Snunn,
Good job on finding a better part for driving the outputs. Since the impedance is not know for the output, and we don't know the length of the run, we generally just put a small source termination resistor in line with the signal (22/33 ohms). Since you are running slow the 33 ohms would probably work fine.
If you look at the RS-232 cable length recommendations, you will see as you go up in frequency, that a shorter the recommended cable length. Since there are so many factors is hard to tell how far you can run the cables, so you will have to experiment. However, another trick is that you can run the signals as twisted pairs (like ethernet). Run the clock twisted with a ground, and the data twisted with the ground.
Another thought is that you could run your tree as one long string. Each of the WS2801 re-buffers the input to output. The maximum clock rate for the device is 25MHz. So you could feed a little faster clock to your tree string and then slow it down going longer distance to other strings.
http://www.analog.com/en/switchesmultiplexers/multiplexers-muxes/adg726/products/product.html
This could be a single chip fix for routing the clock and spi via a single chip. However, it would require a circuit board be designed which is out of my comfort zone.
snunn wrote: http://www.analog.com/en/switchesmultiplexers/multiplexers-muxes/adg726/products/product.htmlThis could be a single chip fix for routing the clock and spi via a single chip. However, it would require a circuit board be designed which is out of my comfort zone.
For development you could use this, the video is quite good.
mrpackethead wrote:For those that are not followign this, this project could be the killer proejct for 2013 in the DIYC space.
If I could understand what everyone is talking about!
I just got a (well...3) Raspberry Pi('s) to play with and I am totally lost.
^ +1
but it's interesting as heck!
