The Priority Scanning Concept for Non-Schneider Devices

[aashraf]

In one of our projects we have more than 750 devices,

The global scan interval was set to 3 min, But the customer is not satisfied with this scanning interval and he is asking for live monitoring (instantaneous update), I know it's not applicable and the min. scan interval is 1 min. but for sure this interval is very small and may lead to performance issues.

So, please advise for the recommended scan interval.

Another thing which I need to discuss,

As far as I know all Schneider devices send a Trap to DCE in case of alarm to do priority scanning and don't wait for the scanning interval.

But unfortunately the UPSs and Cooling units are 3rd party devices and they have to wait for the scanning interval which is leading to some update delay and this is not accepted from the customer side, So please advise if there is any wok around to get instantaneous update at least in case of alarms (We need some workaround to apply the same concept of priority scanning like Schneider devices). Your advise will be well appreciated.

Thanks,

Ahmed Ashraf

[Gabi]

first of all, customer must understand how the things are working in a monitoring system and if he want to monitor a device instantaneous just look on the equipment web page 🙂

default time of scanning is set to 5 minutes for some reasons(you can go down but for sure you will have a lot of performance issues with 750 devices) 

I did not make this kind of tests  but I`m almost sure that DCE cannot read 750 devices in a time less de 3-5 min(some devices have more then 50 parameters and DCE read all this parameters)

second problem is also equipment's management cards which is not designed and maybe limited to receive/supply a lot of requests(technically, the management card is a different/optional part the equipment`s "brain"......since also this card read itself from  the "brain" of course there is a priority(100% serial/modbus slow reads) and "brain's" priority is to keep device in parameters and after that to supply status data)  

I manage the same request from a customer and because of slow/bug management card after 2-3 days the card is freezing and they must restart the cooling unit(since is a 250kw chiller unit takes a lot of time to restart in parameters) 

in this case the problem was linux software of the card and the hardware(sessions remains open and every session use a small amount of RAM memory ...when the RAM memory has been exhausted, the system start to kill random processes   

regarding the traps....normally every equipment(at least brands) also have traps available.....but sometimes these are not implemented in the DDF file(you can open DDF file in a editor and check if there are traps at the end of file)  

also on many customers I implemented  a "redundancy" for the DCE by using dry contacts of the devices(of course is available and reliable on critical devices like HVAC/UPS) 

you can use netbotz if you have in infrastructure or a dry contacts gateway and you can see status changes of the devices instantaneous by mail/sms/beacon/siren/etc (of course you cannot see the exactly error but a general change of status )