zanydroid
Solar Wizard
OK. Custom connectors also implies another special use disconnect tool to keep around. LOL.
How to AC couple Enphase IQ8 w/Battery Storage to Schneider XW Pro 6848 w/rack batteries, off-grid?
- Thread starter Gubman
- Start date
fromport
Solar Addict
I have more than 1 MC4 connector go bad. If you need to reconnect them a couple of times, one might fail.
Having a removable pig tale with an mc4 will make you can get a new pig tale instead of a new micro inverter.
fromport
Solar Addict
You can use
no specials tools needed although the official tool makes it a breeze
Last edited:
The AC connector is a custom connector as well, and I believe the disconnect tool from enphase does both the AC and DC connectors.
zanydroid
Solar Wizard
I thought there was a code requirement for it to be harder than that to remove
zanydroid
Solar Wizard
Oh right, Enphase MC4 is right on the body and not changeable? I messed up one MC4 on my non Enphase micro during install, fortunately it was on a pigtail so I could just lop it off and crimp on a new one.
fromport
Solar Addict
Hedges
I See Electromagnetic Fields!
- Joined
- Mar 28, 2020
- Messages
- 21,393
Like using SMA "connector savers" on instruments. Brass connectors good for 50 mating cycles, stainless 500 cycles.
With the connector savers, could reach 2500 to 250k uses before the instrument has to be sent in for repair.
I've got some REC 325W panels with Stabuli MC4 so I'll use those for the tests. Couple of them in parallel and oriented differently for flat power curve during the day.
I'm hoping my scopes support "track and trend" which would let them graph calculated signals like frequency and RMS current/power. Otherwise, 2Meg samples and transfer to PC for processing (an excuse to try the free version of LeCroy's "Maui Studio", the software of their scopes.) I've been learning about features watching Teledyne-LeCroy's free webinars, and at least some of the functions are on my Tek TDS-784D scopes.
It'll be slow going in spare time, but hopefully I get scope traces and come up with parameters that make these things work. Using Sunny WebBox to configure Sunny Island I should be able to alter frequency limits and ramp rates, find how the microinverters can respond. Offline from the house, so the Official Loads Tester doesn't get annoyed (except by my absence, having disappeared into my dark cave.)
You're correct, the XW is raising frequency and voltage to stop micros from producing, anything is allowed in off-grid mode, no regulations to be complied with. This is not a violation of the on-grid behavior according to the selected grid profile. The 62 Hz applies only to on-grid.
The Hoymiles PV backfeed into off-grid was around 1200W - much higher than the 300W I was using for testing IQ inverters - when the frequency jumped to 66 Hz off-grid. As you've said, the XW will do anything to curtail production to prevent a shutdown.
XW uses the selected on-grid profile as a starting point for off-grid operations. I can change the various power curtailment properties, but it is not clear if the changes apply also to the on-grid profile or only to the off-grid use. Have an inquiry into Schneider SE to get clarification.
There was a 1 sec delay how the XW would respond to a power curtailment. I have set it to 0 and redoing all tests, getting much better results with frequency shift. Need to wait for full PV production on Sunday to repeat the Hoymiles test.
I re-installed the latest firmware for the InsightHome gateway and can modify the grid profile properties as shown in the AC coupling guide and the latest XW manual.
Changed p(f) PreDisp to a 10 point non-linear curve starting at 60.1 and stopping at 62 Hz. Also changed response delay from 1 sec to 0. Big difference in AC coupling. The single IQ8+ inverter test at 300W with no load runs fine for any SOC from 90% to 98%, not much difference. At 61 Hz the IQ8+ shuts off and it always restarts after 5 minutes. I do see slightly higher shut-down frequencies with higher SOC, nothing significant.
SOC is not an absolute measure, SOC of 90% for a 5 kWh battery leaves 500 Wh for charging, while SOC = 98% for a 50 kWh battery allows 1000 Wh charge reserve.
There are properties in the MODBUS specs for power gradient over time, but I have not found them yet in the InsightLocal GUI,.
Would be great to specify FW power curtailment over time.
What is missing is a definable SOC hysteresis, it is 1%, causing frequent "nuisance" cycling.
Planning to test power curtailment tied to battery voltage, there is a definable hysteresis for the on/off curtailment voltage.
Interesting question and helpful link.
