diy solar

diy solar

Solved! SolarAssistant with 6x EG4-6500EX and 9x EG4-LLv2, individual connections

F

fmeili1

Solar Enthusiast
Joined
Jan 19, 2022
Messages
334
Location
Arizona, Mohave County
I’d like to share my findings with you so you may not run into the same problems.

After trying for quite a while, I finally have reliable communication between SolarAssistant (SA) and all inverters and batteries.

My setup:
  • 6x EG4-6500EX-48 inverters in split phase
    • 3 inverters per phase
    • all manufactured in 09/2022
    • firmware (stationary) updated in all inverters to versions DSP=79.63 and MCU=61.12
  • 9x EG4-LLv2 batteries (all bought in 11/2022)
    • 2 racks, one with 5 and one with 4 batteries
    • firmware updated in all batteries to version Z01T13
  • common neutral (no SDS); bonding screws removed in all 6 inverters
  • still in building phase (still no PV panels connected, no load so far and no AC-in connected but the AC-outs are already connected together for the three L1 and three L2; just using the pre-charged batteries to do some testing)
My goals for SA:
  • Get ALL possible data from ALL inverters and ALL batteries in SA in a reliable way.
Summary:
  • It’s not possible to get ALL available data if SA is only connected to the master inverter (and master battery). Inverter values like temperature, BUS voltage and MPPT2 are only available if connected individually to SA (it’s documented in SA).
  • To get ALL inverter data, each inverter has to be connected individually via a separate USB-to-RS232 cable to SA (so I need 6 USB-to-RS232 cables!).
    • EG4 inverter to EG4 battery communication (RS485) can’t be used when you want to see ALL data in SA. Instead you have to set the inverter battery type setting (program 05) to “USE” in all inverters (even in the master inverter) and start with #2 dip switch address for the first battery and up (as documented in the SA manual for EG4)!
    • Use “working” USB-to-RS232 cables (some types are NOT working), these adapter cables have to be bought separately and needs to be connected to the special EG4 serial DP9-to-RJ45 cables (included for each inverter) !
  • To get ALL battery data, each battery has to be connected individually via RJ45. Because it’s RS485, all of the batteries could be wired separately to one passive RS485 hub(in my case I use two cascaded RS485 hub’s, one in each rack and they are connected together)
    • One USB-to-RS485 adapter cable (Signature Solar include one adapter usually in your battery order) has to be connected from the passive RS485 hub to SA.
  • Use an external powered USB hub to connect all individual USB adapters to SA (in my case 7; six for each inverter and one for the batteries)
Some details:

I had very unreliable and unstable SA connections at the beginning. Also the values are not updated reliable (Sa showing old values over long period of time). It looks like the Pi has not enough power on the USB connector to power 7 external USB adapters with a passive USB hub. So an external powered USB hub is a must! Also I found that one type of USB-to-RS232 adapter cable type is not working reliable even with an external powered USB hub. I’ve changed to a different adapter and this works perfect, now.

For what reason ever, this USB-to-RS232 adapter cable is working (https://www.amazon.com/dp/B000A1KD1K)!
For what reason ever, this USB-to-RS232 adapter cable is NOT working (https://www.ebay.com/itm/255666926991)!

The working USB-to-RS232 cable uses a pl2303 chipset and the cable which fails uses a ch341 chipset. But that's not the explanation because the USB-to-RS485 cable for battery communications uses also a ch341 chipset (with the same driver in the SA Pi) and this adapter is working without issues. So it seems it has to do with the additional electronic in the failed USB-to-RS232 adapter... but I'm not sure. The kernel log does not really helps here.

