Interfacing: Caddx > Elk

[blafarm]

For reasons too lengthy to describe, I need to "forward" the status of every zone of a Caddx NX-8e to an Elk-M1Gold. I've got nearly 100 zones and I know I can max out the Caddx with Relay Expanders (giving me only a maximum of 64 outputs) -- but I'm wondering if there is any other possible way to interface the Caddx to the Elk that is more elegant and not so limiting. I believe there are serial protocols for interfacing a Caddx to a Creston -- but I don't know if I can take advantage of them.

Thanks in advance.

 

[BraveSirRobbin]

hmmm, question for ya...do you want to do this without a computer involved?

 

[blafarm]

Hi BSR,

Yes, I'd really like to "hardwire" and eliminate any potential for failure -- however, if there is no other way, I'd certainly entertain it.

 

[BraveSirRobbin]

Well the relay board would probably do the trick (Caddx Relay outputs turn off and on based on Caddx input zone values, then that relay is tied across an Elk input zone). But, I didn't realize you had a limited number of relays you can place on an NX8e board.

I don't think you can just tie the two inputs (of the Caddx and Elk) together as they are open collector (I think I tried something like this to get an Ocelot/SECU16I interfaced to an NX8e and it did not work). Maybe Spanky can give some insight on this one (can you tie both inputs maybe to the same EOL reisistor that is terminated back at the panel???)

 

[blafarm]

But, I didn't realize you had a limited number of relays you can place on an NX8e board.

Yes, the NX8e is limited to 8 Relay Expanders (NX-507E) -- each of which have 7-Outputs and 1-Open Collector (which, using a relay, could be converted into an 8th output -- thereby providing either 56 or 64 outputs. Not really enough for the application.

I don't think you can just tie the two inputs (of the Caddx and Elk) together as they are open collector (I think I tried something like this to get an Ocelot/SECU16I interfaced to an NX8e and it did not work).

Yes, I agree -- not a good idea -- even if it were possible.

Maybe Spanky can give some insight on this one

I look forward to his comments

 

[hucker]

I know that you can setup the nx8 to do zone reporting over the serial port (that is how my hvpro gets status). Could you connect NX8 RS232 to the M1 and decode the zone status in the M1?

 

[Steve]

I guess I'm just not getting why you can't just go straight to the Elk (I know you don't want to describe the whole thing). But assuming you still need the Caddx in place, would it be possible to just move the input zones to the Elk? If you still need the Caddx to sound the alarm then you could have the Caddx monitor the Elk - like when any zone in the Elk opens, turn on an ouput which could be seen by the Caddx to react.

Since the Elk can do everything the Caddx can, plus more I don't see why you can't just to a direct replacement, but maybe the above can work???

 

[BraveSirRobbin]

I know that you can setup the nx8 to do zone reporting over the serial port (that is how my hvpro gets status). Could you connect NX8 RS232 to the M1 and decode the zone status in the M1?

This would be interesting. Have the Elk look at the Caddx's serial port and then act on certain ASCII strings. You would have to write a bunch of rules and probably only toggle flags/variables (I don't think you can control the input state of a zone via rules).

Blafarm, when you say "forward" are you saying you want each of your 100 zones to act like input zones to an Elk M1? In other words would these Elk zones would show up as regular inputs (would you also then have to purchase expansion input cards via the Elk?). This could get expensive. Hopefully Spanky can give some input on the proper direction to take.

 

[mustangcoupe]

This is not a real soloution but if your zones are normally closed you might beable to run the zone wires through a relay coil and trigger the elk of the ouput of said relay...


[codebox]zone---------------------------cadex Z1
|
--------relay coil--------------cadex Z1 return
common NC
| |
| ----resistor--------Elk Z1
-------------------------------Elk Z1 return[/codebox]

 

[upstatemike]

This is not a real soloution but if your zones are normally closed you might beable to run the zone wires through a relay coil and trigger the elk of the ouput of said relay...


[codebox]zone---------------------------cadex Z1
|
--------relay coil--------------cadex Z1 return
common NC
| |
| ----resistor--------Elk Z1
-------------------------------Elk Z1 return[/codebox]

Change relay coil to opto-coupler and this might be the best way to go.

 

[Spanky]

It would be far easier moving the input zones to the M1 rather than trying to cobble together relays to trigger the M1's inputs. You will need 5 or 6 input expanders at 16 zones per expander, but the cost of relays would offset that expense. Check to see if the EOL Resistors are being used at the end of each wire run or are they put on the terminal blocks of the Caddx control. If the EOL Resistors are not being used, you can program them out of the M1 by using a normally closed zone input definition.


The binary protocol of the Caddx bus cannot be monitored by the M1 for matching data.

 

[BraveSirRobbin]

Thanks for the reply Spanky. FYI, there is an "ASCII" option for the Caddx serial port when setting it up (vs. binary).

Also, if you are going through all the trouble of having a contact sensor trigger a relay for each zone, why not just use a double pole (single throw) relay and have one pole go to the elk, and one to the caddx? This sounds like the simplest (and I use that term very loosely here) way to do this! (I actually like this method the more I think about it). This would be an involved hardware wiring project (considering the 100 zones) but could be managed. Also, it would eliminate a lot of software/programming problems (i.e. where do you want your pain, in the hardware or software).

Just some suggestions...

 

[mustangcoupe]

Thanks for the reply Spanky. FYI, there is an "ASCII" option for the Caddx serial port when setting it up (vs. binary).

Also, if you are going through all the trouble of having a contact sensor trigger a relay for each zone, why not just use a double pole (single throw) relay and have one pole go to the elk, and one to the caddx? This sounds like the simplest (and I use that term very loosely here) way to do this! (I actually like this method the more I think about it). This would be an involved hardware wiring project (considering the 100 zones) but could be managed. Also, it would eliminate a lot of software/programming problems (i.e. where do you want your pain, in the hardware or software).

Just some suggestions...

Actually I thought about this option as well when making my recomendations earlier. But I figured you then would need an external power supply for the relays. With my option above I figured you might beable to use the zone voltage to power the relay (or opto isolator.)

with your option BSR you would need something like this....

[codebox]
zone---------------------------Power supply Positive
|
|
---------Relay Coil------------Power Supply Negative
| | | |
| | | |
| | | |
| | | ---------------------Cadex Zone +
| | |
| | ------------------------Cadex Zone -
| |
| | ---------------------Elk Zone +
|
--------------------------Elk Zone -


[/codebox]

You could then parallel the zone/relays so you run them off one power supply.

 

[BraveSirRobbin]

Right, my method has the advantage of taking the relay coil resistance out of the station's "loop" but the disadvantage that you would now not have supervised inputs (since they would have to be at the panel/relay itself, no sense in even doing that).

Also, if there were EOL's at the end of the line for the sensors, you may have to pick a relay with less than 12 volts for the coil spec.

You can use the power supply from either system to operate the relays as the coil is now separate from both inputs.

 

[blafarm]

Thank you all for these suggestions.

Yes, BSR, I basically need the state of every input zone of the Caddx to be effectively "replicated" on the input zones of the Elk.

As much as I would like the Caddx to be downstream of the Elk, it's really not advisable given their respective functions (Caddx=Security - Elk=Automation). Frankly, there's no reason for the Caddx in this equation -- it's just an unfortunate requirement of this situation.

The relay idea is interesting, but with it comes with an inherently higher chance of failure (for both panels) and it presents EOL issues.

It would be cool if the "ASCII" option for the Caddx serial port represented a potential solution -- but I recognize it's a longshot at best.