Niagara LON 'ping fail' alarms

[dapper]

Some of my sites generate ping fail alarms on the LON networks. In most cases the ping returns to normal within a second or two.
Is there a way to delay the ping response time of to make the system generate the alarm after maybe 2 or 3 ping failures?

[lwarren]

I don't know of anyway to do that, but you could increase the ping frequency up from the default of 5 minutes.

[shovelhead]

Look at the poll scheduler statistics, what's the dibs bucket doing?

[risbo]

Have you got your EOL terminations done?

[Chris_Worthington]

Assuming the LON Utilities Manager yields no errors.... and No one needs to know immediately on ?.... why not just change the alarm class?

From there you could do something with an active alarm count or something along those lines to indicate a comm failure to the default/whatever alarm class. Including a delay.

[Chris_Worthington]

Better yet change the class, add a status demux to device, add a delay to the down, link to whatever alarm.

[Chris_Worthington]

I do both, but VAV's I don't really care about unless several has given up the ghost.... Plant controls diff story

[MatrixTransform]

Some of my sites generate ping fail alarms on the LON networks. In most cases the ping returns to normal within a second or two.
Is there a way to delay the ping response time of to make the system generate the alarm after maybe 2 or 3 ping failures?

sounds like poor comms to me.
are all links bound or polled?
nciSndHrtBt and nciMinOutTime set on all the the field devices?

[dapper]

Thanks for the replies guys. EOLs are all installed, Dibs are at 0%, points are polled. Will try some of the other suggestions and see if something works. Chris, most devices are not critical but a ping failure is the only way to tell if a unit goes down from a tripped breaker or blown fuse etc. All controllers use the unit 24v for power.

[Chris_Worthington]

Thanks for the replies guys. EOLs are all installed, Dibs are at 0%, points are polled. Will try some of the other suggestions and see if something works. Chris, most devices are not critical but a ping failure is the only way to tell if a unit goes down from a tripped breaker or blown fuse etc. All controllers use the unit 24v for power.

I'm with ya there, but who cares about a VAV going down in the middle of the night or a weekend? Not me and not any building engineer that I know, but if I lost multiple then that I would like to know about.

Once you change the Alarm classes and by adding status demux to each critical device, from there you can do many things with the Boolean outputs to COV a alarm.

[MatrixTransform]

... points are polled.

Recommend that you dont poll

[dapper]

Thanks Matrix, I have since bound all the points.