LCD's should be driven with an AC signal - DC bias will kill the display in short order. Look at some of the mid-end PIC's which have built in LCD driver circuits:
Marks256 is right though - it's probably lots more trouble then it's worth to try to use them. Zebra strips and custom PC boards would be the most reasonable ways of dealing with "bare glass"...
No it's very difficult to drive an LCD without a PIC designed for it- and the LCD-drive PICs are not very versatile nor is there a wide selection at all.
This question has been asked time and time again. The bottom line is that recycling displays like this just isn't practical. Wish the opposite were the case.
We use LCDs all the time though- ones with drivers built onto the display. Then you just write commands across a bus to display an ascii character or (for a graphics display) write a display byte.
Hey, i had a 50-50 chance of getting it right. Gimme a break.
So you are saying that a calculator uses the pic to create a ac(oscillating) signal to power the crystals? That is wild... I guess ya learn somethin' new every day!
I just pulled up-part an old +20 keys calculator. (I forgot where all the buttons go back).
At 3 volts with direct current the segments will only turn on for about 1/2 sec and then they dissapeer.
Next I programed a pic to alternate two outputs from positive & negative to negative & positive, 100 times a second.
The Pic is powerd with 4.5 volts. One of the outputs was connected with a 1k resistor.
I Then connected the 2 outputs to the calculator display electrods.
The display lights up nicely.
Just got to work out the pinouts for the display, as it's not very logical.
You drive the segment out of phase with the backplane to turn it on. In phase to turn it off. You don't need a very high frequency either: somewhere in the 50-100Hz range I think.
I did this a long time ago with a display designed for a watch using a 68HC11. The problem was that the 10 minute digit was different than all the others with two of the segments joined together to save a pin. This was possible because the minutes digit only had to display numbers between 0 & 6.... So a calculator display would be a better choice than one from a watch.
what cuurent, or do they nead current limitting resistors?
They don't much current beyond that to charge/discharge the capacitance of the segments so no resistors are required. That's why the tiny watch battery lasts so long.
The problem was that the 10 minute digit was different than all the others with two of the segments joined together to save a pin. This was possible because the minutes digit only had to display numbers between 0 & 6.... So a calculator display would be a better choice than one from a watch.
I can't work out the pinouts for that calculator display. there are about 30 pinouts when there should be around 20 if it was in LED mutiplexed mode.
when I connect two pins, 6 segments light up. I'm confused.
Don't forget about speciality symbols like + - fix sci etc.... If you hold the display at the correct angle to the light, then you should be able to see all the "segments"... How may digits and symbols does the display have?