Hi,

Is there a maximum number of input DataPoints that may be configured on a DataLogger?

Regards,

Daniel

 Hello Daniel,

 I've only setup data loggers with less than 100 points, but I'm sure you can get more than that.  I'm working on an FPM to help with testing the iLON's performance.  It shouldn't be too much trouble to build a multi-point output function block to help answer this question.

Richard

 

Daniel, We have dataloggers with above 360 datapoints. We have other cofigurations with 5 dataloggers each having in average 200 datapoints. Both work fine so far. But anyway, knowing a little bit (sic) about the datalog mechanism that polls the datapoint on the network (powerline of FTT), the problem may not be too much about the number of datapoint that you ask a datalog functional block to watch but more about your datalog strategy: how often you datalog ? on change ? every X minutes ? mixed ? ... and the (dramatic) impact on the network bandwith. Tell us more about your application and your datalog strategy, maybe it is something we or someone else has seen. So that we can share with you our experiences. C.

Hi Richard

thanks for your hint. I'm still working hard to make run our old e3-network on the SmartServer (migration being only a little part of the process... ). I'm not yet into FPM programming. I will let you know as soon as we are further on with that.

Regards, Daniel

Hi Christophe,

Thanks for that information.

We are working on a PL network, the SmartServer in standalone mode. In future we'd like to switch to the v40.XIF in order to be allowed to have more datapoints in the network. (More datapoints --> more data to be logged...)

Our datalog needs: Once a month, a scheduler sends a command to all the nodes (lamps) in order that they carry out their automatic selftest (=selftest of the lamp). After the test we want to write all the test results (= the nodes output variable) in a log. The writing of ALL datapoints has to be carried out on an exactly defined time (some time after the selftest has been compleded), whether the nodes output variables have been changed or not. It's a DataLogger that is neither polled nor 'on change'. So what we need is a 'Scheduled DataLogger'. (Maybe Echelon will read that post and that will be part of the next version of the SmartServer...;-)

Our datalog strategy: Each DataLogger is linked to a Scheduler. The Scheduler output is bound to the nviDlEnable. The input datapoints are configured, PollRate=0s, MinDeltaTime=2min, MinDeltaValue=0.

Each time we want the data to be written to the log, the Scheduler output is set 'ON' for one minute (=the minimum time frame of a Scheduler), that enables the DataLogger which causes it to write all its input variables. One minute later the scheduler is set back to 'OFF', disabling the DataLogger. As the MinDeltaTime is set to a value greater than the scheduler 'ON'-time, only one entry is done. (If the MinDeltaTime were smaller than one minute there was another entry at the moment where the logger is disabled).

That's how we got it to work. All the datapoints are logged at exactly the moment in time we want, and never else.

Please tell me if that was not clear enough. Best regards,

Daniel