MIDNITE THE ONE Inverter NEW
- Thread starter Nobodybusiness
- Start date
Let's see how my solar production is today. I put the inverter in standby, disabled AC coupling and moved my AC solar to my loads panel. Turned on the inverter and waited. Looks like SA is picking up solar production (Current and Voltage on the MPPTs) now. One bug in SA is dealing with negative number from the load now. It's still early in the morning but here is what I see now.
You can see the blue graph to the right spiked up to 400MW due to negative numbers I believe...
You can see the blue graph to the right spiked up to 400MW due to negative numbers I believe...
Adam De Lay
Solar Wizard
Yeah, I've seen the battery value showing as "discharging" when it's actually charging and visa-versa.
I did see on someone else's install that "Mil" rating when they have AC Coupling enabled as well. Something's confused when AC Coupling is enabled.
This inverter hates me. Somehow Solar PV data is not coming out right again. At the time of this picture, my DC clamp showed at least 2A of DC solar and SA shows nothing. Load power is wrong. The battery power of 215W charging the powerflo16 is wrong as well. I have a Victron shunt that can measure in realtime and definitely not charging. It is actually discharging at -33W constant. This seems to be the power draw of the inverter as it does not take from the AC grid it looks like.
Adam De Lay
Solar Wizard
That's what I saw on the other system using SA with AC Coupling. I wonder if having that enabled messed up the data even though you disabled it? It's gotta be how SA is interpreting the data unless you're seeing the same thing in your HA feed and the MidNite app?
orangesolar53
New Member
It seems my solar production was smoother today. The first big spike downwards was when Midnite updated the FW on my inverter. The second was when I changed power control settings in the inverter.
I too have noticed bad data being presented on the various MN apps/website for solar production. Today was the first clear sunny day in a while here, so I could monitor things without other variables. Just posted over on the MN AIO forum about this, but in short I found several discrepancies with the power numbers being reported.
Similar to @goldserve, the current values displayed for the MPPTs are flakey. Often showing 0 at the same time that it shows PV power being produced??? I don't think the voltage or current readings we see in the app can be trusted; at best just out of sync with each other to just plain bogus. There are also issues with total PV power being off, and the various power totals not adding up with each other.
Some of this might fall into the general category of the app's power numbers "needing work", like the MWH being reported. But I'd encourage folks here to go the the MN AIO forum and post the problems you've seen there too, even if it is just a "me too" reply. I think Logan sees posts over there more so than he does here.
I too have seen drop outs in PV production during the day. Not just the MPPT running its test cycle again (some 5-20 secs, see those too) but for many minutes. For example today at 11:10AM thru 11:20AM PV production dropped to near 0, then returned to its previous level. This according to the app's graph. I did not happen to be monitoring at that time, but there were no clouds or other obstructions to lower PV output. So not sure if it really happened, or if there is a"glitch" in the MPPT FW, or just another app reporting bug...
FWIW, my "installer" said MN told him that MN is working on getting SA on the AIO. They actually plan on selling the HW setup to run/connect it at some point. He also reported that Logan said he thought it would be OK to extend the CTs with a cat5 or 6 cable, as long as you aren't cutting and soldering leads (guess they have seen some problems with people rolling their own CT cables).
Otherwise, the AIO has been running well. No repeat of it shutting itself off again.
Similar to @goldserve, the current values displayed for the MPPTs are flakey. Often showing 0 at the same time that it shows PV power being produced??? I don't think the voltage or current readings we see in the app can be trusted; at best just out of sync with each other to just plain bogus. There are also issues with total PV power being off, and the various power totals not adding up with each other.
Some of this might fall into the general category of the app's power numbers "needing work", like the MWH being reported. But I'd encourage folks here to go the the MN AIO forum and post the problems you've seen there too, even if it is just a "me too" reply. I think Logan sees posts over there more so than he does here.
I too have seen drop outs in PV production during the day. Not just the MPPT running its test cycle again (some 5-20 secs, see those too) but for many minutes. For example today at 11:10AM thru 11:20AM PV production dropped to near 0, then returned to its previous level. This according to the app's graph. I did not happen to be monitoring at that time, but there were no clouds or other obstructions to lower PV output. So not sure if it really happened, or if there is a"glitch" in the MPPT FW, or just another app reporting bug...
FWIW, my "installer" said MN told him that MN is working on getting SA on the AIO. They actually plan on selling the HW setup to run/connect it at some point. He also reported that Logan said he thought it would be OK to extend the CTs with a cat5 or 6 cable, as long as you aren't cutting and soldering leads (guess they have seen some problems with people rolling their own CT cables).
Otherwise, the AIO has been running well. No repeat of it shutting itself off again.
I'm glad you are able to corroborate some of the things I am see but what makes it difficult to track down is the fact that there is no reliable and trust worth data! I was trying to find some equipment to log DC current over time so I can see what the MPPTs are doing but that's much harder than CT clamp meters with AC voltage.
I am thankful I have a bigger AC coupled array or else, I would be pretty pissed my upgraded inverter is not really an upgrade and costing me solar production!
@Steve777 Out off curiosity, are you running Grid Tie, sell back as poriority? I feel like a lot of issues that aren't seen stems from a setting or something that the guys at MNS are not catching easily. They told me they primarily run off-grid so it's not easy testing on-grid functions.
orangesolar53
New Member
Pardon my ignorance, but what is the advantage of "grid-tie, sell back as priority" over "self consumption" with "discharge end SOC(on-grid)" set to 100%?
I was hoping that grid-tie sell back as priority would sell to grid and only use the excess PV power over the grid sell-back limit to charge batteries. I'm not seeing this behaviour so clarification on the exact differences would be useful.
