question

peter-bris avatar image

Smart Sense with smart MPPT causing large charging voltage over-shoot?

Hi, I noticed the Smart Sense voltage displayed by phone correlates pretty good with the battery voltage displayed by a BMV-700, so let's call this the actual voltage (confirmed by multimeter reading).

But there's up to 0.3V difference between actual voltage and displayed voltage of the MPPT on the VictronConnect app.

The MPPT voltage shows up to 0.3V less on VictronConnect which consequently causes the MPPT to over-shoot its target by the same amount. It can take 15 minutes for the voltage to settle at the correct level (see attached graph taken with the BMV-700 through BMV Reader). So yes, the error disappears after a while.

I've tried this on a smart MPPT100-20, and on a MPPT100-30 with b/t dongle both showing this weird voltage over-shoot. Voltage drop issues can be ruled out since I used short runs of B&S6 wire for this test. Temperature coefficient set to -22.20mV/degC, ambient temp 19 degC, settings for Vabs=14.6V, Vfloat=13.8V.

Has anyone else noticed this behaviour with another MPPT/Smart Sense combo?

MPPT - Solar Charge ControllerSmart Battery Senseovervoltage
2 |3000 characters needed characters left characters exceeded

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

2 Answers
Mark avatar image
Mark answered ·

I can't comment on the delta between voltage displayed on the MPPT (plug in screen I assume?) & VictronConnect - I used a Smart Battery Sense initially & did not notice a difference in the voltage readings.

But about the 'overshoot' you have logged - are you able to quantify the time between when the correct absorption voltage is reached and the peak voltage?

I believe that currently there is a time delay to confirm the absorption voltage has been reached, before ending bulk. So with a high charge current %, this could possibly cause a small overshoot relative to the absorption voltage set point. I also believe that this delay will be shortened significantly with the next MPPT firmware release. So it would be good to retest then.

1 comment
2 |3000 characters needed characters left characters exceeded

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

@Mark

......But about the 'overshoot' you have logged - are you able to quantify the time between when the correct absorption voltage is reached and the peak voltage? ......

Hi Mark, it's about 2~3 minutes as can be seen by looking at the bottom left corner of the purple shaded area. By coincidence there's a conveniently exposed voltage logging point shortly before the peak. Fingers crossed you're right it'll be addressed in an update.

Guy Stewart (Victron Community Manager) avatar image
Guy Stewart (Victron Community Manager) answered ·

Do you get the same effect if you restrict the charge current to 10A?

Can you also confirm that you have been through the procedure in Victron Connect to synchronise Smart Battery Sense and the SmartSolar with a Smart Network?

6 comments
2 |3000 characters needed characters left characters exceeded

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

@Guy Stewart (Victron Energy Staff)

Thanks Guy, for you quick response.

Yes confirmed, the MPPTs are linked with the Smart Sense (app shows temperature readings).

On the MPPT100-20 with current restricted to 10A I get less over-shoot than on unrestricted 20A. Also, the update frequency of the battery voltage in VictronConnect seems to vary quite a bit between 3 seconds and 1 minute. When the actual voltage changes faster, update frequency is lower. Although this might be a sign the control loop is struggling it doesn't totally freeze it for a minute, only the display updates are slower.

What I really don't understand is why it can take 20 minutes for the output voltage to come down. And this time seems to be linked to the absorption time and SOC at begin of charging. With shorter absorption times it settles back on the target voltage faster.

On the MPPT100-30 with b/t dongle, update frequency stays constant about 1 second.

When I restrict current to 15A on this unit, I don't see any over-shoot, only when left at unrestricted 30A.

It looks like the smart MPPT100-20 isn't quite as stable as the non-smart MPPT100-20 with b/t dongle as it over-shoots more, and also takes longer to settle down on the correct voltage level.

This doesn't really bother me when using AGM batteries.

But my LiFePO4 with internal BMS take offence and goes into O/V disconnect.

The potential problem, on O/V disconnect, the MPPT charging output spikes a bit on top of its over-shoot which could affect sensitive loads. IIRC I've seen 17V when I looked at this with a CRO.


Hi @Peter Bris,

Thanks for the detailed reporting.


Is it possible to get a drawing/picture of the setup?

Can we get some logs from VictronConnect while the overshoot occurs?

Ideally close VictronConnect and start it (this way you start with a fresh log). Then connect to the MPPT for a few minutes during the overshoot. Next create a service report (in the settings menu of VictronConnect).

You should be able to attach the .log file to the community answer.

Thanks again.


peter-bris avatar image peter-bris Guy Stewart (Victron Community Manager) ♦♦ ·

Hi Guy, I've no idea how to create a service report in VictronConnect. All I can find is an export function for settings? Am currently looking at some trend developing during a test. Intend to do a screen capture of the over-shoot.

peter-bris avatar image peter-bris Guy Stewart (Victron Community Manager) ♦♦ ·

Guy, I was able to follow the instructions re service report creation, thanks.

I don't know how to read the data in the log though. And attaching it here doesn't seem to work (file type not permitted). Please use contact form instead http://batteryvalue.com.au/#feedback. According to the multimeter, actual battery voltage rose to 15V (again) during charging and then slowly dropped back to the correct level of 14.88V. Vabsorb=14.6V, -22mV/degC, 13 degC battery temp. Although the multimeter and Smart Sense readings differ by max 20mV, when VictronConnect shows bulk on the MPPT screen, the displayed battery voltage consistently is 0.2V below actual.

Over-shoot peaked out at 15.04V (multimeter). Am going to run another test with Vabsorb set to 14.3V to see where it peaks then.

Cheers, Peter

peter-bris avatar image peter-bris Guy Stewart (Victron Community Manager) ♦♦ ·

FWIW I've attached the csv file of the graph in my original post. Row #118 shows the peak. The setup consists of an MPPT100-20 linked with a Smart Sense. Data was collected with a BMV-700 connected to laptop running BMV-Reader via VE.Direct cable. The actual battery voltage as reported by the BMV-700 and double checked with multimeter was almost 0.3V higher than it should have been, like 14.70V in row #146.

Repeated the test tonight, but don't know how to export data log out of the MPPT with VictronConnect. BTW, as mentioned earlier, the MPPT never shows the over-shoot in the VictronConnect app. The information appears to get lost somewhere between the MPPT and Smart Sense. There's no over-shoot if Smart Sense isn't involved.Change the file extension to .csv (it didn't upload so I gave it .txt) 100Ah 20A closed loop.txt

smart-sense.jpg (174.2 KiB)
15v.jpg (2.0 MiB)