The working adapter shows (e.g.) the following information in the kernel log of the SA Pi (for each adapter):
  1. new full-speed USB device number 16 using xhci_hcd
  2. New USB device found, idVendor=067b, idProduct=23a3, bcdDevice= 6.05
  3. New USB device strings: Mfr=1, Product=2, SerialNumber=3
  4. Product: USB-Serial Controller
  5. Manufacturer: Prolific Technology Inc.
  6. SerialNumber: xxxxxxxxxxx
  7. pl2303 converter detected
  8. pl2303 converter now attached to ttyUSBx
The failing adapter shows (e.g.) the following information in the kernel log of the SA Pi (for each adapter) - as I already mentioned, the USB-to-RS485 adapter is showing the same log, but it's working - so the reason is not the chipset and also not the missing "break control" which is mentioned in the log:
  1. new full-speed USB device number 20 using xhci_hcd
  2. New USB device found, idVendor=1a86, idProduct=7523, bcdDevice= 2.54
  3. New USB device strings: Mfr=0, Product=2, SerialNumber=0
  4. Product: USB2.0-Ser!
  5. ch341-uart converter detected
  6. ch341-uart ttyUSB1: break control not supported, using simulated break
  7. ch341-uart converter now attached to ttyUSB1

The passive RS485 hub's I'm using are from growatt (e.g. SignatureSolar offer these, https://signaturesolar.com/growatt-rs485-can-hub-with-8-ports/).
I'm using this external powered USB hub with 10+1 ports (https://www.amazon.com/dp/B0BH252R3J)

Open issues:
  • The inverters serial number are not showing reliably (sometimes I see only 2 or only 4 and they will never displayed if not filled after first connect).
  • In my situation this is really annoying because I have no chance to identify which of the inverters #1 to #6 does correspond to which physical unit (this is discussed in a different thread in detail)
In this picture the individual inverter temperatures and BUS voltages are shown
SA-temp-after-restart.png

In the following picture the "missing serial number problem" is shown - in this example only the first 2 inverters SN are displayed:

SA-serial-no-missing.png

As an example, the following grafana screenshot shows the temperatures of each inverter (I've activated the "show dot" just to be sure that I receive reliable and updating data) represents the temperature values of all 6 inverters in idle (no load, no PV, no AC-in) together with the ambient temperature (I receive this data via SA MQTT bridge in my OpenHAB for further processing and visualization).
Grafana-temp-after-restart-2.png

Off topic:
I need the reliable temperature curves because I'm still in the phase of modding the really loud fans in the EG4-6500EX inverters to make them much more silent via temperature controlled PWM fan control - but this is a different story and I will post the results when it's done. Without reliable inverter temperature I will not be able to proof the modding and do a comparison with un-modded inverters.
 
Last edited:
Awesome! Thanks for sharing!

fmeili1 said:
Open issues:
  • The inverters serial number are not showing reliably (sometimes I see only 2 or only 4 and they will never displayed if not filled after first connect).
  • In my situation this is really annoying because I have no chance to identify which of the inverters #1 to #6 does correspond to which physical unit (this is discussed in a different thread in detail)
Click to expand...
I do know that myself and others have requested of Pierre to give a way to manually identify the different inverters in SA. While there’s no current timeframe, it was mentioned that it will be fixed at some point.
 
w8dev said:
Awesome! Thanks for sharing!


I do know that myself and others have requested of Pierre to give a way to manually identify the different inverters in SA. While there’s no current timeframe, it was mentioned that it will be fixed at some point.
Click to expand...
Yes, Pierre gave me the same answer.

I think it's even more urgent to get this feature because the serial numbers are not shown reliable in the SA overview page (I have not noticed this problem when I've done the first tests with just 4 inverters). With the serial numbers showing in SA it would be possible to manually identify the physical inverter but without you have no chance. Now I try to start the inverters one by one with a longer delay between each other to watch in SA which will be the next with "green" status and write down the inverter number... that's annoying - at least. Btw. even more problematic is the situation when receiving the inverter values via the SA MQTT bridge because the serial number are not published over the message broker at all (even if they would be available in SA), so there is NO chance to identify the physical inverter in an automated way.

I hope this bug about not displaying all serial numbers will be fixed soon. Also this new feature to assign an inverter number permanently to a specific (and always the same) physical unit will hopefully also arrive soon. When this will be fixed, I would also be able to use the data via MQTT in a meaningful way.
 
