[SOLVED] 128x64 LCD Guide. Can't see the components in Picture. Please identify.

Ok, I tried to just get the backlight working. I realized that I can't do anything else with getting a display on the LCD until I get the uno programmed to work with it via the glcd library. I will make a new thread for that.

But now since the backlight isn't working with pins 20 to ground and 19 to resistor+vcc maybe the lcd is now bad.

EDIT: IT WORKS!!! I had the library in the wrong folder. After unplugging the lcd to load the sketch as you suggested worked. Uploaded the glcd diag and it's passing the tests.
YAY!!!!

Ok, now all that needs figured out is why the backlight isn't working.

If diags is reporting success on the serial port then that means that all the data and control lines are hooked up
correctly.
The only wires/pins left are Vo (contrast in), Vee (contrast power supply out), and the backlight
pins.

Which pinout are you now using? (which data sheet)
Where did you buy this glcd?
Are you now seeing any pixels on the display?
Can you adjust the contrast with the pot?
You should be able to see pixels and adjust the "blackness" of the pixels using the contrast pot
without a backlight on that display.

Contrast: (if not working)
How is the contrast pot hooked up?
It should be hooked up as in reply #2
If the glcd Vee power-supply signal, which is an output, is connected directly to gnd or VCC
it can burn out the Vee power supply.
If the Vee power supply gets smoked then there will be no way to display any pixels.

Backlight:
The backlight was previously working. What changed?
Do you know how it was hooked up before vs how it is hooked up now?
(This is where careful attention to details and even notes can really help)
What value resistor are you using?
What other value resistors do you have available?

For the LED, incorrect polarity won't damage it, but a series resistor
that is too small might.

The backlight typically needs a series resistor to limit the current.
If it is too low or not used, that can allow the LED to pass too much current through it and burn it out.
(That can happen quite quickly - much less than 1 second under certain conditions)
If it is too high, the backlight LED won't have enough current to light up.
The value of the resistor can vary a lot depending on the design of the board.
It can be as little a few ohms or as much as a few hundred.
And some (very few glcds) don't need a series resistor.

The best is to have a datasheet that specifies the current needs of the backlight so a safe
resistor value can be calculated.
Missing that, the safest thing is to try high, then go lower but not too fast as you can go too low.
Multi turn potentiometers are good for this kind of experimentation but there are other
ways to proceed as well depending on the values of the resistors you have at hand.

--- bill

I am using the attached datasheet at the bottom of this post.

I am seeing the pixels display. They are showing as they should with the sketches loaded. The clock sketch has a running clock. Looks nice.
The contrast does not change when I adjust the potentiometer. It appears to be full on.

I haven't had the backlight working yet. I have a feeling that pins 1-20 don't feed the backlight. I see two holes in the board to the right of the display. I believe some LCD's have their backlight power supplied by those. Not sure if this is the case. The datasheet says 19 and 20 though.

The way I have this hooked up for GLCD pins 18-20 are:
GLCD pin 18(VEE) to one leg of pot(not the wiper).
GLCD pin 19(LED A) goes to a 330ohm resistor then to +5V(Arduino).
GLCD pin 20(LED K) goes straight to GRND(Arduino).

The pot is hooked up:
One leg to GLCD pin 18(VEE).
Center leg to GLCD pin 3(V0).
The 3rd leg of pot to GRND(Arduino).

Other connections for power:
GLCD Pin 1(VSS) to GRND(Arduino).
GLCD Pin 2(VDD) to +5V(Arduino).
GLCD Pin 3(V0) to wiper of pot.
GLCD Pin 17 to reset(Arduino).

datasheet attached:

DG-12864-53-S2FBEB.PDF (809 KB)

I'm pretty sure that datasheet you referenced is not for the glcd you have.
I looked around and there appears to be several DG-12864-xxx models
that are all slightly different. The differences seem to be in the color of the lcd and backlight
with the main power, data, and control lines all being in the same positions.
That datasheet you referenced is for a EL backlight. EL (ElectroLuminescent) backlights use a high frequency high AC voltage.
The one in this datasheet uses up to 150 VAC at 1Khz with a typical setting of 100 VAC and 400hz.

Since the backlight previously was working. The panel you have isn't using an EL backlight.
There would be no way to get an EL backlight to light up with just DC voltage.