I have tried Grid Priority for a bit, but with the cloudy weather over most of the last two weeks since my AIO has been up and running, can't say that I really understand it or have learned its in and outs. Mostly ran it to see that sell back work and to test out some more settings.
At first I though Grid Priority would help keep the batt at a higher SOC while still running loads. It did seem to improve that one day, but not in general; at least not more so that Self Consume mode. I did not run it that long since even though I have a sell back agreement with my PoCo, it was for a 120v sell back and did not want to raise any eyebrows with them with the AIO.
I did go back to Self Consume mode. And that seems to be working OK, although from what I have seen there appear to be inconsistencies. Have not seen enough to figure out when it direct PV power to loads and batt charging and when not. And an example from today (from memory), getting 2KW of PV, 0.6KW to batt charging, and pulling .3KW from the grid and sending 1.7 to the load. OK, not bad, but if it is prioritizing self consumption, why pull anything from the grid? Send less to the batt and take nothing from the grid, would appear to make more sense, to me at least.
There are other instances, where the AIO seems to be making power distribution decisions which appear to be contrary to the overall setting (or I just don't know the settings well enough to make them correctly).
In general, until this spring, when the rest of my PV gets installed, I won't be able to really test some of these setting/features.
I think you are right, most of the testing seems to have been done in off-grid or grid assist but not selling back it seems. I suspect there are still a bunch of bugs there to be discovered.
Your experience matches mine almost 100%. I can’t quite figure out what setting does what. There are inconsistencies in its behavior. Anyways, the 0.3W constant grid draw is a setting I believe. Default is 300w and lowest can be set to 30w. I believe this is buffer for zero sell back but also the point around where loads are considered zero’d out.
Adam De Lay
Solar Wizard
Once spring arrives and I have more consistent sunlight on the panels, I'll try and see what mine are doing as well. I'll admit I didn't look that closely at the actual production trend in the fall.
I don't have any sell-back agreement, so I'm afraid I won't be any help there with those options.
I don't have any sell-back agreement, so I'm afraid I won't be any help there with those options.
I'm running Self Consumption profile now with battery discharge to 100% while on-grid and it is charging up my batteries consistently for the past two days now. I think I just had a grid outage event but the critical loads was not seamless. It actually shut off and then after 5 seconds, it turned on. Has anyone tested shutting off their grid in to see how the loads are handled?
Just tried flipping off the 125A breaker feeding grid to the inverter to simulate a grid down. Seemed to work fine. No flicker or loss of power at the transition, loads kept running, grid lite on the inverter went off as expected. Turned the breaker back on after a couple mins. Things came back as expected, though it took a couple of mins to get reattached to the grid.
Possible differences: My batt (one Powerflo16) was at 99% SOC. Loads were fairly low at this time, ranging between 1 and 2KW.
Possible differences: My batt (one Powerflo16) was at 99% SOC. Loads were fairly low at this time, ranging between 1 and 2KW.
I have one powerflo16 too but I guess the difference is that I have AC coupled enphase on the loads panel and not ac coupling relay and i'm generating more power than loads...
That could explain this spike as the inverter tried to cope with all the extra power.
Okay, that was it. I just disconnected my ac solar first before simulating a grid down and it was a seamless transfer to backup. Now I need to test if ac coupled solar is less than loads. Since my battery was only at 96%, it should have dumped the power and ramped up frequency. More to test…
I think switching my working mode to Self consume, allowing export (grid feed-in) was the key to get my PV production back to full potential. The Grid feed-in priority mode is wonky and I always suspected the throttling of PV was just bad software. @Steve777 Something to check but when switching working modes, check the max feed-in setting is not changed to 0.
I have noticed that the PV production is at time lower than I would expect. Today for example, PV ran at ~2Kw most of the morning, But in the afternoon it was running ~0.6Kw even though it was just as sunny if not more so than the AM (based on visuals and my weather station showed a higher W/m**2 reading in the PM). And there was plenty of load to push the extra power to. I do have grid feed-in off now, so think I will give it a test with it turned on, and a safe max level, to see if it changes anything.
Turned on grid feed-in (with limits), in self consume mode and tried it out for a few days. As @goldserve hinted, it does improve overall PV power production. With feed-in on, I am now seeing fairly even power production throughout the day, and daily production has improved as well. Unlike with feed-in off, where as mentioned above, afternoon PV production was reduced.
Not sure if it has to be feed-in, or if adding some other loads to the mix (such as powering the loads this side of the CTs) would have the same effect. But having feed-in on greatly improves overall power production.
Not sure if it has to be feed-in, or if adding some other loads to the mix (such as powering the loads this side of the CTs) would have the same effect. But having feed-in on greatly improves overall power production.
What you say is true, but wasn't the case on the afternoons I was watching.
There were plenty of loads, and the inverter was pulling in ~.8KW from the grid and some from the batt too, to feed them. It could have been pulling that grid & batt energy from the PVs, and has been doing that these last few days while feed-in was on.
I suppose could turn feed-in off again, and see if it goes back to the old behavior.
There were plenty of loads, and the inverter was pulling in ~.8KW from the grid and some from the batt too, to feed them. It could have been pulling that grid & batt energy from the PVs, and has been doing that these last few days while feed-in was on.
I suppose could turn feed-in off again, and see if it goes back to the old behavior.
I set 30W as buffer to draw from grid and rarely does it only draw 30W so that setting is off but this capture from the envoy CTs show that feeding into the grid on a load release is very much possible. Luckily I have a NEM agreement but i'm just warning you guys out there...
Talked to Rick at MidNite he recommended if you were to run zero export and not want to backfeed you need to set constant draw from grid if you have a lot of heavy loads at about 500w continuous
Similar threads
