This is a continuation of a project that started of as a Renard Pixel String that would use the Renard protocol that can be found here. Essentially, the goals of the project remain the same, to control multiple RGB LEDs with a minimal amount of hardware. What has changed is the way I want to "talk" to these boards. Originally I was going to use Vixen (I may end up using it anyway) to control these boards and I could get my head wrapped around the Renard protocol pretty easy. However, as the project progressed the realization as to the amount of time it would take to hand encode all 4,008 channels that I would like to do this year just became overwhelming to say the least! I then came across MPH's solution that he came up with which I referenced in the above thread. Basically, using a program on the Macintosh called Quartz Composer I could control all 4,008 channels with a drastic reduction in programming time and even better use my Mac! I am a long time Mac user and the prospect of controlling my blinky-flashy with a Mac warms my heart.
To that end, the patch for Quartz Composer outputs the channel info via Art-Net which is an older method of sending DMX data over Ethernet. This of course would require the boards to be running on DMX instead of Renard hence the move of the project thread here. I posted a "first draft" of the DMX based board in the Renard Pixel String thread and big surprise I had some glaring errors. So attached is the updated and hopefully error free schematic. I am continuing to work on the firmware for the board as well as the PCB and will post these when I am finished.
P.S. Speaking of controlling blinky-flashy with a Mac a very exciting thread was started yesterday here that will hopefully result in a proper sequencer for Macs. Exciting times... exciting times!
Why are there two RS485 transceivers? Are you planning on regenerating DMX?
On the receive side I'm not sure what the story is with diodes and all the resistors, none of that is needed.
DynamoBen wrote:Why are there two RS485 transceivers? Are you planning on regenerating DMX?On the receive side I'm not sure what the story is with diodes and all the resistors, none of that is needed.
That is all from its Renard heritage - if going completely DMX then it's not needed. I'd suggest a DMX-THRU if possible though to aid in cabling.
I agree with budude. Provide a DMX THRU connection if it's DMX only. Typically, a DMX device doesn't modify the stream if there is a DMX out connection. That way you can daisy chain devices, each with their own address in the DMX universe, and you don't care in what order you place them.
Now, you do need to have a way to set the address of each Pixel String in order for this to work.
mschell wrote:I agree with budude. Provide a DMX THRU connection if it's DMX only. Typically, a DMX device doesn't modify the stream if there is a DMX out connection. That way you can daisy chain devices, each with their own address in the DMX universe, and you don't care in what order you place them.Now, you do need to have a way to set the address of each Pixel String in order for this to work.
Which could be done via the data stream (if you can convince yourself that there are certain patterns that you will NEVER send to the device).
Look at the modifed JEC Pixel firmware in the File Library. 🙂
