Right now I'm testing the firmware with a PIC18F2525 that I had laying around. The firmware difference between that part and the one that I want to use in the long term is mostly in the processor initialization. The readability of the code isn't too badly affected IMO by the conditional assembly statements that allow either processor to be used.
The board design is still up in the air. While I have a more or less finished design, there has been a request for a more square form-factor. So there is still the possibility of change by me, as well as the possibility of others creating their own board designs. For powering the board I'm considering leaving it the way that it is (just feed it 5V), adding an input jack and LDO regulator to allow it to be powered from a 6V DC wall-wart (not uncommon in the US), or going to the full-fledged low-voltage AC-input design (input jack, full-wave bridge rectifier and H11AA1 and SMPS), or perhaps all three. There's a trade-off here between designing it to be non-intimidating and non-confusing for the total newby versus more useful for everybody.
Nice work phil.
if you need someone to test here in AUS with 50hz, let me know. I still have lights up for another few days at least (bit slack)
Phil, this is a minor detail but I'd hate you see you put in all this work and have a typo on the silkscreen.
I think the 4th RJ45 across the top is labeled wrong. It is labeled as channels 25-32 and I think it's supposed to be 25-28.
P.S. I love the concept of this board. I started with the Ren16(and I'm happy I did) but I probably would have leaned it the direction of something like this when I started out if it was around. The idea of easy and upgradable(by adding ZC, ect) is great.
ctmal,
Thanks, I'll fix that. I haven't started fab on any boards, so your correction is very timely. There are a few other things that I've fixed on the board as well, so maybe I'll post a new version in a few days.
scorpia,
Right now everything seems to be working except the ac line frequency detection. That is mostly working, except that it seems to be causing quite a bit of jitter in the PWM output signals (everything is clean when I hard-code the relevant variable). So maybe in a day or two. BTW, what part would you be using?
ok i have the following 28 pin pics.
18f2420
24fj64ga002
18f24j10
18f2550
40 pin i have
18f4525
18f4550
16f877
let me know if any are close enough otherwise i can order both the different pics your playing with today.
edit: looks like the 18f2420 is close, same pinout etc just less flash/ram etc. still has a fair bit so maybe you could add processor codes for it.
and the 18f4525 is the 40 pin variation of your 18f2525. should be an easy swapover if required. Let me know otherwise ill order later today.
Peter
I haven't looked at the PIC18F2420, but based on what you say it should work. The firmware only uses a small amount of RAM and Program memory, so if that is the main difference it should be fine. I'll take a look at the register set later on today or tomorrow.
The auto line frequency detection now seems to be working, so the only left for now is to check the LED operations. I should be able to email you something later today or perhaps tomorrow. Could you send me a pm with your email addy in it?
pm sent.
easiest way to check the pic would be the add the definition to the code and recompile i would have thought.
anyway good luck with it and hopefully ill have an email tonight/tomorrow maybe
Peter
with this code does it do some sort of channel flash on turn-on?
i have been talking to a few people and we all agree it would be a good idea. not sure if its been mentioned before.
the thought was to show that the pic is running and the channels are ok. the original idea came from the jec pixels whichflash all colors on and off at power up to show that it is working.
Should all of the channels come on at once to test the power supply? Or a one-cycle chase sequence? Or the lights coming on one at a time until all 16 lights are on, then turning them all off and going to computer control?
P. Short wrote:Should all of the channels come on at once to test the power supply? Or a one-cycle chase sequence? Or the lights coming on one at a time until all 16 lights are on, then turning them all off and going to computer control?
The last option sounds good - coming on one at a time would be helpful as a quick verification that the lights are connected to the correct channels, and then all on at the end is a good check for power supply and GFCI.
don
