Lutron Homeworks oddity

ericmedley

Hey all,

I have several projects using a Homeworks master. I have this oddity that is puzzling. It is not causing any program problems but it is occuring pretty regularly. When you monitor the serial port connected to the Lutron processor, I get regular messages that consist of a massive string of $FFs. It happens about every 2-3 seconds. If I watch the Diagnostic messages from the master it shows an error "String too long - truncated to 274bytes."

Everyone of the projects that I have a Lutron master connected does this. Lutron acknowleges it but doesn know what its from. As I said, it is not causing me any programming problems.

I have this pet theory that since $FF = 1111 1111 binary, that I'm actually dealing with a voltage spike. I'm thinking that I'm getting some kind of positive voltage bias on the RX pin and the AMX serial port is inturpreting it as a string of $FFs.

Has anyone encountered something like this with Lutron? (non-amx version)

alexanbo

I haven't noticed anything like that on my Lutron processors. Are you just using pins 2,3,5 and have the handshaking turned off on the Lutron side?

ericmedley

alexanbo wrote:

I haven't noticed anything like that on my Lutron processors. Are you just using pins 2,3,5 and have the handshaking turned off on the Lutron side?

Yes,
we're only using 2,3,5 and no handshaking. I am running them at 115K baud. I wondered if that might also be one of the problems.

Like I said, it's working fine. there's just this mystery string coming in.


Thanks for the response.
ejm

alexanbo

Yeah for what it's worth I run mine at 9600.

RicardoSiqueira

I remender trying all the baud rates with Lutron while ago and the 38400 was our sweet spot. The highier rates would have some feedback problems. I don't know if it is still the same, but try lower speed and see if it works better for you.

BCalderwood

Just a thought and I may be way off here, I'm using the Homeworks module on an NXI w/ ME260. I know it's older but no strings like what you have stated come back on that and I'm using 115200 baud. I know that the newer NI series don't like 115200 on port 1 but you already may know that. What port # are you using and what series of NI? I've notice some weird things with the serial ports on the newer NI series controllers. Nothing to kill running code but some things that seem weird. Going with a slower baud rate will probably help but that's been stated already in this post so I sound like a repeat.