Is there anyone who has extensive experience with these controllers? I was always curious to know how many controllers can be handled by each version of the FX Supervisory Controllers....although I'm sure the capacities are not based on controller count but more so on type of written programs as well as FX alarms, trends, etc. A majority of the field controllers will be controlling CW/HW/VFD-controlled AHUs and VAVs. One of the FX Supervisory Controllers will also be integrating to an existing Trane Summit System via BACnet IP (Chiller Plant, AHUs and VAVs) and possibly future Honeywell Spider Field Controllers via Lontalk.
Thanks.
Generally, it's the bus that limits the number of controllers under a Jace.
RealEyes
Realize
RealLies!
Some busses allow a fixed max. number of devices. With a Jace tho you can add a second or third network and increase the device count that way, or adding a router to your network may allow an increased count on a bus. I don't think I've ever seen or heard of an "absolute" number of controllers that a Jace could handle. You've mentioned some of the features that could add to that limit though,
i.e., "written programs as well as FX alarms, trends, etc."
RealEyes
Realize
RealLies!
This is a real job. The FECs will be controlling AHUs, VAVs and Hot Water System equipment and will communicate to the FX Supervisory Controllers via BACnet MS/TP.
I'm figuring it will be best to use FX Supervisory Controllers vs Johnson NAEs because I feel the future integration to existing 3rd party Honeywell LON Controllers will be much smoother using FX20/60/70 Controllers.
...so, youre from a JCI branch ... or an ABCS?
The FX will easily communicate to as many FEC as an NAE. What sort of unit load do FECs present on a MSTP bus? Same constraints for FX.
we have FX60 that concurrently interfaces 5 different RS485 bus including modbus and Bacnet, SNMP, consumes RS232 strings from the fire system, and serves graphics for the site....and it still has capacity.
How many devices do you need to communicate with
1 + 1 = 3 ( *** for very large values of 1)
...everybody wants a box of chocolates and long stemmed rose
Be brave. You cannot get eaten by an imaginary tiger.
We are an ABCS but currently have access to both FX & Metasys Lines. I'm looking at around 450-500 FEC field controllers (400 of which are FEC VMA VAV Controllers and the others for AHUs and Hot Water System equipment control). I'm thinking 2 or 3 FX Supervisory Controllers would be needed....not sure if to go FX60 or FX70s.
NightrainATC,
Put some info in your profile please.
While you're at it apply from pro membership. I know I would be more willing to assist in the pro forum.
If sense were so common everyone would have it !You cannot protect the Stupid from themselves !"Experience is the ability to recognize a mistake Before you make it again!" (Stolen Quote)
I've got a site with 3200 points on a single FX60. The N2 bus is split in halve, on two different physical ports. I have the server software package doing the graphics and trending. The FX60 is maxed out. I have similar sites with 2300-2400 points and everything runs at a much lower level. So, the consumption is defiantly not linear.
Mechdorn,
Thanks. This is the type of information I was looking for. I too will be implementing a FX Server so is it safe to use 1600 pts per trunk as a max guidline?
At this time, we are used to using NAEs which have controller count limitations. Because this is a big facility which may require 4 to 5 FX Supervisory Controllers, I'm assuming we will use the RS-485 adapter for each FX Supervisory and run (2) BACnet MS/TP trunks per FX with repeaters as needed.....is this correct?
Lastly, I will be bringing in existing Trane pts from (2) BCUs...approx a total of 600-800 pts...over BACnet IP into the FX Server. Also future Lontalk integration to approx (15) Honeywell Spider field controllers which would require a FX LON card.
I omitted the fact that all of these jobs that I'm referring to all have the extended memory license installed. With a FX60, 1600 points maybe a push without the extended memory, no problem with the extra memory and a supervisor software serving up the graphics.
NightrainATC,
Did you submit that Pro member app/request yet ?
The FX-60 with extended memory should match with an NAE5500 point for point with capacity to spare if you put all your graphics on an FX-Server.
I can't remember the figure but the limitation is "File Descriptors", I think a 60 w/o extended memor has a 2000 limit but I am not sure. You should be able to easily handle 200 devices on multiple trunks in a single 60
Freddy-B may be able to give you more info on massive numbers of devices/points in an FX-60.
Apply for Pro Membership !
If sense were so common everyone would have it !You cannot protect the Stupid from themselves !"Experience is the ability to recognize a mistake Before you make it again!" (Stolen Quote)
Thanks. I figured you had extended memory. I'm going with extra memory anyway to accommodate future controllers.
Question: If I had a job without the FX server and it had (3) FX Supervisory Controllers, could the customer hit all three controllers on their network from under one controller or do they have to go into individual access points to see what's under each controller? Also with this scenario, is there anything I should no regarding the graphics?
Thanks again!!!
Propagating the formula. http://www.noagendashow.com/
Control folks may go Pro by simply requesting to do so, to the CFC...
The reason is the regular Pro route is based off of things such as CFC cards and a like that several control folks obviously do not have.
The Controls committee then votes based on ones posts to wither or not we believe they are a "Pro"
Should this be your desire to go Pro, all I need is a request from you via this thread
Chris
Good day Mechdorn,
Indeed, consumption is not linear and in fact has have more to do with physics than the performance of the N2 supervisory controller. Remember that the N2 bus communicates at 9600 baud and so the maximum bandwidth (theoretical) is 960 characters per second (10 bits per character). Talking to each device on the N2 bus requires a packet size which is dependent upon the the N2 device itself (N2 device, DX devices, and VMA devices with N2 protocol having the smallest packet size in general), the number of (mapped) points in the N2 devices, the rate with which the points change, how fast the device gets off the bus, how fast the supervisory gets on and off the bus, and what priority the supervisory/user set for particular devices... all this determines how much time each field device gets attention (i.e. transferring point data to/from the supervisory) from the supervisory controller... Thus, the number of points is a guideline, but is very site specific for the reasons I previously mentioned... and so you are 100% right that consumption is not linear.
Cheers,
Sam