Nokia 5110/3310 LCD...?

[JimboZA]

Ok, I didn't go about this too scientifically, didn't record the combinations I tried, but I tried to get this damned thing going with 2 or 3 libraries, to no avail. Then I spotted  this link to Stuart Lewis earlier in this thread and tried it. Still no luck. It's the one with no libraries: I thought I'd try that, reducing the number of places for errors.

So then I tried the sketch his code linked to and was based on, namely the Hello World sketch in the Playground and Voila! it worked.

I thought, ok, let's go back a step and try Stuart's again and trouble shoot a bit... no trouble to shoot, it worked all of a sudden.

So are these displays a bit temperamental? Did I have persistent loose breadboard pins which a bump set right? I have no idea... At least I know the display works.

But now I'm going to switch to a library-based method, since that's a better way innit? One thing I will do though, is use the breadboard / LCD setup that I currently have, not going to change the hardware at all, and just set the pins in the sketch. Try and disrupt things as little as possible.

Jings, this is frustrating, butSaturday arvie not totally wasted.....

[Nick_Pyner]

I found the Stuart Lewis the best way to get things going but I'm sure it makes a lot more sense to use a library in the end. Although I didn't like digging into it to set the pins up, the standard Philips PCD8544 library is an obvious choice.  I don't think I had to make any bias adjustments in order to get the text as clear as Stuart Lewis's.

The Philips library enables you to set the cursor by pixel. You can also use their on-line glyph editor. The two together enable you to  tart-up the display and often get an extra character on a line.

But now I'm going to switch to a library-based method, since that's a better way innit?

[JimboZA]

Since then I tried Henning Karlesen's library, which I know I had tried unsuccessfully before, and it worked.

Then  powered down, went out for an hour, powered up and zilch.

This is really weird..... that said, I'm putting 5v straight in everywhere except for 3.3v to the power pin, so I'm wondering if that has anything to do with it? A previous post spoke of the LCD taking an hour to recover. So tomorrow I'll cobble some voltage dividers together and see if that makes for reliability.

[Nick_Pyner]

I run all mine off the 3v3 with no dividers or shifters. Only one gives trouble.

I think Buggsy's reply #13 is noteworthy. It implies a mechanical problem. It occurred to me that heat from the four LEDs may be getting into the act. I have already tested connections and ensure there is no shock while under test.

While I reckon this random shutdown stuff is arrant nonsense, recovering after an hour may also imply heat stress. That seems a hell of a long time but, when dealing with such low voltage and current in the norm, maybe that can happen..

[JimboZA]

At the risk of flogging a dead horse....

This morning I hooked the LCD up with series resistors as suggested in the second Fritzing here, and all is good. I can switch it off and straight back on or off and leave it, it comes on ok whatever.

Of course, in fitting the resistors I completely re-did the breadboard, so I'm not 100% convinced it wasn't a connection problem all along.....