T3 and T3000 Bacnet limits

[clyde_butler]

Not sure if this is the correct category or if the post should be split up. I could see it in a couple of different categories. I am considering the T3 TB or BB to monitor and control a Bacnet MSTP network consisting of approx 15 Honeywell VAV controllers, a few environmental sensors such as the AirLab and a couple of misc.controllers (maybe T3’s, maybe not). I have a few questions to help me evaluate your system.

The Honeywell VAV controllers have fixed point numbers in excess of 1500, so your T3 will not work if the the max point number is 255 as stated in the current data sheet. I have seen forum posts indicating the number has been increased. What are the T3’s current limits for maximum point number per device, number of 3rd-party devices on an MSTP net, and total points (if any)?

I plan to add a second similar Bacnet MSTP network in the future, and it would be good to use the 2nd T3 RS-485 port to control that network. I have seen forum comments that the T3 did not support Bacnet MSTP on both RS-485 ports at the time of the post. Is that currently a limitation?

I want to trend log several points from each of these devices. What are the limits on trend logging in terms of number of points per device, number of devices, total number of points logged, and maximum number of values stored? I also believe that the T3 will log to SD card while it isn’t connected to a T3000 computer and then upload the results when it connects. Is that correct?

Your programming examples show two methods for addressing 3rd-party BACNET devices, one using cabinet, MSTP device address, and point, and another using instance number and point. Has the former been replaced by the latter? Also, does Control Basic have a way to programmatically address an MSTP device, i.e. where the device address is a variable? I’d like to read devices sequentially in a loop to check points against limits and the only way I see to do that is to read each individually using inline code.

I would like to build a table of devices containing info such as MSTP address, device type, parameter upper and lower limits, etc. that Control Basic can access. It appears that the system builds a similar table but I don’t see a way to add fields or to access it from Control Basic. Do you have any suggestions?

Your user manual mentioned a T3 simulator but the download link was TBD or something similar. Is there a way to test Control Basic programs, and particularly with 3rd party MSTP devices, without having a T3? I downloaded and installed the T3000 software and connected an RS-485 adapter to the computer, but T3000 would not see 3rd party devices (which I discovered was normal without a T3) even though Yabe did see them.

Lastly wondering if the above-mentioned limits for the T3 Nano are different than the T3 TB or BB, since it has a different processor and presumably a different codebase. Also, one forum post indicated that adding an SD card to the Nano was in the works which would allow it to log. Is that feature now available?

Regards.

[maurice]

The maximum point number of the T3 applies to the number of IO, and in some cases it might be referring the the number of variables on board the T3. This is separate from the 1500 bacnet points on the Honeywell VAV controllers, they can have as many bacnet objects as you like and the T3 will blindly pass messages on to the VAV controller from anywhere in the system. If you want to process those 1500 bacnet objects on the T3 that would be a different matter, you have to work within the 255 variables on the T3. So for example, if you want to control the occupied mode of the VAV controller you will need a variable on the T3 to hold the occupied state and some logic associated with that. That’s one VAR on the T3 used up. Then you might want to manage a few setpoints and log a few IO on the VAV, that will require a few more vars on the T3 for each of those.

LOGGING EXTERNAL BACNET OBJECTS: You (might) be able to log 3^rd party devices straight to a trend log without this intermediate VAR but I don’t think we have that feature for now, set up a trend log on 3333AI1 (where 3333 is the bacnet instance of the 3^rd VAV controller and AI1 is the air flow reading for example). I will post an update here on whether this is possible. For now, the way to log 3^rd party devices is by setting up a small program on the T3 to pull the data into a local VAR and then set up your trend log on that VAR. Data flows to the SD card on board the T3 and is also mirrored to your hard disk through the T3000 software, its stored as a Microsoft Access format database. T3000 needs to be running to get the data flowing to the hard disk and you need one var on the T3 controller for each item you plan to log.

TWO BACNET SUBNETS: Each T3 controller supports only one Bacnet subnet at a time, I know we have promised two separate MSTP subnets but we’re not quite there yet. So you can have one Bacnet and one Modbus, or two modbus but not two separate bacnet subnets, apologies for that.

NETWORK PROGRAMMING: The example is pretty clear, you can use either method and both are supported. The general rule is for 3^rd party devices use the bacnet reference scheme (ie 3333AI1) and for other T3 controllers use the native T3000 scheme, ie 2IN1. Unfortunately we don’t have loops in our programs as you mentioned, it’s long overdue I admit.

SIMULATOR: The simulator lets you do EVERYTHING in our system with one important exception, the control basic programs are not actually running. To see programs running and turning outputs on & off, that sort of thing, you need some live hardware such as a T3 controllers or the Tstat10, etc.

T3-Nano: This device has been updated to take advantage of a new CPU called the ESP32. The rest of the lineup uses an Arm based cpu from ST and is slowly being switched over the the ESP32. The repositories for both versions are on github if you are interested in poking around in the source code. The ESP32 repo is here:

https://github.com/temcocontrols/T3-programmable-controller-on-ESP32

[larnold]

Confirming Maurice’s comment above.

The 255 number is a setting that specifies how many Bacnet Points may be in a specific Bacnet/MSTP space. Many systems, such as Contemporary Controls, will use 127 as a limitation even, and shortening that number does simplify some of the communication tasks (such as Who Is scans).

Bacnet Instances though can be much higher and multiple MSTP networks can be joined together into a Bacnet system.

Should be not a problem getting to the Honeywell VAV controllers.
If you indeed have 1500 actual objects in a network and 255 (254 actually) points per MSTP network set, you would generally need 6 “lead controllers/routers” that would allow those networks to all work together.

IE 6x T3 which each have a Bacnet/MSTP subnet below of 254 units.
Could also have 12x T3 with each having Bacnet/MSTP subnets below with 127 points.

I am quite sure that it will be a complicated and rather slow scan of the network with that many points involved. Expect that a large Niagra type controller could keep track of all that, but perhaps T3000 with Yabe/ “Temco Bacnet Tool” Could do it as well. I would be inclined to build or have an external database which would pole through T3000/Yabe and get the data in short focused queries once the extent of the network is known.

[maurice]

There is one setting to help with the who-is scanning activity, you can set this max master parameter to a low number. The bacnet who-is scan is different than the device discovery scan which T3000 uses to find devices during commissioning.

[carlton.brown]

I’m not sure I follow. Example: I am looking at a project that will include 130 VFDs w/ BacNet MSTP. I will need to send the Speed command an On/Off digital and an Open/Close(OA Damper) digital (Outputs) to each VFD and I need to know what the KW(Inputs)is for each of those drives. How many T3 controllers will I need?

[maurice]

The T3-nano could handle the job, a single device costing $61. That would be one long and complicated network cable to commission though. I’d break the system up into subnets of say 30 VFD’s, one T3-nano in charge of each group. Therefore 4 units of the T3-nano’s could handle the job quite nicely.