question

westen avatar image
westen asked

Weird Voltagespike causes BMV to reset to 100%?

A couple of days ago I started draining my batterybank, slowly, in order to do some tests with my newly acquired Globallink 520 (which I am very much in love with). I drained the battery to about 95%, leaving a fridge running overnight. Then all of a sudden (apparently out of nowhere) there was a Voltage/Amperage "surge" (see 11.30u.) and on this very same moment the SOC jumped from 95% to 100% on the BMV.

I think the GL520 graphs show this nicely.

The MPPT however was not to be fooled by this and has been dutifully "bulking away" after the anomaly untill in fact fully charged, 24hrs or so later.

My question: does anybody recognise this behaviour?

Any clues to what may have caused it?weird-spike.png


Any insight or thoughts on this are very much appreciated,

Jan

BMV Battery Monitor
weird-spike.png (194.7 KiB)
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
pwfarnell avatar image
pwfarnell answered ·

A small gap in the clouds for a few minutes and the sun shone on the solar panels. Check the trends from the MPPT for panel voltage and current and charging current and see if they increased at that time.

This points to the SOC synchronizing too early. You might like to increase your synchronizing (charged) voltage to 0.1 to 0.2V below absorption voltage and reduce the tail current to something lower than the default 4%.

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.

kevgermany avatar image kevgermany ♦♦ commented ·
That spike may be longer than you think due to sampling intervals. Also the timing of the spike may not coincide.


Couple of other things to check

Charged detection time. This defaults to 3mins. For solar, increase depending on local conditions. Values of 10 minutes or more may be appropriate. This seems the most likely solution here.

Charge efficiency factor. This can lead to the monitor thinking the battery is less charged than it really is. It's trial and error based on different batteries. If it's too low this can give the jump.

Peukert exponent. This describes the behaviour of the battery under load. If loads are generally heavy Peukert may be too low. In this case more likely that it's too high and your loads are light. This again is trial and error depending on how you use the batteries and how your batteries perform.

Battery monitors can only give an indication of SOC, usually it's good, to within 5-10% in my experience. But there are too many other factors affecting the actual SOC to expect anything better. In general an early synch to 100% will not stop the battery from charging if it needs to. The BMV is not a charge controller.


1 Like 1 ·
westen avatar image westen commented ·
Thanks!

At the moment charged Voltage (on the BMV) is set to 13.40V and absorption Voltage (on the MPPT) to 14.30V

I will raise Charged Voltage on the BMV to 14.10V and see what happens.

0 Likes 0 ·
jolir avatar image
jolir answered ·

This was my experience too. The BMV712S was synchronising too soon. I increased the charged voltage to a couple of points below the absorption voltage and it solved the problem.

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.

westen avatar image
westen answered ·

These are all very insightful and helpful answers. Great advice, thanks!

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.