Guide Make your own USB LCD controller!

Scorpia,

Thanks for the input. I'll work on the pots and R1. As far as enlarging the pads, as far as I know, it's not something that can be done....easily that is (in Eagle anyway). I would have to go in to each component library file and enlarge the pads there. It's not something that can be done directly in the board layout window. Unless you know of another way?

 

It's easier than that, I can't remember off hand, but I'm pretty sure you can click on a component and edit the properties from there without affecting the library. You can also just place a pad over the original pad and it'll change the pad size for you.

I'll check tommorow when I'm in the office.

 

I think it's done if you set it up in your DRC, under shapes or restring. I can get it to enlarge the pads by changing the top/bottom setting for restring. I was hoping to enlongate the pads (for easier solderability), which is under shapes. But changing those settings don't seem to help at all. I've also tried creating a general elongated pad with larger drill and pad size and place over the holes, but it won't show up like the pad I created once I place it on the board . I'll play more with it later.

SG, if you do find any more info, please let me know. Thanks.

 

You're right - it's not as easy as I thought. I mainly use Proteus which makes stuff like this a LOT easier. What a pain...

 

Woo! Let's hear it for Proteus!

Seriously though, I've just realised that those pads are definitely too small. Also, there's no way I could put a track between the pads on that USB connector, using the equipment we have at school. I guess it'd be OK if the board was being made by pros though?

Thanks
ch424

 

Easily achievable by a PCB manufacturer, but probably not by most people's home equipment.

 

Hi again

Sory for the delay, but I was a bit bussy modding my PC. Now I've got the LCD, the controller and 4 GPOs wired up and working. Here are some pictures:


Thanks ch424 for this great project!

About modifying the size of the pads on the PCB, you need to modify "Design Rules". It is described here

 

I've figured out how to enlarge the pads. The only thing I can think of doing to fix the USB connector is something that you had suggested to me before. I could just put in a 4-pin header. That way the user can either just hard wire it or use a connector. Plus, the more I have to move around, the more "wasted space" I seem to be generating.
Sorry it's taking so long to everyone waiting on the Eagle version.

 

Would someone like to give me an idea what trace width and drill/pad size would be easiest for home-made boards? I'm in a class right now and don't have access to a printer to see the actual size of everything, so it makes a little hard to judge the sizes when I'm drawing it on the PC.

 

im not sure what sizes they are etc, but if you use the pads on the pic socket as a reference then you should be ok. I have allways found that the pad sizes on the IC sockets seem to be a good size.

Also as for the trace though the USB try turning the capacitors under the USB socket by 180 degrees, that should allow the traces to come directly off of the USB. i think you might need to add a 0 ohm resistor (jumper) to the board to make the traces easy to run.

See if that helps

Scorpia

 

I have been having a look at making an eagle board myself,

since pins 11-14 on the LCD seem to swap over to pins 24-21 on the pic this seems to make trace routing a nightmare, im going to try and reconfigure the I/O config in the firmware so that pins 11-14 line up with 21-24, this should make running the PCB heaps easier.

Also im thinking about making an LCD header with pins 1-14 being the LCD and pins 15-16 being the backlight. this is how my LCD is connected so i thought it might be easier with all 16 pins in a row. do other peoples LCD's have the backlight as pins 15-16 on the connector?

EDIT: hhhmm, not as easy as i thought. Ch424, i guess you use the bottom 4 bits of port b as bits 1-4 (0-3) and thoughts on swapping then to bits 4-1?

is there an easy way of swapping 4 bits around so that it can be outputted backwards?

Any thoughts?

Scorpia

 

I think you have to leave the D0-3 bits as they are... consider moving RS/RW/EN bits onto PORTA instead? I've managed to fix it on my revision of the board. I'll post how tomorrow... It should be easy enough to reconfigure the bits in the firmware: io_cfg.h, except the D bits, because of all the masking going on in lcd.c

ch424

 

yeah, once i had a read though the lcd.c i figured it wouldnt be as easy as i thought

 

OK, it looks like I've moved the 4 data bits (D4-7) onto RA0-3. Then on RB5, RB6 and RB7 are RS, RW, EN respectively. That allows PORTB to be used for buttons, which is convenient as PORTB supports internal weak pullup resistors.

Hope that helps!

ch424

 

if you put RS/RW/e on RB5-7, you have moved the backlight? where to and how did you change the LCD.c? because it looks like it reads port b directly and then assumes it on rb7.

 

A 16 thou track should be fine, and DIY PCB manufacturing, with toner transfer or photolithography, will easily handle this.

If somebody points me to an up-to-date circuit schematic, i'll knock up a PCB design, i'm not too bad at it.

Cheers.

 

Brilliant project guys,

I have got hold of a 18F4550 and was wondering if anyone had the source code , hex, inf file etc. so that I can give it a try. Scorpia's links no longer work, and I'm not sure whether anyone tried the code out. If not the source code for the 2550 would be good and I will give the conversion a go.

Also, has anyone tried expanded the circuit to include a fan controller? I noticed in one of ch424's photo's that there was fan headers in the background....

Many thanks,

John

johnpugh24(at)yahoo.com

 

I posted a diagram for one fan header earlier in the thread. You'll have to add it yourself, I'm afraid: I didn't include it in the actual published design because I haven't found a failsafe way of making it work with LCDSmartie and didn't want to be responsible for peoples' fans starting/stopping randomly!

Scorpia: I haven't been working on it I'm afraid, so haven't sorted the backlight stuff yet.

 

I have fixed the links, they should be ok.

i havnt had any time to look at the other stuff since the last post either

Scorpia

 

What I did on my fan controller was to have hardware that ensured a minimum of 5V on each of the fan outputs, which was enough to keep them ticking over in the event of the PIC messing up. The PWM waveform was therefore between 5V and 12V, not 0V to 12V.