Negative Power Indicated

Ocassionly I observe a small negative power reading on one leg of the Mains.
Any ideas on what causes this?

You have solar, correct?

I have solar panels.
After 50+ days of "working " with the Sense SR team and what I perceived as
incorrect diagnosis, they were able to get the Signal Check to complete and
a few devices have identified.
Their last diagnosis refered to the Midday Solar Dip. I see no indication
my Sense data which reflected that phenomenon. In one email there was a
comment about negative power readings causing the automatic signal check to
fail.
If so Sense SRs or supervisor may have overidden the automation in order to
get me going.
The negative power is 7-15 Watts.

Alan Kirk GMail

If one of your mains is reading negative I would guess there is a good chance that one of your solar CT’s is not fully closed.

Been through that exercise, twice, sent them photos and they found
another rabbit hole to send me down and another factory reset or two. None
of those exercises were productive. Part of the reason it took them 50 days
to get basic functions going.
There was no believable technical explication on what was "discovered "
when Signal check passed and Device detection started.
I’m just happy it at least appears to be fixed!

Alan Kirk GMail

If you turn off your solar when the main is showing negative does it go positive? If it does then either one Solar CT is either not fully closed or there is something else wrong with it.

It does go positive!
I have a perhaps unusual solar install which draws 100W off the Mains all the time but is on the Load side of the solar breaker. I am guessing this may have given them a problem, though the Factory would have known this because they see all the data.

Edit:

Alan Kirk GMail

Something is DEFINITELY wrong!!! We all know you can’t have negative on the main unless solar is pushing power out. Thus if you are still seeing it when the solar is off then there has to be a calibration of physical problem with the CT on that main. Have you gotten any assistance from Ben on this?

He is aware of the startup issue but I don’t think he is aware of the negative power though he may be now!
It’s only on one leg. The constant load on the solar side is balanced across both legs.
Tried another experiment and as I load that Mains leg the value swings positive. So! It appears my Solar side load, which is accounted for at Reset / Calibration doesn’t fit the factory power calculation model