This project has moved and is read-only. For the latest updates, please go here.

Date/time stamp and data in Spot log files

Jun 30, 2014 at 2:58 PM
I am working on a script to send data created by the SBFspot files to emoncms. I've got a sript which works with the daily 5 minute interval files. Now I'm looking at the "Spot" files. At the moment I am using 123Solar which (I haven't looked closely at the code) seems to run sbfspot every second. I have the sbfspot configuration set to generate the various spot files even though 123Solar doesn't use these, but my scripts do.
But I don't quite understand the data in the spot file so I'm hoping someone may be able to explain to me what is going on?

Here is an extract of todays spot logfile ( I have omitted several fields that are not relevant for now):
dd/MM/yyyy HH:mm:SS;PacTot;EToday;ETotal;
30/06/2014 14:40:33;905.000;19.038;5945.789
30/06/2014 14:40:38;903.000;19.039;5945.790
30/06/2014 14:40:38;908.000;19.039;5945.790
30/06/2014 14:40:38;907.000;19.039;5945.791
30/06/2014 14:40:43;910.000;19.041;5945.791

Looking at the time stamp, there are 5 seconds between each time stamp, and looking at the overall logfile that seems to be consistent. Each unique time stamp has 5 seconds between them. However, there are 3 lines with the same time stamp. First I thought that since 123Solar is requesting the "now" data so often I was simply seeing some logfile entries that would have the same data until the next 5 seconds have gone, and I could ignore the additional entries with the same time stamp.
But when you look at the data, that is not quite the case. In this sample some of the values DO change for entries that have the same time stamp. Does anyone know what might be going on?

On a slightly related note: Does anyone know of a link to SMA information that possibly describes the Spot data in more detail?
Jul 1, 2014 at 6:27 PM
Edited Jul 5, 2014 at 5:14 PM
Dear BSG63,

interpreting data from the inverter at "seconds" interval will not deliver meaningfull information

the inverter is designed to handle in first priority a "BEST" operate/treat for the generation of energy from your PV-install
  • the highest priority is for:
    -> control on correct operational conditions
    -> the decision system on the best MPP point
    for the MPP algorithm, the decision-cycle is repeated about each 3 seconds
  • the extraction of data is treated as lower priority
from my experience/observation:
  • the handling/extraction of the "SPOT" data for "reading purpose" has a delay of a few seconds
    AND it includes a "freeze" periode of about 4<->10 (20) seconds,
    mostly the data refresh needs less then 10sec but often (about each 50--80seconds) this refresh interval is longer
    within one freeze period - the changes on the extracted data are "not predictable"
    keep in mind that your inverter include several modules (DCpart, ACpart, . . .) 
    the data transfer module extracts the data from each module in a step-step mode (which explains the delay and time issue)
    
eg: see value of operat-time and feed-in time (expressed as 1/1000 of an hour) vis-a-vis to your interpretation

from observation I found that:
-A- a 10seconds (or slower) interval is "BEST" choice to catch "stable" info from an inverter
         very short intervals make interpretation difficult
-B- if you look in the detaled output form SBFspot you will even find that the time-stamp in the inverters-info might be "wrong"
  take the "deboug-output" options -d5 -v5 
  eg in my inverter the DC info has a time stamp that is rounded to the 5minute
          each module delivers info with "own" DateTime assumption
-C- warning - some of the values that are extracted from the inverter might be "calculated"
cfr several of the power values
-- as a result the interpretation of these values might give "strange" effect

-D- we have not yet found 'all-in' SMA documents with information about the interpretation of its data,
on the SMAwebsite's (germ and others) you find limited information,
  most of the interpretation of the data by SBFspot is found via "reverse engineering"
kind regards wim
Jul 2, 2014 at 12:44 PM
BSG63 wrote:
On a slightly related note: Does anyone know of a link to SMA information that possibly describes the Spot data in more detail?
This one has a lot of info: Measured Values and Parameters
Sep 9, 2015 at 4:52 PM
Hi,

at BSG63: Sorry for the offtopic question, but how did you obtain data with 5 seconds intervall?

Many greetings,
Simsa
Sep 9, 2015 at 10:17 PM
SBFspot is not designed for this.
There is too much overhead for connecting and logon/logoff to the inverter. If you want higher frequencies for spotdata retrieval, the connection to the inverter should be kept open all the time.
Sep 9, 2015 at 10:29 PM
It wasn't anything I was trying to do on purpose. But I'm using 123solar.org and it continuously polls the inverter at that speed.

On 09/09/2015 16:52, iwes1 wrote:

From: iwes1

Hi,

at BSG63: Sorry for the offtopic question, but how did you obtain data with 5 seconds intervall?

Many greetings,
Simsa

Sep 10, 2015 at 11:35 PM
to BSG63,

after reading the FAQ for 123solar I found that there it is stated: (see quoted text below)
  • a 5minute logging frequency is enough
  • i expect that the "123solar log" master can inform you whether / which "parameter" is to be set
    to call for sbfspot at a moderate tempo
kr wim

Quote-123Solar: - - - - - - - - - -
Is a 5 minutes logging enough ?
A 5 min logging is more than enough to have relevant data.
Still, the pooling of gauges and live data updates as much as possible !

No, it's a logger. It should run continously OR at least before down and after the sunset to get precise production datas.
Also, a serious monitoring needs to run continuously.

Sep 11, 2015 at 10:08 AM
@SillieWimons
Quoting 123Solar :
<< Still, the pooling of gauges and live data updates as much as possible >>

===> 123Solar has gauges. That's why it reads data continuously !

Logging is a different thing -> A data sample is added to the 123Solar inverter CSV daily file at a 5' pace.
These CSV files are the data source of 123Solar detail charting page.

The gauges will not be too bad in single inverter configuration. Altough they will perturbate or prevent other applications to access the inverter.

Gauges just don't work in multi SMA inverter configuration.
Sep 11, 2015 at 7:51 PM
See here: http://www.123solar.org/phpBB/viewtopic.php?f=8&t=195

It is possible to call only once every 60 seconds sbfspot