I’ve got a T3-TB unit that I’m trying to determine its’ viability as a replacement for a ‘standard’ BACnet router (CC BASRT-B). I’ve got two BACnet/MSTP devices installed in my test network: a Dent Power Meter and a Badger Ultrasonic flow meter. When I connect this trunk to the T3-TB module, it does not discover either unit. If I take the network trunk and plug it into my BASRT-B router, all of the devices are discovered using the internal Yabe tool:
I believe that I have my TB configured properly, and the ‘Network Health’ shows a ton of TX/RX traffic with no errors, but no devices are discovered in the T3000 tree under the TB, and no devices are visible when running Yabe natively.
Our team is out for the local holiday till Feb5 but there’s a few things I can suggest and comment on here;
At Tab1, this looks like a UDP ethernet network scan of bacnet devices using Yabe. However in your troubleshooting description above you are mentioning RS485 subnet scanning. Could use some clarification on that please.
Tab2: I see a T3-22i set up directly on the ethernet network. Normally you would set up expansion IO under one of the T3 controllers using RS485, that way you can directly use the intrinsic commands like VALVE101 = PID1 rather than having to integrate as external bacnet objects: 44434AO4 = PID1 for example.
Tab3 shows a scan at the office here, all temco subnet devices show up in a sub-tree below the associated T3 controller. Note that T3000 is not well set up to discover 3rd party devices like Yabe does.
Tab4: One small thing I notice is the gateway should probably be 192.168.86.1, this doesnt affect your ability to discover devices though.
Tab5: Are you able to contact any of your subnet devices using the communications settings showing there, by direct cable connection to your PC for example. This setting needs to match your network devices exactly.
Tab6: You can try putting in a test program in the T3 controller to read a few IO from your 3rd party devices into local variables and watch what happens here in the network points dialog.
Tab7: Each device on the subnet requires a unique address, including the T3 controller itself. Make sure the panel ID is set to a unique number.
Thanks for the reply - I’ll try to reply by topic…
Tab1: Yabe detects all BACnet devices on my network - the RS485 devices on the trunk while it’s connected to the BASRT-B router are exposed just like the Native BACnet/IP devices are shown (the router is at 192.168.86.27 in this image below:
My goal with the T3-TB is to determine if I can connect a BACnet MSTP trunk to it and communicate/control the RS485 devices via BACnet/IP, like I can with a standard router like the BASRT-B unit.
Tab2: The other Temco controllers are just bench samples - I won’t be using them to implement any programming (they’re not associated with this issue)
Tab5: If I connect the MSTP trunk to my computer, I can see the devices using a Yabe scan of the serial port. If I connect that trunk to the T3-TB, I can’t see any devices when I scan the IP network.
I’m assuming, based on my tests that this unit doesn’t function as I would expect, like any other BACnet/IP to MSTP router.
I can confirm that the T3-TB functions as an IP-to-MSTP router. It transparently passes BACnet traffic to BACnet instances on its subnets, allowing you to refer to subnet BACnet objects as if they were on the IP network.
It seems that Yabe/BASRT-B takes this a step further by displaying the subnet devices in its tree view as if they were ‘IP’ devices. We can work on implementing similar functionality once everyone returns from the local holiday.
I’m sorry for my shortsightedness - I use Yabe as my first ‘go-to’ to test for BACnet functionality and when the T3-TB didn’t reply with any downstream MSTP devices, I assumed that it wasn’t passing the WhoIs command or the Iam responses. When I use my Tridium Jace to query the BACnet devices on my network, the T3-NB shows up on my network, but not the two devices on its’ MSTP trunk. If I connect the trunk to my BASRT-B, the devices on my trunk show up, and I can map them into the Niagara database, as expected. Also, if I connect my trunk to the Jace’s Com port, the devices show up in a scan and I can map them in. They don’t discover at all when the trunk is only connected to the T3-TB and do discover in all of the other scenarios.
We will add some features to get the subnet devices to show up better using Jace and Yabe. Stand by for good news in a week or so. In the mean time you can still do some integrating using the bacnet instance and object ID. The data will get through.
Aren’t the two devices on its MSTP trunk from Temco? If not, they can still be displayed normally. However, due to factors such as the command response speed, network command latency, and Yabe’s logic for processing sub-devices and master devices, there will eventually be differences in the display on Yabe. At the same time, issues such as network packet loss cannot be ruled out. The following picture shows you the display result of Yabe under normal circumstances.
No - none of the devices on the TB’s MSTP trunk are Temco devices and none of them respond to a BACnet/IP discovery using Yabe, Tridium Niagara (Supervisor & Jace8000) or an Obvius AcquiSuite gateway. If I move the MSTP trunk to a BASRT router, all devices are discovered on all four systems, as well as by the T3000 software. This proves that there is no ‘command response speed, network command latency or Yabe processing logic issues’. If I connect the MSTP trunk directly to the Jace, I can also discover the devices, as expected. It’s only the TB that doesn’t pass the discovery request. If you can’t use the BACnet discovery tools for integration, it’s not very useful as a gateway. (BTW, I have used the TB as Modbus/RTU gateway without any issues, but those devices have to be manually programmed anyway.)
Fandu: Put some non temco devices on the subnet, they should show up in both the Yabe and the T3000 trees. I beleive that is what jon needs to see.
Jon: For your info, the T3 controller Fandu is testing with is functioning like a T3-nano, they have the same routing capabilities and share the same firmware.
We have a BASRT-B and several 3rd party subnet devices, we will connect them up and make sure both the BASRT and our T3 devices can show the subnet devices equally well. My screen shots show some testing, I need to add 3rd party subnet devices. Stand by for more results shortly.
Important Note: The T3 devices indeed act as a router, packets pass to and from the subnets transparently with no programming required. And this includes temco devices and 3rd party devices. Its just the tree view we want to show better.
Thanks for the continuing effort here - it’s not just the visibility on the Yabe tree that is important: I need for the devices downstream of the TB to respond to a discovery command issued from the network level so that I can integrate any of the BACnet objects just as I can when they are connected to a BASRT-B or other router. I have a few open projects with this need and would love to be able to use the TB instead of a BASRT-B if the functionality was equivalent.
The devices work as you describe but as Fandu says its a bit complicated with the timing. Devices on the subnet and devices on the main ethernet lan are all responding to the whois command at various time stamps. Yabe and T3000 have to make sense of all that traffic and display it logically. We will get it done.
Just re-emphaising again: This is separate from the router functions, our T3 controllers do indeed work as routers from IP to MSTP and back. Just like the BasRT-B
I’m not sure how you can state that it functions just like the BASRT-B when it doesn’t when connected to a network. When I connect my meter to the BASRT-B and perform a discovery via BACNet/IP, all of the devices’ objects are displayed in both Yabe and Tridium. When I connect the trunk to the TB, no objects are displayed using either system. Only the TB’s native objects are displayed, none of the objects on the MSTP are displayed.
If you were to get a wireshark capture going you would see the packets better. That said, we will get you going. Fandu has all the tools he needs to demonstrate the flow of data and eventually a nice looking tree. More news soon.
Can you confirm the MSTP MAC ID of TB and the two sub-devices? TB automatically determines whether it is master. Only the one with the smallest ID can proactively send who-is and become a BIP-TO-MSTP router. Therefore, it is important to set TB to have a minimum MSTP ID. In addition, when your MSTP device is connected to the RS485 network of TB, do not connect to the Ethernet at the same time, so that software will recognize it as a network device.
Chelsea will work on this,any panel ID should be able to handle the routing without special configuring like that. Its just a matter of smartening up our firmware a bit which we will do.
