SmartShunt SOC Automatic Synchronization not triggering

Hi All, I'm having trouble understanding why my SmartShunt 500a won't automatically sync SOC to 100% even when all parameters appear to be met. I'm seeing this across a few installations and it isn't just a one-time occurrence.

Tail current should be 24A - and the maximum recorded in VRM is ~6A.

Charged voltage is 13.2V and the lowest recorded is ~13.35V.

Detection time is 6m however the chart below shows a period of 6 hours where synchronization did not occur and my SOC was sitting at ~90%.

Batteries are LiFePO4.

All I can think of is that there are transients from my load or charger which are not being captured by VRM but are causing the shunt to re-start the synchronization period.

Any other ideas?

2 |3000

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

klim8skeptic ♦ commented · Oct 03, 2023 at 11:05 PM

@John Richardson I would not expect a 4 cell, 12v LiFePO4 battery to be charged until it reaches 14v (3.5v/cell).

What is your battery? What are the manufactures recommended charge voltages? What voltages have you set your chargers to?

0 ·

John Richardson klim8skeptic ♦ commented · Oct 03, 2023 at 11:34 PM

Hi @klim8skeptic - I'm assuming you would expect a higher charge voltage. The previous shot was showing the end state of the charge - I think the charger was in some kind of power supply mode or popping in and out of charge. Here's the end of the actual charge where the voltage is significantly higher. Batteries state recommended bulk charge - 14.6v and maximum charge 15.2v. My charger's max charge setting is 14.4v.

0 ·

charge-end.png (56.9 KiB)

klim8skeptic ♦ John Richardson commented · Oct 03, 2023 at 11:47 PM

@John Richardson 13.4v is a good storage/float set point.

For the BMV try the following,

Charged Voltage 14.2v

Tail Current 3%

Charge detection time 3 mins

Peukert exp 1.05

Charge eff 99%

See if that works for you.

1 ·

kevgermany ♦♦ commented · Oct 04, 2023 at 04:25 AM

There's a setting in VRM which controls how often data is sent to the server. If you reduce this to 1 minute you'll get more detail. Might not capture all transients, but you have a better chance.

These will be more visible if you look at graphs with a shorter time period, say for one or three hours, reduces the effects of averaging.

0 ·