Guide Make your own USB LCD controller!

I've built mine up on some veroboard so I can play around with it a bit but I'm having some timing issues - when the text scrolls, it sometimes locks up or drags out the text. Ie. if it was displaying "hello" it might come out as "heeeeeello"

Would you mind if I also had the source code.. wouldn't mind playing around with it too.

hr350 [at] adam.com.au

Cheers

Hoops

 

with the text errors , try shortening the cable lenght to the LCD (5cm), also use a ribbon cable if possible.

Seems like the people that had character errors (my self included) seemed to have longish cables or used individual wires instead of a ribbon cable.

Also show us a photo if ya can.
I allways like to see what others are doing

Scorpia

 

OK.. I have the LCD (or actually its a VFD) connected directly to the veroboard via 14 SIL pins. I think there are two issues with my board at this point:

1) I do not have the pull up resistors on the inputs where the rotary switch would go (I don't have the rotary switch connected either)

2) I have relatively long and un-shielded wires from the USB connector to the PIC.



I will address these issues and let you all know how I go..


Hoops

 

If you don't plan on using the rotary encoder, you don't need to connect the pull-up's. You can just leave those pins unconnected.

 

unconnected logical inputs is a big NO-NO, you know...

 

Yes
But it also depends on how these pins are configured. I imagine a safer practice would be to tie them high or low, but since I don't use the encoder either, not sure what the results would be as far as function. CH424 would be able to answer it better I think as I'm not entirely sure if they are all configured as inputs or only some.

 

Hoopstar, looking at your board it looks like you're missing a capacitor. The capacitor on vusb actually needs to 220nF (ch424 has put 100nF on his schematic), but I can't see one on your board, unless it's one of those next to the electrolytic. You can get away with using two 100nF ceramic caps in parallel if you don't have any 220nF's. You also need one of your decoupling capacitors right next to the PIC at the supply pins. Solder it directly to the pins on the socket if you can.

This may help your problems.

How are you getting power for your VFD? If you're taking it from the USB lead, you might be asking for trouble - the power supply section will need a bit of re-working for it to work well.

 

Just wondering, would someone be willing to make me one for say $40-60? No enclosure necessary. Just a feeler to see if anyone is willing, I'm not 100% sure if I'm going use one or not.

 

I'd be willing, but it's still going to take a few weeks at best.

 

That's fine, I'll keep it in mind, like I said, I'm not 100% positive yet, Hopefully I'll be getting started on my next project in a month or so, depending how much my tax return will be. I'll send you a PM & we can work from there. Hope that's alright with you.

 

SteveyG,

Would you mind letting me know where you got the figure of 220nf from? Im trying to understand more about electronics and i figured that you didnt just pull the number out of the air,

Thanks

Scorpia

 

It's in Microchip's documentation

 

I have a 100nF on the vusb pin.. I will increase it up to a 220nF and report..

I am also only running one 100nF and one 100uF on the power supply and as you can see they are both a distance from the supply pins. Do both the Vss pins need to be connected??

I am powering the VFD directly from the USB port.. I think you're probably right about it drawing too much and might need a little more filtering on the supply side of things..


Hoops

 

Definitely!!! They aren't there for show!

Capacitor placement is cruciual. The bulk capacitor should be right at the entrance to the board (i.e. at the USB socket), and decoupling capacitors right at the power supply pins for the PIC. Odd stuff starts to happen if you don't do this.

Hopefully this'll take away your problems.

 

Thanks for pointing out the 220nF thing SteveyG, I've updated the schematic and parts list.

It's definitely best to leave the pull-ups in place. When I was designing it I didn't realise there were only pullups on PORTB - I should have used that for the buttons.

ch424

 

OK, here is the first attempt in Eagle. I've kept it single sided so people could etch their own boards if they wish. I'm only posting the picture right now in case anyone notices any mistakes that I may have missed. Once everything seems good, I'll go ahead and post the board file (Eagle .brd) as well.

 

You can get around those non-optimal tracks between pins - just mirror the crystal and capacitors so that the crystal is connected directly to the PIC pins. The capacitors will then line up with the Vss pin for easy connection to ground.

Run the track from R1 to the left hand side of the resistor and along the middle of the PIC to Vdd.

Remove the track going through JP1. It's a potential short circuit across the USB connector. Route it round the left side of Q2 and along the top.

The spacing between the 3 tracks going to the LCD header will probably be too close for most people to manage also.

If you're going to connect the USB shield to the 0V pin (technically you shouldn't) don't do it through that potentially nasty current loop. Take another track from the 0V pin to the top anchoring point of the USB connector . Also you can remove that track between pins near the rotary encoder header by taking the track between C3 and C5.

 

Thanks for the input SteveyG
I'll make those changes and post a new picture.

 

I uploaded the new layout with SG's suggestions.

 

Just having a quick look,

if you turned the contrast pot 90 or180 degrees then i suspect that the tracks could be cleaned up around the 2 pots.

Also i suggest moving the r1 to the end of the board or next to r2 to allow easier access to the I/O pins later.

Also for easier soldering i would suggest replacing the pads on the LCD header with the same size as the ones on the backlight pins jp1.
infact looking at it most of the pads could be increased in size to allow easier soldering as i suspect quite a few of the boards will be home made single sided boards.

hope that helps.

scorpia