even at a 100 channel limit it still cuts the amount of cable runs from inside the garage by a factor of 25.
I'm not sure what RL is proposing, but it is probably almost the same as what I had in mind.
The easiest choice is to use an 8-channel controller (and firmware) and leave off the extra four channels. Another possibility would be to customize the firmware for exactly this board. The extra pins on the PIC could be sued for a crystal and possibly some method of setting the DMX address. The benefit of the crystal is that it is a lot cheaper than the oscillator used on the mainstream Renard design, and probably uses less power as well. While I am no fan of DIP switches, this might be a place where they are useful There should be room on the board for a DIP switch to allow the address to be set by the user. This would take less programming than adding code to allow it to be changed through the DMX channel and would probably be easier to use in the field. The disadvantage of all of this, though, is that it is a new code image that would have to be created (not too hard) and maintained (a PITB IMO).
Brian,
Do you have an enclosure scheme in mind?
Just want it to fit in my current ssr enclosures.
P. Short wrote:I'm not sure what RL is proposing, but it is probably almost the same as what I had in mind.
Basically, it boils down to a Ren8 with on-board voltage regulator and I would put a header for reprogramming on board. Also jumpers to allow for either Renard comm out or pass-thru of DMX.
This approach (IMO) probably would be the quickest to get done. A known design that would be able to control 2 existing SSRs, 1 in the same enclosure and another via CAT5 cable (if desired).
Creating a new 4 channel only would mean each SSR would have to have a board, which means (probably) more costly. Also would have a longer development timetable due to new code requirements.
Just some thoughts....
On my bench I have an eight channel DMX control board that might fit the bill. A couple of months back I had the same thought. After rereading an old nuts and volts article about a dimmer based on the Parallax SX I decided to see what I could come up with. ("Dimming The Lights Fantastic" by Jon Williams)
While the 8 channel dimmer in the article was good it wasn't DMX controlled. So instead of just writting DMX code for it I decided I wanted to create a DMX controller that was like a swiss army knife. I would allow me controll SSRs, relays, LEDs, Servos, Stepper motors, or anything else I could dream up. The board runs off of any external power supply, has a zero-cross opto onboard and I plan to use either dipswitches or and LCD/push-buttons for addressing.
I have the dimmer and DMX code finished and working. The PCB is partially laid out in ExpressPCB. I haven't started the LCD code yet so I'm not 100% sure if that will work but I do have the dipswitch code ready as my fall back. The DMX interface is wired for bi-directional half-duplex communications. If I ever get around to playing with RDM.
The schematic is at the bottom of the page (the current version hasn't changed a lot):
http://doityourselfchristmas.com/forums/showthread.php?t=3227&page=2
I kept the Parallax SX because it is easy to learn, and geared toward hobbyists. It has proven to be a very fast add capable micro, and at $2.79 its hard to beat.
http://www.parallax.com/tabid/248/Default.aspx
There seem to be a lot of options. Let's see which one(s) will be completed.
Just FYI, i am working on a DMX 8 Channel Dimmer/SSR. but its not for Mains voltage, its for low voltage, which is what we use down here...( Aussie/NZ ).. Its going to use DC, at 12/24/36V.. Dimming is by 8 bit PWM on a FET, with active over-current shutdown..
Am doing this for myself mainly, cause there was'nt anything i coudl find that quite met my needs.. If its useful to others great..
Well it seems as though there are a couple of options here, DMX and Renard. I am wondering which will be the better of the 2. Phil and I spoke a little about it last night in the chat. DMX is a standard and can handle other types of off the shelf devices. Renard is proven and used by lots of us. The ability to interface with other types of devices is a big benefit for DMX. The fact Renard is proven and currently in use by lots of us is a big pro for it. Now a question I have before I go with one or the other is about power. Each ssr has 110v power to run the lights. Will this make things easier to power the controller side or will power still need to be run through the control cabling, preferably cat 5. I am concerned with voltage drop in regards to the small conductor size and distances. 24 awg conductors with a 1 amp load will loose .053 volts per foot of cable, for 6 or 12vdc or single phase ac. This would be a significant limitation for cabling distances. This is where the 110vac supplied at each location could come in handy. Another concern is the signal degradation over long cabling distances. Currently I am sending only voltage to trigger the ssrs over a max of 120' linear cable lengths. The data signals may suffer at distances like that.
Just some thoughts to ponder.
FYI, I am looking at this for next year, not this year. That gives lots of time for development and testing.
"The fact Renard is proven and currently in use by lots of us is a big pro for it."
While the DMX dimmers made here may not be "proven" the DMX protocol has been around for twenty years. The protocol is very robust and it a good match for the type of stuff we are doing.
As far as data distance DMX (which is actually RS485) will give you about 1500'. You are allowed 32 devices on a single run, which would give you 256 channels (there are ways around this limit).
Currently you are making a home run from your SSR to the house. With DMX you will daisy chain each dimmer to the next.
The dimmer I've got would typically use a power supply locally. Running power down a DMX data cable is never suggested.