Last edited:
I want to let you know, that I also found a solution about the usually ugly cable wiring for parallel inverter communication and current sharing between the inverters which was even worse after connecting 6 inverters individually to SA. I found a low profile, open slot, cable channel (https://www.amazon.com/dp/B07C9ZWH1H) which just fits (with a bit of pressure) between the vertical short EMT conduits, which runs between the cable channel and the inverters, and the wall. They have a height of just 0.8 inches and have just enough width to wire 6 inverters with parallel communication, current sharing and individual SA communication via RS232 cables. Now the system looks much cleaner than before.

Monitoring-Cable-Channel.jpg
 
Oldtimmer69 said:
Any light flickering at all?
Click to expand...
As I wrote in my original post, I'm still in building phase and PV, AC-out and AC-in are still not connected - just the batteries (which had 50% charge when arrived). In the last picture you may notice, that the 200A breakers in the AC-in panel (on the very right side) and the AC-out panel (left of the AC-in panel) are still not connected! Until now, I'm only able to do some "dry" (idle) tests without load and without charging with the remaining battery charge. I still wait (since 6 weeks) for the electrician to connect my AC-in (permanent to the grid) and AC-out (to the manual transfer switch to feed the existing house panel). because this is the only work, which I don't want to do by myself.

But until the electrician will show up, I've decided to use the time to finalize the SA communication, inverter and battery setup and tuning, firmware updates, try to reduce fan noise level, cleanup wiring, install SoftStart for 5ton HVAC, install mini splits in the garage, etc. - but no load tests so far!

I can barely wait for the first real load tests. As soon as I will have first results, I will post them!
 
fmeili1 said:
As I wrote in my original post, I'm still in building phase and PV, AC-out and AC-in are still not connected - just the batteries (which had 50% charge when arrived). In the last picture you may notice, that the 200A breakers in the AC-in panel (on the very right side) and the AC-out panel (left of the AC-in panel) are still not connected! Until now, I'm only able to do some "dry" (idle) tests without load and without charging with the remaining battery charge. I still wait (since 6 weeks) for the electrician to connect my AC-in (permanent to the grid) and AC-out (to the manual transfer switch to feed the existing house panel). because this is the only work, which I don't want to do by myself.

But until the electrician will show up, I've decided to use the time to finalize the SA communication, inverter and battery setup and tuning, firmware updates, try to reduce fan noise level, cleanup wiring, install SoftStart for 5ton HVAC, install mini splits in the garage, etc. - but no load tests so far!

I can barely wait for the first real load tests. As soon as I will have first results, I will post them!
Click to expand...
I did see the panels not connected yet.
I'm definitely interested in seeing your performance out of these. I have 4- 6500's, and 15- eg4 batteries.
 
Oldtimmer69 said:
I did see the panels not connected yet.
I'm definitely interested in seeing your performance out of these. I have 4- 6500's, and 15- eg4 batteries.
Click to expand...
I'm curious how satisfied you are with your system because I'm a little nervous how these 6 inverters will work when everything will be finished...
  • How long has your system been in operation?
  • Does it work reliable?
  • Do you use SBU or SUB mode?
  • Do you used EG4 specific inverter/battery communication or are you using them as "USE" in program 05 without communication.
  • Have you noticed that the first inverter of L1 (usually your master) and the first inverter of the L2 pair will get warmer than the others (I notice about 3-5 °C more for these - at idle).
  • Because you've asked, I guess you have light flickering! Is it very disturbing or acceptable?
Thanks!
 
I've had my system up since this past Thursday, but my #1 inverter has some problems right out of the box. Signature Solar is working on it so i hope to know what they're going to do about it by the end of this week.

So i only have 2 inverters run. This will not run my entire home. All neutral /ground bonds are correct in all my panels but im still have TERRIBLE light flickering, and pulsing when the washing machine is going. Ive replaced LED bulds to different ones but no change.
It is unacceptable.. If it can not be resolved I'll purchase something else.
Ive been running the Growatt 12k for a couple of years but i couldn't run my whole home on it either without only using one 240v circuit at a time. But never had the light flickering.
Updating firmware now to see if it corrects the flickering and pulsing.