**Do you know how you wired it up when the backlight was turning on? ** - This would really help even if it was when
the power supply was shutting down and the backlight was going back off.
Was the backlight connected up without a resistor? (A to +v and K to gnd)

When you say:

GLCD pin 19(LED A) goes to a 330ohm resistor then to +5V(Arduino).

I assume you mean you mean pin 19 is not connected directly to 5v but that power goes
through a resistor like this - (I couldn't see this part form your earlier photos)

glcd pin 19 ---------- ( 300 ohm resistor ) ---------------- +5v

For the backlight, It comes down to about 1 of 3 things at this point:

  1. incorrect wiring, broken wires, wires not making connection.
  2. the backlight is burned out
  3. the resistor is too large and there isn't enough current to light the backlight.
    You could try hooking up two 330 ohm resistors in parallel - both connected to pin 19 and both connected to +5v.
    That will reduce the resistance to 165 ohms.
    Turn off all the lights and see if you see any kind of glow at all.

There are some ways to diagnose this but it depends on what you have available.
Do you have a voltmeter?
What other resistors or potentiometers do you have?


The contrast pot not working concerns me.
When rotated to one side, all the pixels should be off.
When rotated to the other side all the pixels should be on - regardless of what the sketch is doing.

Do you have an ohm meter?

I'm still curious where did you get/buy this glcd? link to purchase site?

--- bill

Nah, never had the backlight working. I think it's burned out from my previous wiring errors. I tried with two 330ohm resistors in parallel and then with no resistor at all. No hint of lighting at all.
Still looking for battery charger for camera so I can post a pic of the back of this LCD.

Yup: glcd pin 19 ---------- ( 300 ohm resistor ) ---------------- +5v

Yes, I do have a multimeter.

I didn't buy it, I got it with another lcd(smaller) that I couldn't get to work. The only other lcd I've gotten to work is the one that comes in the Arduino kit. It has a backlight; that lcd worked like a charm.

I guess I misunderstood. I thought the backlight was previously lighting up.
It could be that backlight is burned or perhaps the glcd does have an EL backlight.
If so, to make it work you need high voltage high frequency to drive it.
You can buy EL power inverters for as low as around $5 USD.

What about the contrast control. Is that working?

Is the multimeter digital?
If it is connect the meter black/- wire to gnd, then connect the Red/+ wire
to pin 3 of the glcd.
The voltage you see should be negative.
Do you see the voltage change when you rotate the contrast pot?
At one end it should be like negative 7-10 volts and as you turn the pot the voltage should
rise up to zero at the other end.

--- bill

Ok, the range of LCD pin 3 to GRND is: +1V to -7.1V

EL inverter like this one?
http://www.ebay.com/itm/12V-Battery-or-Transformer-Power-Inverter-EL-3-meter-10-feet-Foil-Wire-Constant-/251107233479?pt=LH_DefaultDomain_0&hash=item3a77284ec7

Foolios:
Ok, the range of LCD pin 3 to GRND is: +1V to -7.1V

Hmm...
A positive voltage should not be possible.

How is the pot, and pins 3 and 18 wired up?
The voltage should vary between the voltage on the 2 legs of the pot.
1 leg should be ground
The other leg should be pin 18 or Vee.
The wiper (the center connection) is what connects to pin 3 and where the
voltage is being measured.

Vee is normally negative like negative 7 to 10 volts and then
ground is well ground.
So the voltage should vary between gnd and whatever Vee is (pin 18).

EL inverter like this one?
http://www.ebay.com/itm/12V-Battery-or-Transformer-Power-Inverter-EL-3-meter-10-feet-Foil-Wire-Constant-/251107233479?pt=LH_DefaultDomain_0&hash=item3a77284ec7

Yes. But there are some out there that run off lower voltages that might be better if you want
to power it from something other than 9-12 volts.

--- bill

Ok, checked it again today.

Connections are:
Pot leg 1 to LCD pin 18
Pot leg 2(center leg/wiper) to LCD pin 3
Pot leg 3 to GRND

With knob of pot to farthest twist to the right possible:
Leg #1 of pot is -9.6V
Leg #2(wiper) is -9.6V
Leg #3 is 0V

LCD pin 3 is -9.6V
LCD pin 18 is -9.6V

With knob of pot to farthest twist to the left possible:
Leg #1 of pot is -9.6V
Leg #2(wiper) is 0V
Leg #3 of pot is 0V

LCD pin 3 is +0V
LCD pin 18 is -9.6V

Thats more like what I would expect. ( -9.6v to 0v)
What is different between things now and in reply #20 which was (-7.1v to +1v) ?

