Trouble adding Omnistat 2000 to M1

[gcimmino]

I have an existing M1 install, running 5.1.20. It had the following serial devices:

0: XEP
1: XZW (1.0.16)
2: XSP (1.0.46) hooked to RCS thermostat (in bldg 2)
3: XSP (10.0.2) hooked to Uplink

I wanted to replace a non-integrated stat with an OmniStat 2000 in bldg 1. In the XSP docs, it seems as though an OmniStat with address 1 needs to connect to an XSP addressed as Serial #1. So I renumbered the XZW as serial 4 and put a new XSP with 70.0.2 firmware as serial #1. Result is:

0: XEP
1: XSP (70.0.2) hooked to Omnistat 2000 (in bldg 1)
2: XSP (1.0.46) hooked to RCS thermostat (in bldg 2)
3: XSP (10.0.2) hooked to Uplink
4: XZW (1.0.16)

The RCS, which is "TStat 1" in ELK RP continues to work just fine.

The new stat, which is "TStat 2" is flaky. I've seen the temps change under M1 control twice, but with many failures to change in between. I've been able to send text strings to the stat's display with the "Send Text" rule.

I'm looking via eKeypad on an iPad and I don't see any current temp values returned for the new stat.

70.0.2 was released on 27 Oct 2008 and 5.1.20 on 3 Jun 2009, so I'm presuming there's no issues with the M1 firmware level supporting the Ominstat 2000.

I've double checked the jumpers on both the XSP and the Ominstat as well as the Omnistat setup.

Any ideas on where to look?

 

[Del91574]

Verify JP5 and the cable has continuity on the white wire? The whites are coming "pre-cut" now as many people have been blowing up stuff by feeding 12V through that cable.

Also verify the baud rate and 232? Re-enroll the devices?

You can also verify the serial for functionality by loading the default software into it and running hyperterminal with a null cable to make sure it's working properly.

 

[ddennerline]

I didn't get my OmniStat2 working right off the bat either. This post may provides some troubleshooting steps. I have managed to get 1200 baud working afterwaiting a week. 2400 baud definitely fails to communicate. I believe distancemay play a factor. You will need a common "white" wire. My ancient thermostatwire was mess as the original house A/C installer clipped the common wire andleft inside wall. I slightly opened up wall and reran new 18g-7 wire/line back tothermostat.



http://www.cocoontec...=1

 

[ddennerline]

I have my XSP set up differently. I believe the OmniStat requires 4 identifiers to be reserved.


[font="'Courier New"]****SERIAL PORT EXANDERS (BUS TYPE 5) ****[/font]

[font="'Courier New"]1,M1XSP, 0.3, 1.0.1, 70.0.2, Firmware support HAI OmniStat[/font]

[font="'Courier New"]5,M1XSP, 0.3, 1.0.1, 20.0.4, Firmware support Leviton ViziaRF Z-Wave[/font]

 

[trmitche]

You do have the red and black wires running to the Omnistat2 back-plate "swapped" right? In other words, the red wire coming from the XSP should connect to the terminal labeled "Black" on the back-plate, and the black wire should connect to the terminal labeled "N/C", with no jumper between the two. It's in the XSP instructions, but it's easy to miss. Having these connections reversed (which would be intuitive), you can sometimes achieve partial communication at a lower baud rate, but having them correctly wired (assuming no other problems) should allow you to set the t-stat to 2400 baud and achieve full and constant communication.

Also, it's mentioned in the thread that is linked to above, and you said you checked the jumpers on the Omnistat in your original post, but just to be sure: You do have the "Comm Jumper" (JP8, bottom left corner of the t-stat circuit board) in the OFF position, right? It comes from the factory in the ON position, and if left on, it will prevent reliable communication.

Hope this helps.

 

[gcimmino]

Del, d.denerline, Trey, thank you all for the responses.

I think I have the setup right, double checked. J8 on the stat was definitely removed. I don't remember removing the N/C-Black jumper, but double checked and there's none there.

My XSP is in the same can as the M1 and the OMNI is about 10 feet of Cat5 cable away. The XSP came with the Uplink black connector, which I cut off.

I wired as follows (used both wires of a cat 5 color pair twisted together):

XSP Cable -> Cat5 -> Omni contacts
Black (DB-9 pin 5 - GND) -> Brown pair -> N/C (GND)
Red (DB-9 pin 2 - RCV) -> Blue pair -> Black (RXD)
Green (DB-9 pin 3 - XMT) -> Green pair -> Green (TXD)
White (DB-9 pin 4 - DTR) -> Orange pair -> Yellow (DTR)

I looked up the DB-9 pinouts and Omni stat pinouts (from protocol book at http://kb.homeauto.com/redirfile.asp?id=482&SID= )and included that info above mostly to make sure it made sense to myself.

