P. Short wrote: The Renard24 and Lynx would probably be the best (i.e. simplest wiring) if they are installed outdoors (although the fan might be an issue with Lynx) and the lights are nearby.
Phil
Why would the fan be an issue on the Lynx outside Phil? I know of lots of equipment with heatsinks and fans outside in cases.
Mostly personal bias. I view mechanical items like fans as inherently unreliable, and (assuming that it is truly required) dislike putting them in an unattended, unmonitored situation. Also, to be effective, the enclosure probably requires some sort of venting, which would allow Florida lizards (or snails or slugs, lol) to come in contact with high voltage unless care is taken to keep the openings small or screened off.
But mostly just personal bias.
--
Phil
I like the idea of the small-channel count with integrated SSR controller, but I think that a daisy-chain of so many controllers could easily be just as messy in wiring as other setups. I would go for it if you could use a chain, star, or some kind of hybrid topology that used a bus with all units attached to it. The computer would broadcast data on the bus and units would be assigned individual addresses. For example, standard 2 or 4-conductor phone wire could be used for data, and you could use the same equipment you would use for a telephone system (line cords, splitters, those punch-down wire things, etc.).
If the communication protocol were technologically feasible, it would allow flexibility without adding much complexity, and the wiring would be cheap.
-Matt
With DMX you could do that i think because each device get its own id.
I think a 4 channel Renard with onboard ssr's, in something like a 4 outlet box or pod ssr enclosure, would be perfect for long series' of 3-4 color mini trees. Just daisy chain them all together. And if they're all led, then you could do the same with the power cord, too.
What might the benefit be of separately addressable controllers? Each one would still need to connect to the computer, either directly or via the previous controller. I guess that would be necessary for power line control or wireless.
To really make Christmas lights easy we either need to come hub "system" that doesn't cost much or go wireless. Honestly, I'm looking at both. After 2008 lights, my goal will be to go wireless. I don't really care if it adds $20-$30 to a box, that is what I want to do. By going wireless, the only issue left is running power...and that is pretty easy to distribute (you can splice a ton of things together). Right because I am wired I need to punch a hole through the wall between our house and garage. My wife doesn't like that one bit.
Frankz wrote:What might the benefit be of separately addressable controllers?
answered it myself. separate addresses would allow for any number of multiple topologies, even intermixed with one another, without having to be in series.
What does a 1000' spool of 2 or 4 conductor wire run? and then the subsequent connectors, plugs, jacks, for it? compared to cat5?
I don't see much cost savings there, if any. I think the availability of cat5 cable and connectors makes it comparable in price.
There are differences in price, although price is not everything.
Suppose you have a cluster of twelve mini-trees about 100 feet from your house, each using four channels. If the controller is in the house and the SSRs are by the mini-trees, you might end up using 1200 feet of cable (12 cables times 100' each). In the other design that I'm talking about here that would end up as 100-200' of cable (call it 100' to reach the cluster, and 100' to inter-connect them). In my opinion this is a much cleaner setup, and possibly less expensive.
--
Phil
This idea is practically being done now by alot of folks. Alot of the multi-color mini-trees out there have the SSR boxes under them now and CAT5 running to the controller.
This new device definitely would cut down on the amount of CAT5 running back to the house (controller). It would be particularly useful for a row or "forest" of mini-trees.
You would only have one data and one main power cord coming from the house - Nice!
If you could keep the board size down then you could always mount them in the 4" PVC housings that some folks are building now.
The main issue I see would be that if you build for the worst case scenario, then you would be limited to about 10-12 devices off of each main power cord. 4 channels each with 100 minis would be somewhere around 1.2 to 1.5 amps for each unit.
Of course the current setups have the same issues but I just wanted to mention it so that any newbs reading wouldn't get to thinking that they could hook 20-30 of these onto a single main power.
If you are going to also be using the AC to create the DC power for each board then you going to reduce the total number of devices per main power cord you can do also. Probably only lose 1 but I haven't done the math.
So the next question(s) would be...
What to do if you need more devices than 1 main power cord can handle?
Do you run a second main power cord and just continue to daisy-chain the data signal?
or
Do you run a second main power and a second data line?
or
Just dependent on the layout?
Then of course which protocol to use... but that can wait til later in the developement (if it gets that far).
RavingLunatic wrote:This idea is practically being done now by alot of folks. Alot of the multi-color mini-trees out there have the SSR boxes under them now and CAT5 running to the controller.This new device definitely would cut down on the amount of CAT5 running back to the house (controller). It would be particularly useful for a row or "forest" of mini-trees.
You would only have one data and one main power cord coming from the house - Nice!
If you could keep the board size down then you could always mount them in the 4" PVC housings that some folks are building now.
Right now I have a PCB design that is 1.7" x 2.4", smaller than the current SSR boards. One limitation of this design is that I'm using Z0103 triacs (TO92 package) , which are only good for 1A each. It would not be a big deal to use a larger part.
The main issue I see would be that if you build for the worst case scenario, then you would be limited to about 10-12 devices off of each main power cord. 4 channels each with 100 minis would be somewhere around 1.2 to 1.5 amps for each unit.
Of course the current setups have the same issues but I just wanted to mention it so that any newbs reading wouldn't get to thinking that they could hook 20-30 of these onto a single main power.
If you are going to also be using the AC to create the DC power for each board then you going to reduce the total number of devices per main power cord you can do also. Probably only lose 1 but I haven't done the math.
What I'm thinking of doing is sending low voltage power down the same two wires as the data signal. What I would use for the transmitter is a L293 H-Bridge chip, intended for motor control, which is good for 1A. The receiver circuitry on the controller would use a full-wave bridge rectifier (and cap and zener) to get DC power, and would look at the polarity of the incoming signals for serial data. The model railroading DCC standard uses a somewhat similar scheme.
So the next question(s) would be...
What to do if you need more devices than 1 main power cord can handle?
Do you run a second main power cord and just continue to daisy-chain the data signal?
or
Do you run a second main power and a second data line?
or
Just dependent on the layout?
All of the above. The data rate is slow enough that you don't really want more than maybe 48 channels on a given data line anyway.
Then of course which protocol to use... but that can wait til later in the developement (if it gets that far).
Edit...although this is much more detailed than I wanted to get in this thread.
--
Phil