I use SBU i only want to use grid power as a last resort to charge batteries if we've had some long periods of no sun.

Ive been just using the EG4 setting until i get everything else going. I'll probably change to USE so i can setup my limits on #12, #13 #26, #27

I cant really tell about the heat yet.. Lol

Hopefully everything works out and the system performs as it should and I'll be happy with it. There's a few of these systems i see online that are working well so im hopeful. SS is working to fix the known issue.. And Customer service seems to be getting better ( and faster) to finding resolutions to the problems.
 
Oldtimmer69 said:
I've had my system up since this past Thursday, but my #1 inverter has some problems right out of the box. Signature Solar is working on it so i hope to know what they're going to do about it by the end of this week.

So i only have 2 inverters run. This will not run my entire home. All neutral /ground bonds are correct in all my panels but im still have TERRIBLE light flickering, and pulsing when the washing machine is going. Ive replaced LED bulds to different ones but no change.
It is unacceptable.. If it can not be resolved I'll purchase something else.
Ive been running the Growatt 12k for a couple of years but i couldn't run my whole home on it either without only using one 240v circuit at a time. But never had the light flickering.
Updating firmware now to see if it corrects the flickering and pulsing.

I use SBU i only want to use grid power as a last resort to charge batteries if we've had some long periods of no sun.

Ive been just using the EG4 setting until i get everything else going. I'll probably change to USE so i can setup my limits on #12, #13 #26, #27

I cant really tell about the heat yet.. Lol

Hopefully everything works out and the system performs as it should and I'll be happy with it. There's a few of these systems i see online that are working well so im hopeful. SS is working to fix the known issue.. And Customer service seems to be getting better ( and faster) to finding resolutions to the problems.
Click to expand...
Because of time constraints, the inverters and batteries have been sitting unpacked in the garage for more than 4 months. I am now happy that they all work (at least at idle) after initial startup (programming, flashing, paralleling, etc.) - I've feared the same what happened to you with one of the inverters (one or more not working out of the box).

I think you plan to run your whole house with the inverters as soon as you'll get the one unit fixed?! Do you also have a 200A service? My plans are the same as yours, use utility as a last resort, just if SOC too low and no solar. With 12 batteries, I think you will have a realistic chance to not using the utility at all - my 9 batteries may not enough, I will find out...

About the light flickering problem - have you tried to disconnect some other consumers while your washer runs and the lights flickers (e.g. notebook, tablet, phone chargers, 3D printers or some other low energy consuming (but switching) power supplies? I read, that just one "dirty" (cheap, unfiltered) switching power supply is able to interfere with the HF-inverter unit and may produce a lot of noise to your inverter's output signal which may cause light flickering.

I wish you good luck that your inverter will be fixed soon.
 
fmeili1 said:
Because of time constraints, the inverters and batteries have been sitting unpacked in the garage for more than 4 months. I am now happy that they all work (at least at idle) after initial startup (programming, flashing, paralleling, etc.) - I've feared the same what happened to you with one of the inverters (one or more not working out of the box).

*Im glad they all worked for you, that's a plus.

I think you plan to run your whole house with the inverters as soon as you'll get the one unit fixed?!

*Yes, i plan on running everything. 2 - inverters wont start my air compressor maybe 4 will if not, Im going to install a 100amp transfer switch for my shop so when i want to use the air compressor ( 5hp 25amp) i can just switch to grid then back when im done. Other than that i should be good to go.

Do you also have a 200A service?

*I have a 200amp grid service, but i only installed a 125amp panel for all my 240v circuits, ( 4- 60 amp breakers, 1 from each inverter) then a smaller 240v panel that has all my 120v circuits. ( this is the panel i started with when i first installed solar)

My plans are the same as yours, use utility as a last resort, just if SOC too low and no solar. With 12 batteries, I think you will have a realistic chance to not using the utility at all - my 9 batteries may not enough, I will find out...

*I had 9 batteries, and it just wasn't enough when we had cloudy days especially in the winter i don't get sun but 4 hrs or so. If the sun came out every day it would be good. So i now have 15, and im buying one more.

