Trying to figure out the pinout of an LCD display (no driver)

[Michael Lloyd]

I picked up some 24 pin displays for $1.00. I thought it would be easy to find a data sheet  but that has proven to be incorrect. There is no part number and a fairly lengthy search of the web turned up nothing that looked like it.

It supposedly was used in a weighing scale that displayed 99lb.99oz. It has 28 pins and is 2" L X 0.86" W not counting the pins. Any ideas?

[Maxlor]

You'll need some sort of AC signal (rectangular is fine) in the 50-150Hz range at probably 3V. If you have a signal generator, use that, otherwise an arduino might do in a pinch. Test various pin combinations until you can identify the 3 or 4 COM pins, then test all other pins vs. those COM pins to identify each segment.

Do not apply DC voltage to it! If nothing happens at 3V, slowly increase the voltage. If still nothing happens at 5V, it's probably broken.

[Michael Lloyd]

I've got a couple of different options for a square or sine wave so that's easy. Thanks for the comment. I'll see what I can come up with. I bought a tray of 24 for $1.00 so I'm not out a lot of cash for them. They appear to be NOS.

[Michael Lloyd]

The LCD didn't behave like I thought it would. I used an HP33120A Function Generator. Set to sine wave. 3.0V p/p.

I have a BNC connector on the function generator with a set of test clips on the other end. It doesn't matter which lead I use. One lead will light a segment, sometimes two segments. Sometimes the word "oz". Sometimes the word "lb". And the 4th position has 1/2 sized characters. Sometimes the segment stays on for seconds after I lift the lead. Sometimes it goes off right away. I didn't write anything down   

The pins are oriented 17 pins on the bottom and 11 pins on the top (orientation determined by the text that pops up and the decimal points)
Pins 5-8, 16, and 17 on the bottom row don't do anything when individually probed
Pins 1, 4, 9, and 11 on the top row don't do anything when individually probed <-- this kind of mimics the orientation of the common on other LCD displays that I found datasheets for but using the pins as a common doesn't affect the way the bottom row of pins responds.

[Lukas]

You'll probably need to connect all pins of the LCD.
I've successfully reversed 'dumb' LCD like this: Connect all pins to an MCU, toggle all these pins in-phase. Now, invert one pin at a time and watch how many segments turn on. If many segments turn on, it's a COM pin, if not so many, it's a SEG pin.

[Ian.M]

LCD segments have considerable capacitance.  If you are testing a bare glass LCD, you really need a leakage path to discharge the segments you aren't probing.  Try putting it in a breadboard with a separate 1M resistor from each pin to a common bus.  Leave the bus floating.

Probe a pair of pins with an AC voltage.  If multiple segments show, you've got two segments or two commons,     Once you find a segment/common pair, mapping the rest should be fairly simple.

N.B. the test signal should have no DC offset.  If you are generating it from an Arduino or similar and you aren't 100% sure you've got exact 50% duty cycle antiphase signals, capacitively couple it!

[Brumby]

N.B. the test signal should have no DC offset.  If you are generating it from an Arduino or similar and you aren't 100% sure you've got exact 50% duty cycle antiphase signals, capacitively couple it!

YES.

Current draw is, effectively, negligible - so you won't need much in the way of capacitance.  Enough DC on a segment will make it go black - and stay black.

[Michael Lloyd]

Thanks for all of the replies. I checked the function generator output and it's DC component is 0.005VDC with the output set to 3.0V P/P and sine wave. I've got it on a breadboard so it will be easy tie the pins to a common buss with 1M? resistors. I'll check back in when I've tried that.