I can't believe that I missed this thread earlier. I've been anxiously awaiting to hear more about omzig's progress on the Renard Servo Controller. It's looking pretty good thus far.
One suggestion I have is to connect the 7/8 pins on both of the IN and OUT RJ45 sockets together. Even though this board will not be using Zero Crossing, it would be nice to pass it along the line. This way you can connect other light controllers down-stream and still be able to dim lights after it.
ppohlman wrote:One suggestion I have is to connect the 7/8 pins on both of the IN and OUT RJ45 sockets together. Even though this board will not be using Zero Crossing, it would be nice to pass it along the line. This way you can connect other light controllers down-stream and still be able to dim lights after it.
Good point. I will do it.
Thanks for the suggestions for handling the channel configuration. I've thought a lot about the best way to do this, and this is what I've come up with:
- All servos will start out disabled. Once any data is received for a channel, that channel will be enabled and its control pulses will start.
- If the very first byte of data received for a channel is 0xFE (254) that channel will be set up in the extended mode (and the 0xFE will be discarded). If the first byte is any other character, the channel will be in the standard mode. Once in the extended mode, a channel will remain extended until the controller is reset.
I think that this is the simplest approach, but it is still effective. If you aren't using any extended range servos, the you don't have to do anything as far as configuration is concerned--just start sending data.
In the unlikely event that you would want a 254 for your first position in the standard mode, you could use a 253 or 255 first, and then 254. Being off by 1 for a duration of one Vixen timing interval is hardly going to be catastrophic. If you needed you first position to be 254 in the extended mode, you could just use two consecutive 254's.
Comments?
The more I think about it, I don't like the idea of stretching the Renard protocol to handle the extended mode configuration. I would rather do it with jumpers on the board. This requires more pins though, so I'm looking at using the 20-pin 16F687 instead of the 16F688. The new schematic is attached.
There are ways of adding jumpers that do not require extra pins, basically by re-using existing pins. For example, you could add a jumper to ground on an output pin that would normally be pulled up. At power-up you tri-state the output and use it as an input. If you detect that the pin is tied to ground, do the special stuff, and if you detect that it is not shorted to ground, re-configure the pin as an output for normal use.
P. Short wrote:There are ways of adding jumpers that do not require extra pins, basically by re-using existing pins. For example, you could add a jumper to ground on an output pin that would normally be pulled up. At power-up you tri-state the output and use it as an input. If you detect that the pin is tied to ground, do the special stuff, and if you detect that it is not shorted to ground, re-configure the pin as an output for normal use.
Interesting...But if the jumper is in place, how do you then use it as an output if it is shorted to ground?
I suppose really you the ideal option would be to use the "command" protocol options phil built into the protocol. (remember the double 127 byte issue)
the biggest issue with that is the current plugin's dont support them.
I just keep coming back to thinking "jumper's for configuration is just so 90's" 🙂
anyway. just another idea. at the end of the day it is your design and you need to do what you feel happy with. if other dont like what you do they just wont use it. either way you still get a controller that works for you.
scorpia wrote:I just keep coming back to thinking "jumper's for configuration is just so 90's" 🙂
LOL...Yeah well I'm kind of an old-fasioned kind of guy.
scorpia wrote:
anyway. just another idea. at the end of the day it is your design and you need to do what you feel happy with. if other dont like what you do they just wont use it. either way you still get a controller that works for you.
Yeah I know. But I still value the opinion of the community. My reason for doing the board is mostly for myself, because I have an idea for using servos this year. I know Frank said he used DMX for this, but since I don't have any other DMX stuff (that is, unless I get around to building the Pixel boards that I bought), and there is going to be a Ren24 right by where I need the servos, it just seems the easiest way to go, especially since it's >100' from the computer. Not to mention that I'll have a ton of fun doing it (probably the biggest reason). If someone else has a use for the board, then that's great too. Actually, a couple of people have already expressed interest.
I found this last week and have built the circuit included. It seems to work as advertised.
The only input needed is servo number (1-8) and position. IE: 1,175
http://www.rentron.com/SerialServo.htm
My question is could this be controlled by Vixen. It seems to me only a plug-in would be needed to convert the channel number to the servo number and the dimming level to the range (20-220). Or am I missing something.
WireNut wrote:I found this last week and have built the circuit included. It seems to work as advertised.
The only input needed is servo number (1-8) and position. IE: 1,175
http://www.rentron.com/SerialServo.htmMy question is could this be controlled by Vixen. It seems to me only a plug-in would be needed to convert the channel number to the servo number and the dimming level to the range (20-220). Or am I missing something.
Sure a plugin could be written to do that. A plugin could be written for just about any serial servo controller. My design is trying to make a controller that works within any Renard chain, and doesn't require another interface to the host or any software changes. That is the topic of this thread.