About the light flickering problem - have you tried to disconnect some other consumers while your washer runs and the lights flickers (e.g. notebook, tablet, phone chargers, 3D printers or some other low energy consuming (but switching) power supplies? I read, that just one "dirty" (cheap, unfiltered) switching power supply is able to interfere with the HF-inverter unit and may produce a lot of noise to your inverter's output signal which may cause light flickering.

Since we've been on solar for a few years, we don't keep ghost devices plugged in, and i have a few strips on a few devices so i can keep them off when not in use. We haven't been washing any clothes at night so we can keep the lights off.. Lol not the ideal situation but wors for now.

I wish you good luck that your inverter will be fixed soon.
Click to expand...

Thanks and good luck with your build. Keep us posted as you go. It put my replies above under your questions.
 
fmeili1 said:
I’d like to share my findings with you so you may not run into the same problems.

After trying for quite a while, I finally have reliable communication between SolarAssistant (SA) and all inverters and batteries.

My setup:
  • 6x EG4-6500EX-48 inverters in split phase
    • 3 inverters per phase
    • all manufactured in 09/2022
    • firmware (stationary) updated in all inverters to versions DSP=79.63 and MCU=61.12
  • 9x EG4-LLv2 batteries (all bought in 11/2022)
    • 2 racks, one with 5 and one with 4 batteries
    • firmware updated in all batteries to version Z01T13
  • common neutral (no SDS); bonding screws removed in all 6 inverters
  • still in building phase (still no PV panels connected, no load so far and no AC-in connected but the AC-outs are already connected together for the three L1 and three L2; just using the pre-charged batteries to do some testing)
My goals for SA:
  • Get ALL possible data from ALL inverters and ALL batteries in SA in a reliable way.
Summary:
  • It’s not possible to get ALL available data if SA is only connected to the master inverter (and master battery). Inverter values like temperature, BUS voltage and MPPT2 are only available if connected individually to SA (it’s documented in SA).
  • To get ALL inverter data, each inverter has to be connected individually via a separate USB-to-RS232 cable to SA (so I need 6 USB-to-RS232 cables!).
    • EG4 inverter to EG4 battery communication (RS485) can’t be used when you want to see ALL data in SA. Instead you have to set the inverter battery type setting (program 05) to “USE” in all inverters (even in the master inverter) and start with #2 dip switch address for the first battery and up (as documented in the SA manual for EG4)!
    • Use “working” USB-to-RS232 cables (some types are NOT working), these adapter cables have to be bought separately and needs to be connected to the special EG4 serial DP9-to-RJ45 cables (included for each inverter) !
  • To get ALL battery data, each battery has to be connected individually via RJ45. Because it’s RS485, all of the batteries could be wired separately to one passive RS485 hub(in my case I use two cascaded RS485 hub’s, one in each rack and they are connected together)
    • One USB-to-RS485 adapter cable (Signature Solar include one adapter usually in your battery order) has to be connected from the passive RS485 hub to SA.
  • Use an external powered USB hub to connect all individual USB adapters to SA (in my case 7; six for each inverter and one for the batteries)
Some details:

I had very unreliable and unstable SA connections at the beginning. Also the values are not updated reliable (Sa showing old values over long period of time). It looks like the Pi has not enough power on the USB connector to power 7 external USB adapters with a passive USB hub. So an external powered USB hub is a must! Also I found that one type of USB-to-RS232 adapter cable type is not working reliable even with an external powered USB hub. I’ve changed to a different adapter and this works perfect, now.

For what reason ever, this USB-to-RS232 adapter cable is working (https://www.amazon.com/dp/B000A1KD1K)!
For what reason ever, this USB-to-RS232 adapter cable is NOT working (https://www.ebay.com/itm/255666926991)!

The working USB-to-RS232 cable uses a pl2303 chipset and the cable which fails uses a ch341 chipset. But that's not the explanation because the USB-to-RS485 cable for battery communications uses also a ch341 chipset (with the same driver in the SA Pi) and this adapter is working without issues. So it seems it has to do with the additional electronic in the failed USB-to-RS232 adapter... but I'm not sure. The kernel log does not really helps here.