I believe that's correct. I'm going to test the cable continuity from the DB-9 male to the Omni contacts tomorrow. If I'm not seeing data back from the OMNI, then I'll focus on the DTR (supplies power to the OMNI to transmit), and the TXD wires.

Can anyone else confirm the need to relocate all other serial devices to start at Elk Bus address 5 when using the OMNI firmware load? I'm not getting that from the XSP doc. The doc says "If M1XSPs are in use with other devices they will have to be changed to addresses other than 1 to 4.", but that's vague and might mean if one is using more than one XSPs for Omni devices.

Thanks again for the help.

 

[chedemefedeme]

May want to check the quality of your cabling run to the tstat and that it isnt too long. I've noticed that a bad connector on mind causes me occasional misses in communication. If i reseat it i'm good for a while again. Have also found a total reboot of the Elk is required every 2 to 6 weeks when it will just kindof stop communicating reliably.

 

[happy_again]

Del, d.denerline, Trey, thank you all for the responses.

I think I have the setup right, double checked. J8 on the stat was definitely removed. I don't remember removing the N/C-Black jumper, but double checked and there's none there.

My XSP is in the same can as the M1 and the OMNI is about 10 feet of Cat5 cable away. The XSP came with the Uplink black connector, which I cut off.

I wired as follows (used both wires of a cat 5 color pair twisted together):

XSP Cable -> Cat5 -> Omni contacts
Black (DB-9 pin 5 - GND) -> Brown pair -> N/C (GND)
Red (DB-9 pin 2 - RCV) -> Blue pair -> Black (RXD)
Green (DB-9 pin 3 - XMT) -> Green pair -> Green (TXD)
White (DB-9 pin 4 - DTR) -> Orange pair -> Yellow (DTR)

I looked up the DB-9 pinouts and Omni stat pinouts (from protocol book at http://kb.homeauto.com/redirfile.asp?id=482&SID= )and included that info above mostly to make sure it made sense to myself.

I believe that's correct. I'm going to test the cable continuity from the DB-9 male to the Omni contacts tomorrow. If I'm not seeing data back from the OMNI, then I'll focus on the DTR (supplies power to the OMNI to transmit), and the TXD wires.

Can anyone else confirm the need to relocate all other serial devices to start at Elk Bus address 5 when using the OMNI firmware load? I'm not getting that from the XSP doc. The doc says "If M1XSPs are in use with other devices they will have to be changed to addresses other than 1 to 4.", but that's vague and might mean if one is using more than one XSPs for Omni devices.

Thanks again for the help.

I believe moving all other XSP's to start at an Elk Bus Address of 5 would only be needed if you had greater than 12 thermostats in place, but that is just my opinion.

One thing you may want to try, if you haven't as of yet, would be to temporarily disconnect the other XSP's from the M1; program the Omnistat XSP to an address of 1, and try to communicate to the thermostat via that XSP only. I have a Omnistat-2 RC-1000 reliably 'talking' to my M1 system at a 2400 baud rate(setting both baud rates in the Omnitstat to 2400). My system only has one XSP and my cable length between the M1 and Omnistat is roughly 15-20 feet.

I read that you had the RC-2000. I don't believe the M1 passes the humidity info through the XSP. I don't know if that is the issue. Does anyone have a RC-2000 Omnistat-2 successfully communicating to their M1?

Hope this helps...

 

[gcimmino]

Fixed. Didn't change a thing wiring wise.

What I did was to renumber the "network address" of the existing RCS thermostat from 1 to 2.

It looks like the thermostat addresses need to be globally unique. Since they were hanging off of different XSPs, I hadn't thought there would be a conflict with both the RCS TR40 and the Oministat both as address #1 on their respective XSPs.

The final configuration is:

0: XEP
1: XSP (70.0.2) hooked to Omnistat 2000 - set to address 1 (in bldg 1)
2: XSP (1.0.46) hooked to RCS thermostat - set to network address 2 (in bldg 2)
3: XSP (10.0.2) hooked to Uplink
4: XZW (1.0.16)

In the computer world, I'm used to multiple disk drives/devices with the same unit ID, but differentiated by being on different interfaces.

Thanks to all who responded.