More LCD Backlight questions

Skip to first unread message

Vance Briggs

Jul 5, 2021, 9:38:30 AMJul 5
to Reading Hackspace

The character LCD display that I picked up from the giveaway a couple of Sundays ago:


I am trying to work out the requirements of the backlight - connector pictured above leading to white "backlight" panel.  The supply IS NOT reversible, so its either LED or has some built in power conversion that expects the correct polarity.  I have connected it up to my PSU with current limiting set.  I have tried different voltages and gradually increased the current limit until it is no longer limiting.

The panel doesn't light at all until it gets around 4V and draws 0.2A and at 5V it takes 1.25A before current limiting has no effect. (I only applied this for very short periods in case it went bang!)

This looks very much like 1980s technology, so my questions are:

What is the backlight technology?  (Electroluminescent? - if so I guess the inverter circuitry is hidden under the panel)

1.25A @ 5V seems very high to me (6.25W), so I guess that I may need some form of current limiting.  If this were LED backlighting then I would guess that the forward voltage across the LED(s) would be in the region of 4V then I would need a 5 Ohm 0.2W resistor, assuming I  was aiming for the 0.2A current. 

Alternatively I could PWM the supply, but I am not sure how that would affect the life of the backlight.

Any suggestions for further investigation or how I should drive this safely please let me know :-)


Vance Briggs

Jul 6, 2021, 10:17:52 AMJul 6
to Reading Hackspace
TLDR;  Does anyone have any recommendations for a constant current source IC capable of delivering 0.75 - 1.0A with a 5V supply?

After further investigation - and dismantling it looks as if the backlight is an LED matrix with no evidence of any series resistor.

There are 72 LEDs, and the layout effectively puts 36 LEDs in parallel between Anode and a mid-rail then another 36 in parallel between the mid-rail and the Cathode, so the forward voltage needed to light the array is 2 x the single LED forward voltage.  From experimentation this puts the single LED forward voltage around 1.8 - 2.0V.  So I need to aim for 3.6 - 4.0V forward voltage.  Assuming the LEDs will draw between 10mA and 20 mA each, this gives me a safe current of < 36 x 0.02 = 0.72A.  I will need to place a constant current IC or circuit to drive this backlight safely. 

Does anyone have any recommendations for a constant current source IC capable of delivering 0.75 - 1.0A with a 5V supply?


Alex Gibson

Jul 6, 2021, 10:36:02 AMJul 6

Hi Vance


I’ve had some success working with similar boards before and they are usually fairly standard, based on a Hitachi driver from the 80’s.

If you’ve not already seen it, here’s a guide to interfacing:


I can’t answer the constant current question but across various examples of these displays, I’ve ‘just’ fed the backlight 5V from available PSU – normally 5v 1-2A or straight from an Arduino – and been ‘OK’.  Is there a measurable resistance to the array, even if there is no visible resistor component?




Alex Gibson


+44 7813 810 765    @alexgibson3d    37 Royal Avenue, Reading RG31 4UR


admg consulting


edumaker limited


·         Project management

·         Operations & Process improvement

·         3D Printing

You received this message because you are subscribed to the Google Groups "rLab / Reading's Hackspace" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
To view this discussion on the web, visit





Vance Briggs

Jul 6, 2021, 3:10:13 PMJul 6
to Reading Hackspace
Alex, thanks!  I think I am ok with the data side of things, but just designing a board and want to make sure I am able to drive the backlight properly. 

If I just supply with 5V from an "unlimited" current supply the backlight takes 1.5A, which worries me a bit. That would be about 42mA through each LED, which seems high to me. 

LEDs will often run for a short time without a current limiter, but do then burn out. I don't want that happening with this display.  So I am trying to make sure I drive it correctly. 


Richard Ibbotson

Jul 7, 2021, 4:26:58 AMJul 7
If your power supply is stable then I would suggest to
just use a resistor. Simple constant current circuits have a voltage drop due to the current sense and the control element and you don’t have a lot of Volts to spare.
Do a test.

Sent via Richard's Phone

On 6 Jul 2021, at 20:16, Vance Briggs <> wrote:

Vance Briggs

Jul 7, 2021, 11:29:13 AMJul 7
to Reading Hackspace
If I assume that the LEDs will want between 10mA and 20mA maximum
 then I think this gives me a couple of measured operating points:

10mA per LED ~360mA needs 4.2V across the panel  => R = 0.8 / 0.36 = 2.2 Ohms
20mA per LED ~720mA needs 4.4V across the panel  => R = 0.6 / 0.72 = 0.83 Ohms

I don't have any resistors that small :)

Another alternative could be just to put another diode in series?  If I can find a suitable one that has a forward voltage drop of around 0.7V @ 0.5A...

1N4001 looks just a nice fit here - measured as 0.83V @ 0.5A, so its a bit dimmer, but safer.

Any other gotchas I am missing when using a diode - other than the obvious need for a stable 5V supply?


Reply all
Reply to author
0 new messages