The working adapter shows (e.g.) the following information in the kernel log of the SA Pi (for each adapter):
  1. new full-speed USB device number 16 using xhci_hcd
  2. New USB device found, idVendor=067b, idProduct=23a3, bcdDevice= 6.05
  3. New USB device strings: Mfr=1, Product=2, SerialNumber=3
  4. Product: USB-Serial Controller
  5. Manufacturer: Prolific Technology Inc.
  6. SerialNumber: xxxxxxxxxxx
  7. pl2303 converter detected
  8. pl2303 converter now attached to ttyUSBx
The failing adapter shows (e.g.) the following information in the kernel log of the SA Pi (for each adapter) - as I already mentioned, the USB-to-RS485 adapter is showing the same log, but it's working - so the reason is not the chipset and also not the missing "break control" which is mentioned in the log:
  1. new full-speed USB device number 20 using xhci_hcd
  2. New USB device found, idVendor=1a86, idProduct=7523, bcdDevice= 2.54
  3. New USB device strings: Mfr=0, Product=2, SerialNumber=0
  4. Product: USB2.0-Ser!
  5. ch341-uart converter detected
  6. ch341-uart ttyUSB1: break control not supported, using simulated break
  7. ch341-uart converter now attached to ttyUSB1

The passive RS485 hub's I'm using are from growatt (e.g. SignatureSolar offer these, https://signaturesolar.com/growatt-rs485-can-hub-with-8-ports/).
I'm using this external powered USB hub with 10+1 ports (https://www.amazon.com/dp/B0BH252R3J)

Open issues:
  • The inverters serial number are not showing reliably (sometimes I see only 2 or only 4 and they will never displayed if not filled after first connect).
  • In my situation this is really annoying because I have no chance to identify which of the inverters #1 to #6 does correspond to which physical unit (this is discussed in a different thread in detail)
In this picture the individual inverter temperatures and BUS voltages are shown
View attachment 146059

In the following picture the "missing serial number problem" is shown - in this example only the first 2 inverters SN are displayed:

View attachment 146060

As an example, the following grafana screenshot shows the temperatures of each inverter (I've activated the "show dot" just to be sure that I receive reliable and updating data) represents the temperature values of all 6 inverters in idle (no load, no PV, no AC-in) together with the ambient temperature (I receive this data via SA MQTT bridge in my OpenHAB for further processing and visualization).
View attachment 146061

Off topic:
I need the reliable temperature curves because I'm still in the phase of modding the really loud fans in the EG4-6500EX inverters to make them much more silent via temperature controlled PWM fan control - but this is a different story and I will post the results when it's done. Without reliable inverter temperature I will not be able to proof the modding and do a comparison with un-modded inverters.
Click to expand...
This is FANTASTIC info, Id just like to say thanks for sharing.! I had come to a similar conclusion with SA and 4 EG4 LL batteries, albeit 1 Inverter. I too am EAGER to learn how to Quiet the Jet Engine sound of the EG4 6500 inverter. (I have a Live Music Studio live room next door with my Mic'd Drums).... UGH! ..Very eager to hear your findings from the MOD you are working on.! cheers.

RandomPlanet
 
fmeili1 said:
...

Open issues:
  • The inverters serial number are not showing reliably (sometimes I see only 2 or only 4 and they will never displayed if not filled after first connect).
  • In my situation this is really annoying because I have no chance to identify which of the inverters #1 to #6 does correspond to which physical unit (this is discussed in a different thread in detail)
...
Click to expand...
Just an update about my open issues of my first post. When I turn on the six inverters from left to right, one by one, with a delay of not more than 1-3 seconds between pressing their on/off buttons, I always get all six serial numbers from the inverters reliable.

But if I wait between 1-2 minutes between switching on each inverter (for sure I have to switch on at least one inverter of L2 in not more than 30 seconds after the master - otherwise the master L1 inverter automatically shutting down), the last ones will never show their serial numbers in SA. I've played around with "slow" startup procedure to easier find out which SA inverter number will be assigned to which physical inverter... but this results in missing serial numbers.

