grid current varies with solar output

[glenb]

Hello,
I am noticing that the more solar I produce, the higher the grid reading as well.  Is there a way I can fix this?
It is a cloudy day today, and the solar output is going from 6kw down to 1kw at times.  With virtually no change
in house current, I see the 'grid' value going from 550watts to almost 1300watts depending on the solar output...


TIA
GlenB

[pfletch101]

Can you show us some screen shots in Footprints graph view of what you are describing - including a voltage trace? Can you also confirm what your solar configuration is - i.e. does your solar inverter feed into your main panel, with the CTs from your (presumably) two MTUs around the two live feeds from the inverter and the two live utility feeds? If this does not describe your setup, please give us a circuit sketch.

[glenb]

Attached is a diagram of my installation.  I will say that the MTU2 and MTU3 readings match the two inverters, and seem fine.  It is the MTU1 clipped onto the Grid that are not working correctly.
I assigned the Grid mtu as 'net' and the other two as 'generation'.  The Grid mtu does not work correctly as 'net' whether I turn the red dots on the CT's towards the grid or back towards the house.  It actually seems to work better with the CT dots towards the house...
I have verified that the generation CTs are quite close to the inverter output readings.

I know approximately what the house is consuming at 'idle', no major stuff on (well pump, water heater, etc).  less than 1 kw.
As the generation output increases so does the mtu1 or grid increase.  I have seen the generation go over 5kw, and the house
goes up as well, to 2500w or more(with nothing running in the house).
Not sure why this happens.
Also, the remote display only shows the total of the mtu2 and 3, not the 'net' as I thought it would... I do not see where the display gets it numbers from, I cannot figure out a combination of mtu readings that match the display...

[pfletch101]

Your drawing is slightly confusing, since you write that MTU2 and MTU3 are both monitoring inverters, but you show MTU3 on a line going to 'Workshop'. I assume that this is actually a remotely located second inverter, with no local loads. Your configuration sounds correct. The only problem resembling this I have seen described here resulted from the CT clips on the problem circuit not being quite closed round the wires they were on (apparently only a small deviation from 'perfection' was enough to cause a substantial problem). If this is not your issue, and it doesn't sound as if it is, I would raise a ticket on the main site and discuss it with the real experts.  

One other thought - I assume that your gateway is on a circuit coming off the main panel. People have certainly described problems with receiving the PLC signal from an MTU which is not in the same panel as the gateway circuit and/or not on the same 'phase'. Have you looked at the gateway's stats page (http://[Gateway IP address]/stats.htm) to check the signal quality for the three MTUs?