I have 2 Quattros 48V/10kVA/120V running in a master - slave split phase configuration in the US. They have 3x 48v/16 kWh LFP batteries for storage in a closed loop communication system and a cerbo GX device. There is NO solar power. These are set up for time of use arbitrage - charge with cheap electricity at night (11PM-7AM), discharge during expensive power during the rest of the day.
I’ve noticed unstable ac voltage output on L2 for some part of the day (usually from 7 AM to 4 or 5 PM). At times the AC out will drop from 120 V till 115 or even 114V. I see some oscillation in the brightness of the LED lights in my home that I suspect is related to these drops. After 4 or 5 PM, the instability goes away both in the AC out voltage and in the LED brightness. This has been happening for months - atleast 4, which is the entire operational life of the system.
L1 is perfect. No issues at all.
This does NOT correlate with any loads. The largest loads are the heatpump, dryer and oven and the voltage drops happen on L2 even when these are off. Similarly, the instability does not happen even if these loads are turned on after 4 or 5 PM.
I’m a bit lost on how to debug and fix this. Here’s my thinking so far:
Since the issue is consistently time dependent, not load dependent and self resolves after some time of day, it is unlikely to be wiring/connection. If it were connection dependent, the issue wouldn’t resolve every evening. That said, I inspected and retorqued all the connections without any improvement.
This led me to look for configuration issues, and to my knowledge, the configuration through VE.Configure is correct. Also L1 (the other phase) is performing flawlessly, and during configuration I used “copy to all units” so the configuration should be the same for both L2 and L1.
So now I’m out of ideas and turning to you for help. I’ve attached the outputs over time for L1 and L2 here.
Some other information that may help or may be a complete red herring:
- L2 seems to run a bit hotter, its fans come on more often under load
- L2 does have a slightly higher baseline load as it powers the more frequently used circuits.
- The issues seem to resolve when the DC voltage drops to around 52.xx ? Just grasping at straws here, could be correlation without causation.
I appreciate any help and advice.
