question

mark-a avatar image
mark-a asked

How unreliable is the sunset/sunrise detection delay with the Streetlight function?

UPDATE: The sunrise switch-off delay used to be a bit short but today it was late by an hour! See below.

I use the load output of my BlueSolar MPPT 75/15 to power my anchor light and it is set with a 30 minutes sunrise detection delay using the Streetlight function. It works as expected except for the delay being somewhat short. As can be seen in the screenshots in February 13th the sunrise detection voltage was reached at 06:38:37 and the load output is switched off at 07:06:12.


My questions are by which means are the delays enforced and what accuracy can I expect?



screenshot-20230313-070928.pngThreshold is reached at 06:38:37 …


screenshot-20230313-070818.png… load output is off at 07:06:12

screenshot-20230313-071016.pngStreetlight function configured values

screenshot-20230415-070417.pngStreetlight sunrise switch-off overdue by an hour

Streetlightaccuracy
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

2 Answers
kevgermany avatar image
kevgermany answered ·
You should treat load output as off after the drop, not before, in reality the point was somewhere between the start and end of the drop, but it's immaterial here.

Similarly the threshold voltage would have been reached before it was recorded on the graph.

I think what you're seeing is limits of graph accuracy due to inadequate plot point frequency, not inaccuracies in the time delay.

1 comment
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

mark-a avatar image mark-a commented ·

Thanks for your answer but Sorry, I can't agree.

1) when an electric circuit is switched off the current will drop to zero only after the switch has been opened (unless there is a time machine in your circuit ;-). It can drop slowly afterwards because capacitors and inductivities within all store a little bit of energy. Also the drop was completed within a second but the delay is short by more than a minute and it was similar all other days I checked.

2} I did not retrospectively derive the points from the graphs but watched it happen in real time. Furthermore the curve around the 13.45 V point is very steep. Even if there was a peak too short to make it into the graph it would skew the calculations by no more than a second or two but as mentioned above the shortage of the delay is more than a minute.

I understand the BMV doesn't have an RTC so the delays must be enforced by a different method, e.g. a counter or an interrupt, many of which are prone to such problems. What I'd like to know is if there is a limit or if the delays could be aborted much more prematurely under adverse circumstances.

Here is a closeup of the graph at the point where the threshold voltage was reached.

screenshot-20230313-064047.png

0 Likes 0 ·
kevgermany avatar image
kevgermany answered ·

Sorry, wasn't obvious that you were timing this and using an external meter. If it's important, suggest you contact your dealer, who has access to more resources.

1 comment
2 |3000

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

mark-a avatar image mark-a commented ·
Well, you don't need an external meter for this and I didn't. The time resolution of the realtime graphs is well below one second and with 10 mV the voltage resolution is sufficient to catch the points with an error in the range of a second. Especially the load shutdown can be determined with high accuracy.
0 Likes 0 ·

Related Resources

Additional resources still need to be added for this topic