backlight and powerin it directly it works so it's not burned.

Also removing the diode I get the voltage on the two solder pads of the backlight so the problem is most certainly the diode.

My problem though, as per OP, is defining what kind of diode it is.

[Zero999]

The 180R resistors limit the LED current. The diode is just there for reverse voltage protection. It makes no sense having an overvoltage protection diode, because it would never conduct, as it would need to have a higher breakover voltage than the forward voltage of the LEDs. If it's too difficult to find a diode in a surface mount package, just replace it with a 1N4148, with the leads trimmed and bent so it can be soldered to the pads.

[kiltro]

The 180R resistors limit the LED current. The diode is just there for reverse voltage protection. It makes no sense having an overvoltage protection diode, because it would never conduct, as it would need to have a higher breakover voltage than the forward voltage of the LEDs. If it's too difficult to find a diode in a surface mount package, just replace it with a 1N4148, with the leads trimmed and bent so it can be soldered to the pads.

So even a 3,6V would be useless?

As a side note, the silkscreen also shows a diode and not a zener

PS. the two res are 18ohms I mistakenly wrote 180 because the label on the component is "180"

[Zero999]

The 180R resistors limit the LED current. The diode is just there for reverse voltage protection. It makes no sense having an overvoltage protection diode, because it would never conduct, as it would need to have a higher breakover voltage than the forward voltage of the LEDs. If it's too difficult to find a diode in a surface mount package, just replace it with a 1N4148, with the leads trimmed and bent so it can be soldered to the pads.

So even a 3,6V would be useless?

As a side note, the silkscreen also shows a diode and not a zener

PS. the two res are 18ohms I mistakenly wrote 180 because the label on the component is "180"

LEDs require a constant current source to work well. This is because the current increases exponentially for any increase in voltage. If the forward voltage is around 3V, at the rated current, then the current will be far too high, if the voltage were to be increased to 3.6V, so no, a 3.66 zener won't help at all.