Do the pixels now change when you rotate the pot?

--- bill

I have plugged it into a different usb port the second time. I will try to find the point at which I got -7v.

The pixels stay at the same contrast.

It isn't the -7 that seems odd, it is the > zero that is puzzling me.
Maybe the ground the meter was using was not a good ground?

I'm still very puzzled by the pixel contrast not changing when the voltage on Vo is changing
from gnd to -9.6v

It should be going from every pixel being on/black to all pixels off.

--- bill

It's ok tho. I am really glad it has a working display. I'm kinda bummed that I can't get the backlight to work. I am going to assume I blew it out.

I have another of the exact same lcd. I haven't touched it yet. I should be able to get this one lit up without problems now. I am going to go over everything one more time to try not to blow out the backlight this time.

If it doesn't light up, then more than likely the diagram isn't exactly right for this lcd. I will go back to the first one and mess with the two holes at the side of the display to see if those are the backlight pins.

Thanks for the help all.

Foolios:
If it doesn't light up, then more than likely the diagram isn't exactly right for this lcd. I will go back to the first one and mess with the two holes at the side of the display to see if those are the backlight pins.

Maybe not, it might need a very low resistor value - how low did you go?
I have some glcds that use 3-4 ohms.
It could also be an EL backlight model.

The safest way to test an unknown backlight LED after you tried 330, 165, and 110 ohms,
(assuming you hook up 2 then 3 in parallel)
would be to get a multi turn pot say 50 ohms, hook it up
to the backlight and slowly turn it to reduce the resistance to see if the backlight lights up.

From the earlier photos, it looks like there are traces running to header pins 19 and 20.
See if you can see where they go.
All the glcds I've got (3 different ones) have holes on the side like yours but are soldered
to the backlight LED module.
It could be that the PCB can be stuffed with either a LED backlight or an EL backlight.

Can you post a couple of clear closeups of the back of the glcd?
--- bill

Hope this helps

Take a look at this Pacific Displays datasheet:

I think this may explain it.
Take a look at page 2. It appears to have a magic decoder ring for the
information below the model number.

In your case it is: FC-R-ET-6-V-T

While not an exact match - since it isn't quite the right data sheet.
-R seems to imply:
Reflective (no backlight)

Sounds like the panel doesn't have a backlight.

Follow the traces for pins 19 and 20 and see if you can see where they go.
Also look carefully at the side of the display.
Often with displays that don't have a backlight you can see under it all the way through
the long ways.

--- bill

The first part # posted has an EL backlight... but an inverter is only $3.00 - $6.00

Doc

Also look carefully at the side of the display.
Often with displays that don't have a backlight you can see under it all the way through
the long ways.

You're right, there is a gap all the way across between the lcd and the circuit board.

So there's no backlight. Now does that mean there is no EL backlight as well? I don't know the difference.

EDIT:
Or would I need to do something like this:
(See attachment)

I think I got my answer here:
http://www.mp3car.com/general-hardware-discussion/3432-el-backlight-inverter.html

ELbacklight_Guide.pdf (1.11 MB)

Without looking directly, it's hard to tell for sure but from the photos so far, it looks like
there is no backlight.

Attached a blue backlight panel that uses an EL backlight.
You can see the EL layer sticking out on the right and it is soldered
to the PCB.

I've also looked at 3 other glcds and a few non glcds that use LED backlights.
(I attached two other glcds with LCD backlights)
In all cases - even those that have the backlight attachments on the side vs
on the header, the backlight has connections that are soldered to the PCB.
On the ones where the backlight has to be powered from connections on the side,
there the 4 holes. The backlight uses the two inner holes and the wires for power to the backlight use
the outer holes.

In the photos in reply #8 and #28 there is nothing soldered to the backlight connections
on the PCB.

So my suspicion at this point is that there is no backlight.

--- bill

panel_vk5121.jpg

IMG_5709.JPG

IMG_5710.JPG

IMG_5711.JPG

Ok, that makes it very obvious. Thank you for posting those.

Alrighty then. On to soldering this onto a pcb. A new adventure awaits. =)