question

papagoy avatar image
papagoy asked

MPPT 75/15 calendar/day count error

Solar charging on a boat installation: Using solar panels that are put out only when the boat is at mooring.

I noticed a while back that the historic day count and calendar date assignation on the MPPT charger is wrong, and this weekend realised this is because it is actually counting periods of panel voltage high corresponding to sunset/sunrise. In my installation the solar panels are disconnected and stashed below decks when sailing. Then put out and connected when at the mooring and space on deck permits.

Is this fixed on v1.42 as suggested in another thread where they are discussing passing through shade being mis-interpreted as a new day event ?

Or do I need to submit some sort of bug report for the software error to be corrected.

MPPT Controllers
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

3 Answers
adev avatar image
adev answered ·

The only time I’ve noticed mis-count is when Ive disconnected the PV for a period of time or the battery (and pv).


cloud, shade etc never been an issue for me.

1 comment
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

papagoy avatar image papagoy commented ·

I'm going to monitor it more closely, but I think I'm getting a number of very short 'days' logged, possibly when cloud cover changes soon after sunrise when the voltage is close to the threshold. Midday I wouldn't expect problems. (UK at 53 degrees north, so a lot of slow starts or ends to the day under grey skies or rain showers.)

0 Likes 0 ·
wkirby avatar image
wkirby answered ·

The MPPT it's self does not have a real time clock in it.
It detects a new day by the solar panels and starts a timer. If the solar panels are stowed then it cannot detect a new day.

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

papagoy avatar image
papagoy answered ·

Further to answers so far:

I Accept that the MPPT itself is counting 'charging events', doesn't maintain a synchronised clock and assumes a fixed installation. So a logged 'charging period' is a period of solar input.

The link to a calendar is done in the app receiving the bluetooth data for the last 28 events using the clock in the phone or device.

To refine the software bug report/update request.

(i) The bug: It is simply wrong to assign a date or the unit 'days' to the history when they are charging periods that are not necessarily defined by sunrise events.

(ii) The update request: If the raw data includes times for bulk/absorption/float, then surely a time for the 'no input' or night-time period can also be (or is) logged. This can then used in the app software to assign dates to the previous 28 logged charging events accurately ?



2 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

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

David avatar image David commented ·

If it is important for you to log the history on a calandar day by day basis, you could consider connecting the MPPT Controller to a Venus OS system (even just a Raspberry Pi running the Venus OS will be sufficient). This will poll the MPPT on a regular basis (down to 1 minute intervals if you want), and log the data polled at a specific time.

Uploading that data to the VRM Portal would then give your data in both number and chart form and clearly show what time of day harvesting was occuring the level.

0 Likes 0 ·
papagoy avatar image papagoy David commented ·

Thanks.

I probably will be adding an external logger in the next winter maintenance. The calendar thing was really that while getting used to the installation it confused me greatly how I was ending up with 9 'days' in a week between visits, with the best charge happening on the cloudiest days! So it would have helped if the app didn't give dates, or couls assign dates accurately.

There is another problem highlighted in other threads relating to the Bulk/Absorption/Float sequence which apparently it always starts with a minimum period of Bulk, even if the 'night' has only been a few minutes. In that case I think they were talking about an installation of solar panels on the roof of an RV that might drive through tunnels in the middle of the day and trigger a new sunrise event.

0 Likes 0 ·