Tetris on a PIC

[MrDEB]

I realized my mistake BEFORE I etched so its easier to remove the black imprint.
Will post hopefully later today.

[brad]

Look forward to it!

[MrDEB]

Tried something new by using an aquarium air pump and agitate the etching solution (muratic acid/hydrogen peroxide)
Worked very well and I read somewhere that using air for agitation it tends to rejuvenate the solution?
All I know is it works very well but need to use more bubbles. Just directing the air towards the board. My 2x method is pretty accurate but kinda messy. I print one side of the board then paint the other side. Using lacquer thinner, remove all printed and paint. Repeat on other side.
I use 3 stick pins to orientate the two side together. Method works really well including using air for agitation. Tried other methods with some success but tends to mess up some traces or very slow (wiping and/or rocking. This board only has one very short trace missing.
Somewhere here I posted a DIY on home etching as well as over at DDIY and Electro tech.


http://i992.photobucket.com/albums/af44 ... board1.jpg

[brad]

I must say, that's a good looking home made board!

Well done

[MrDEB]

Been getting better each time. Trying different things. This time using air to agitate the solution. Using the PULSAR toner transfer method but I found using the green TRF smudges the black. I print at 1200dpi resolution. Have a laser cartridge just for pcboards only.
Found several design mistakes. The pads for the Matrix are too small and the 22pf cap pattern is wrong. I thought I had some 22pf caps in the 2mm ls but not. Putting the crystal and caps under the PIC socket on same side to conserve space (Been wanting to try this).

Hopefully finish soldering tonight.

[brad]

Been getting better each time. Trying different things. This time using air to agitate the solution. Using the PULSAR toner transfer method but I found using the green TRF smudges the black. I print at 1200dpi resolution. Have a laser cartridge just for pcboards only.
Found several design mistakes. The pads for the Matrix are too small and the 22pf cap pattern is wrong. I thought I had some 22pf caps in the 2mm ls but not. Putting the crystal and caps under the PIC socket on same side to conserve space (Been wanting to try this).

Hopefully finish soldering tonight.

Make sure you post the pics!

[MrDEB]

Here are some pics of the completed board. Note the crystal inside the 40 pin socket frame.
The 30G wires poking out of the board are a simple method to connect the top layer to the bottom layer on a 2x board. Much cheaper and easier than using automotive defrost repair glue or rivets. It works only if you remember to insert in all the right holes and solder one side to a trace. The 18 pin socket has several wires to tie top to bottom layers.



I was going to solder the 5x7 matrix to the board but made it removable in case I need to make changes.

[brad]

Nice work, I also used that method of connecting tracks to the two sides - the hard part was getting them lined up correctly for etching.

Where do the matrix' connect to?

[MrDEB]

The matrix connects to the bottom (opposite side from the PIC and ULN2803's. My problem is the pins on the matrix (wanted to be able to remove seeing how this is just for developing code only) won't contact the socket headers. Have encountered before so on to the soldering iron and a perfboard.

[brad]

Hmm, I still can't make out where they will connect to. Will they be on a separate board?

[MrDEB]

Will post a picture but first I need to figure out why the row 3 LEDs fail to come on. Hopefully its the code but tried several things to test it but to no avail. Have a hunch the row connection to the pic is faulty. Get an output on the pin at the ULN2803 etc
Next I need to check the dang header to matrix connection.
In the mean time, the matrix plugs into two headers located on the opposite side of the board (pictured) adjecent to the 18 pin sockets (note the 9 solder joints at each end of the board)

[brad]

Perhaps you're using a different matrix to the ones I use. The ones I use have 24 pins (12 on each side) 8 x anodes, 8 x green cathodes and 8 x red cathodes.

[MrDEB]

The matrixs has 18 pins (Kingbright TBC40-11GWA)
The issue I have found is I can't get the RD.0 pin to got high. I have yet to find any issues with wiring so hacking at the code just trying to configure the code so all the port pins go high. Using a ULN2803 to sink. Wondering why the LEDs seem not very bright? I started with 330 ohm resistors, put another in parallel but still not very bright. Could be an issue with the ULN2803 wiring?pin 9 grd and pin 10 Vcc.
Going to have another look at the data sheet. Keep thinking I need Vcc connected to the LED as well?

[brad]

Don't connect pin 10 - this simply places a diode in parallel with all your led / resistor combination (limiting the voltage across the two)

Connect them just like this:

screen-capture-41.png (68.22 KiB) Viewed 20494 times

[MrDEB]

good reason why the leds are not real bright.
Still trying to track down why pin 19 won't go high in my code. Seems like an issuer with portD but will figure it out when I get back to it. Been trying to stay indoors but have work outside (lots of smoke due to a forest fire about 10 miles up the road but in a valley so ALL the smoke is in Salmon, Idaho. Contemplating going south until smoke clears.