Now, I switch on all inverters one by one "fast" and because I see the serial numbers now in a reliable way I've labeled the serial numbers to my inverters so I'm able to identify them and can find out which SA inverter number is assigned to which inverter
 
fmeili1 said:
I want to let you know, that I also found a solution about the usually ugly cable wiring for parallel inverter communication and current sharing between the inverters which was even worse after connecting 6 inverters individually to SA. I found a low profile, open slot, cable channel (https://www.amazon.com/dp/B07C9ZWH1H) which just fits (with a bit of pressure) between the vertical short EMT conduits, which runs between the cable channel and the inverters, and the wall. They have a height of just 0.8 inches and have just enough width to wire 6 inverters with parallel communication, current sharing and individual SA communication via RS232 cables. Now the system looks much cleaner than before.

View attachment 146991
Click to expand...
What are you powering that you need 6 inverters ? I have 2 inverters and most of the time they are less than 25% loaded. They run everything just fine, AC, well pump everything. fyi my power use is about 1000kwh / month.
 
EPicTony said:
What are you powering that you need 6 inverters ? I have 2 inverters and most of the time they are less than 25% loaded. They run everything just fine, AC, well pump everything. fyi my power use is about 1000kwh / month.
Click to expand...
I live in a house with 200A service. Everything is running via electric (no natural/propane gas). I've measured the house consumption since one year and my maximum momentary power consumption was 21kW (AC, cooktop, dryer, oven, water-heater, pool-pump, etc. at the same time). I plan to install 2 mini splits in the garage to keep the inverters and batteries cool, also I connect sometimes an RV to the house power. So I want to have enough reserve to be able to run everything in parallel without thinking about it (and be able to use additional tools, if required) - even if 4 units would mostly more than enough. Two are like "spare" ones and no one knows, if they may be available in a couple of years, so I may use two of them later for parts... I also think, that the inverters will last longer if they would not be running on it's specification edge.
 
fmeili1 said:
I live in a house with 200A service. Everything is running via electric (no natural/propane gas). I've measured the house consumption since one year and my maximum momentary power consumption was 21kW (AC, cooktop, dryer, oven, water-heater, pool-pump, etc. at the same time). I plan to install 2 mini splits in the garage to keep the inverters and batteries cool, also I connect sometimes an RV to the house power. So I want to have enough reserve to be able to run everything in parallel without thinking about it (and be able to use additional tools, if required) - even if 4 units would mostly more than enough. Two are like "spare" ones and no one knows, if they may be available in a couple of years, so I may use two of them later for parts... I also think, that the inverters will last longer if they would not be running on it's specification edge.
Click to expand...
You might want to shut 4 of them off during the night. This is because there will be 0 PV power and the EG4 inverter draws about 100w.
6 EG4's will be a 600+ watt draw off the batteries. 2 EG4's will only draw 200 watts. Probably not an issue in the high sun months but definitely an issue in the winter with short days and lots of clouds. I have a system similar to yours except only 4 EG4 inverters but I have 12 batteries where you have 9. I thought that 2 EG4 inverters would not be enough, boy was I wrong. So I now have inverter #3 that I use only to help charge the batteries faster and provide more PV power on low sun days. Inverter #4 is strictly a spare. By the way what do you have for solar panels ? You will need 12 strings for 6 inverters. I have 6 full strings. You can see what I have in my signature.
 
Last edited:
You must log in or register to reply here.

Similar threads

S
SolarAssistant + Sol-Ark + EG4 LP4
Replies
2
Views
339
MaikaiLifeDIY
MaikaiLifeDIY
O
Solved! EG4 Lifepower4, 6000xp, and solar assistant
Replies
3
Views
266
orion7144
O
JWLV
EG4 6500EX F90 Error
Replies
10
Views
476
traffman2370
T
J
EG4 6000XP and SolarAssistant
2 3 4
Replies
65
Views
2K
Ambrotos
A
nkhorman
EG4 6500EX drop outs
2 3
Replies
48
Views
2K
BajaOffGrid
B
Back
Top