Enphase:
When I connect 2-3 IQ8+'s to an XW-Pro, they start producing power in off-grid mode after 5 minutes.
When I connect a live array of 14 IQ8+'s they never start producing, independent of the PV power available.
There is no measurable frequency shift during the anti-islanding testing, 60.0 Hz constant. But I am observing an increase in the power factor (0.02) and off grid consumption (6W) of the 14 IQ8+'s. They never come on-line to produce off-grid.
Will try to split the 14 IQ8+'s into 2 sub-arrays of 7 IQ8+'s each.
Hoymiles:
Hoymiles inverters (7 HMS-700) always start producing power in off-grid mode, independent of the PV power available.
When SOC has reached it's defined limit, XW-Pro inverter performs near perfect balancing of AC coupled power produced, internal power needed and power to loads by keeping the frequency at a level (60.1 to 60.49 Hz) to operate in steady-state without cycling.
The XW Pro does not have sophisticated graphing available to show the role of off-grid frequency modulation and power factor in any of the displays.
fromport
Solar Addict
I think the homies are more comparable with the IQ7 series. Grid tie only.
The fact that iq8 can generate a micro grid from scratch under the right circumstances is impressive and requires a lot of computing power. I have been told that the FPGA in the iq8 is about 100 times faster than IQ7.
Enphase basically has 2 situations in mind.
Work with the grid or with their own off grid setup.
They never intended to let it work with any other equipment than their own and grid.
I think you can not blame the for that approach.
I hope that Enphase with firmware updates could make it more compatible with other grid providing inverters.
I have been told there is a "sunnyboy" grid profile setting for iq8's which improve co-operation with eg sol-ark/luxpower in off grid mode that is much better than the standard rule 21 profile.
You might want to check that out.
The fact that iq8 can generate a micro grid from scratch under the right circumstances is impressive and requires a lot of computing power. I have been told that the FPGA in the iq8 is about 100 times faster than IQ7.
Enphase basically has 2 situations in mind.
Work with the grid or with their own off grid setup.
They never intended to let it work with any other equipment than their own and grid.
I think you can not blame the for that approach.
I hope that Enphase with firmware updates could make it more compatible with other grid providing inverters.
I have been told there is a "sunnyboy" grid profile setting for iq8's which improve co-operation with eg sol-ark/luxpower in off grid mode that is much better than the standard rule 21 profile.
You might want to check that out.
zanydroid
Solar Wizard
Thanks for doing the testing!
I can believe they are way faster but I don’t think they’re using FPGA. Can you share your source on that?
Except nonzero number of people here purchased them for that future purpose assuming they had similar level of compatibility to industry average. And the only warning about this issue is the forum, which could not have known about the issue before people got burned trying to get it to work.
zanydroid
Solar Wizard
Lately I can imagine some scenarios where being forced to operate quickly can be a liability. Imagine if the grid former expects to be able to react faster than a typical grid tie inverter that it is managing. GTI do not have an intrinsic need to operate quickly to respond to load while a grid former does, as does an AIO.
fromport
Solar Addict
The FPGA part is what I would think would be used.
The 100x factor was given by an Enphase instructor on one of the courses I did.
Hedges
I See Electromagnetic Fields!
- Joined
- Mar 28, 2020
- Messages
- 21,393
I'm guessing the anti-islanding testing of others affects each one trying to connect. I'll see what I can see with voltage waveform and FFT.
This is where grid-forming inverter has to be stiff enough, that may vary even between LF inverters, and Enphase is under no obligation to have grid-forming inverters pass the test.
zanydroid
Solar Wizard
Why an FPGA vs an ASIC?
fromport
Solar Addict
Enphase IQ8 Microinverters are the industry’s first grid-forming microinverters. The brain of the semiconductor-based microinverter is our proprietary application-specific integrated circuit (ASIC) which enables the microinverter to operate in grid-tied or off-grid modes.
source: enphase docs
Could someone try the IQ "sunnyboy" grid profile together with the XW?
Just tried to AC couple the 14 IQ8's with AC coupling switched off in the XW-Pro. They still do not start producing power.
Will try to test other grid profiles, but Enphase profiles are limited by region and I do no see a profile for "sunnyboy".
Last edited:
CyboEnergy had gridforming micro inverters for several years. But not a combination grid-tied plus micro-grid forming like Enhase.
Last edited:
Similar threads
