Hi All,
Not a Sense specific question, but with all the smart people here - I thought I would ask.
We had solar installed about a month ago. Due to some complications on how we installed, at the moment, I can’t have Sense monitor the production. But hopefully in the future.
We have an issue where we have 1 panel that sometimes is producing less than its neighbors. In working with the Solar company they say that it is an issue of the Utility Grid voltage hitting a limit and the inverter shutting down. They have not given me a great reason as to why its only 1 micro inverter, but…
So this statement of course led me to go look at the power quality chart. On many days (not all), we do now have this nice curve of voltage going up as production increases. I haven’t’ overlaid charts exactly to see that it 100% correlates with production but it seems close.
The solar company did make a change to the settings on the Microinverter, but it doesn’t seem to be helping. Is my next call back to them, or is there anything in these charts that say I should call the power company first, and if so, what am I telling them / asking them?
I don’t know enough about solar to know what’s going on, but it sure looks like your solar backfeed is lifting your local voltage level when your array is producing. I have a single inverter, not microinverters, and my inverter is only about 8 feet away from my panel / load center where we do the backfeed, but I don’t see similar:
My understanding is that inverters are designed to follow the phase and voltage from the grid, but in your case it looks like they are trying to take the voltage higher. I could see one of the microinverters kicking out it they are aiming for 125V for some mysterious reason and th grid keeps hanging at 120V. A couple questions - how far apart are your feedin and the inverters ? How much do you produce relative load.
Mine does the same, but I dont get one or any of my 14 micro’s shutting down at any time, but I do see the voltage difference during the day when generating solar power.
Thanks for the responses.
Additional details that may have been helpful - 23 Microinverters for a theoretical DC capacity of 8.5kW.
I had not thought about the distance between the inverters and the main panel. It is about 150’ between the micro inverters and where it taps into the line from the street. The array is on the roof of one building, but feeding the power to a second building.
On a whole, we consume more than we produce. During the day, around peak production (noon ish) we can see peak production of around 3500W and usually mid day, our consumption is around 800w unless we are cooking or something. So the time of the day with peak production, we have some of our lowest consumption during solar producing hours. I’m going to make a better effort this week to seeing if I can move things around so I can get Sense monitoring solar as well. Expect another thread from me shortly to ask a question about clamp placement for that…
The solar company did say that the micro inverters do need to “put out a slightly higher voltage than the grid to force power out” . I guess what i don’t quite understand is why the voltage “keeps” going up. I understand the concept of the Micro inverter’s needing the higher voltage, but unless the power from the street is changing as well, I guess I would just expect a little bump as the system started to produce. not necessarily tracking production. Or is this a strange Current / Resistance / Voltage equation due to the length of run between the solar panels and grid tie.
I don’t think the 3 other houses that are on our transformer have solar, but that makes me think that I should go confirm that there aren’t more houses on the transformer.
I was kind of wondering if multiple solar systems on a single feed could produce something like this as each array needs to “beat” out grid voltage, so do I 1 up my neighbor, then they 1 up me and so on.
The next time we have a sunny day, I’ll kill the array for a couple hours mid day and see if there is a change in the curve. But it does seem to be solar dependent
I would talk to your solar company about your voltage rise settings as well. Seems like this is a a concern when your microinverters are far away from the feed-in to the grid due to IR drop. It also looks like each microinverter in the the string might need its own voltage rise adjustment. Simple thinking - if the voltage rise adjustment is set for too high a resistance, you’ll get a voltage output curve from the microinverter that follows your current/power output. But usually the adjustments only dance the voltage by 10ths of volts, not 5V.
The ‘impedance’ as seen at the micro-inverter will be the sum of the grid impedance plus the impedance of wires from the main panel to the micro-inverters. The higher the impedance the higher the inverters must push the voltage to back feed the current. Added to this the inverters will shutdown (without a manufacturer and utility authorized override) at some UL defined upper (and lower) threshold (lower threshold is now subject to ‘ride through’ in later UL standards to help support the grid).
Solutions:
Make sure the impedance in the wires from the panel to the inverter is as low as possible — thicker (lower AWG) wires, or shorter wires. Keep the wire run under the panels as short as possible as the heat increased wire impedance. Make sure all wire splices / connections are really solid.
Make sure the impedance of the grid is sufficiently low – good luck with this it may require the utility to adjust transformers on your block or capacitor banks or upgrade wires. It will require you to find ‘the man that can’ in the utility. I’ve done this before - the key is gathering evidence that the grid is unable to accept the amps without the voltage being raised too far. For boarder line systems this is more likely to happen in spring just before AC session when solar output is a max, and the utility is raising voltage on some blocks to preparation for air conditioner loads kick in. This is where Sense could really help with Sense Labs input – it may be possible to graph grid impedence.
Override the threshold for overvoltage disconnect in the inverters.
Note to SENSE LABS:
In theory a SENSE can calculate ‘grid impedance’, this would be very useful data for such situations.
