Megarevo 8K Firmware Update

fromport said:

Is there any way to know if the RS485 connection is working ?
Will "any" RS485 adapter work or do you need a specific version to work with the program ?
I think got the wrong one, XDS100 vs XDS110
Can you tell me about the special cable that you were talking about ?

Click to expand...

If the RS-485 connection is working, then during firmware update, the hybrid app will show the progress bar making progress. Pretty much any RS-485 adapter will work.
Here is the super hard way to flash the .out firmware and recover from a failed .out firmware upgrade if all the above suggestions fail. The XDS100 should work the same as the XDS110. You'll need Uniflash to upload the .out file. You need to disconnect all power from the inverter, disconnect both power connectors from the control board. Near that two-pin power connector, there's a relay. Remove the two-pin jumper next to the relay. Power the two-pin AC port with DC 220-350V (needs roughly 20W). It must be DC. AC will be bad, even though it runs on AC during normal operation. Connect your programmer to your PC and to the white 4 pin connector on the control board; it's toward the middle of the board on the edge closest to you. Left to right pins 1-4 are TCK, GND, TMS, 3.3V. See pages 97 & 98 for more necessary info on the connections. Flash the firmware using UniFlash. (The first time you use your programmer, it will probably update the firmware on your programmer.) Then disconnect and put all connections back to how it was.

NCSolarGarage said:

You need to disable "insulation d" -> "Leack Dete". I'm not sure if that is the same as the ARC you are referring to.

Setup > Run Setting > Active Rep > Insulation D > Disable

Click to expand...

Hi,
I updated my MR with the firmware from @fromport website (https://www.dth.net/solar/megarevo/firmware/20231102/) but I am not seeing the path Setup > Run Setting > Active Rep > Insulation D > Disable. My Run Setting submenu stops at FAC-MAX. Is there something I am missing? I am also having the rebooting of the inverter in the morning at about 6 am. What is the setting I need to change on the latest firmware?

chayden said:

I am also having the rebooting of the inverter in the morning at about 6 am. What is the setting I need to change on the latest firmware?

Click to expand...

Disable the isolation test setting.

Does the latest firmware not have that setting in the tree ?
If that is the case, can you try:
SYS SETTING -> 7 ARC ENABLE -> 1. DISABLE

Is anyone running the 0605 release in parallel? I'm having an issue with the slave unit reporting twice its load. The Main unit is correctly reporting its load based on the current its drawing from the battery. On the slave unit, however, it is reporting twice its load. What this does is shifts more of the "work' to the main unit, even though its reporting they are sharing it 50/50.

For example: Master/Slave


And in Home Assistant, same reported data:


Power is reporting 9.6kw when in reality its only about 7.5kw of battery draw.

fromport said:

Disable the isolation test setting.

Does the latest firmware not have that setting in the tree ?
If that is the case, can you try:
SYS SETTING -> 7 ARC ENABLE -> 1. DISABLE

Click to expand...

Hi @fromport . Thanks for the response. I do not that in the setting tree. I have made the adjustment to disable ARC ENABLE. I will watch it over the next two days and see how it performs.

chayden said:

Hi @fromport . Thanks for the response. I do not that in the setting tree. I have made the adjustment to disable ARC ENABLE. I will watch it over the next two days and see how it performs.

Click to expand...

(AFAIK) That "check insulation" is part of the ARC check of the MR. So by disabling that, also the insulation test should be disabled [famous last words]

When off grid, the insulation check turns off the inverter for a couple minutes each day. You can hear some relays clicking. It's measuring the resistance between PV wiring and ground.

The ARC fault detection shouldn't have any relation to the PV insulation check. If you follow the PV wires up from the connector, they go through an arc fault detector. With arc detection enabled, if the inverter detects arcing in the PV lines, it will turn off the solar so that the panels don't catch fire. Then you have to go find where the bad connection is.

chayden said:

Hi @fromport . Thanks for the response. I do not that in the setting tree. I have made the adjustment to disable ARC ENABLE. I will watch it over the next two days and see how it performs.

Click to expand...

Just an update that disabling ARC ENABLE worked and MR is no longer rebooting. Thanks very much.

Well this thread really died off. I was ready to buy a 10k MR, then decided to hold off and see if theey fix the quirks in the software. I have not seen much activity on this site about the Megarevo. Either everyone is having 0 issues, or too disgusted with their units to talk about them.

I have also been following the SRNE 10k inverter threads. Not sure if it has all the same features in its software as far as time of use and priority modes.

Well let me know what you guys are dealing with.
Thank you,

So far it is performing as expected for me. single unit with 24kWh batteries

Just popped in for a random check...

Running the eg4-8k at parents house... a year now and seem good to go... (Glad to see @fromport has his sorted out)


.

Our EG4-8KEXP-240 (MegaRevo R8KLNA) Hybrid Inverter & EG4 LP4 Battery/PV Install​

Mpgrimm2 said:

Just popped in for a random check...

Running the eg4-8k at parents house... a year now and seem good to go... (Glad to see @fromport has his sorted out)


.

Our EG4-8KEXP-240 (MegaRevo R8KLNA) Hybrid Inverter & EG4 LP4 Battery/PV Install​

Click to expand...

I was helped a lot by the feedback/help from @WrenchLight who helped me more than support of MR ever did !
As long as it keeps working, I will keep using it. But I would never advise the unit for a new setup.

This is now the 2nd day that my MR is failing to charge the batteries and instead feeds back into the grid:

Last "normal" day, behaving as expected:
During the night using the batteries, sunrise -> take over the house load and then start charging the batteries, as soon as batteries are full feed in into the grid


This is what I expected and what it has been doing for a very long time, and then yesterday:

Instead of charging the batteries, it started feeding back to the grid first and finally at 11AM it started charging the batteries.


And then finally today:

It fed basically everything it had back into the grid and at 4:15PM it finally decided it needed to charge the batteries.
From 8:20AM the battery State-of-Charge was at 48% till 4:15PM and then it charge in the declining sun to 59%.
The house is now being powered from the batteries but it won't make till next morning without reaching my 30% minimum limit I set.
So part of the night I will be using power from the grid instead of from the batteries.

How can this happen ?
It was so stable for such long time.
Last changes in settings were many _many_ months ago.

Incredible....
Almost feels like an overflow of a register of some kind.
If/when tomorrow it again won't start charging the batteries in the morning after sunrise, I will do a complete shutdown and do a cold restart.
If after that it works again as expected, i assume this is a software bug.

Interesting! The MR8k chatter has been quiet lately. I would have probably sent a reset to it once I noticed that happened.

Off Topic, I'm still trying to buy (or warranty) replacement parts for 1 of my 3, no luck yet. I've contacted MR and many sellers on the Ali sites, can't source a motherboard.

I just did an update of the firmware to https://www.dth.net/solar/megarevo/firmware/20240428/

HybridInverter8KSNA_v1.4.13.B03D36_20231212.out
MD5SUM: bbd1a875e70e87f9d9ee04e292698c8c

hybrid_appLNAV1.03.24B01D31_20240401.axf
MD5SUM: 60e81eee877845a802f40a1a31573c550

Compare that with previous stable version dated November 2nd 2023:
HybridInverter8KSNA_v1.4.13.B03D34_20231012.out
The file name suggests the new file is 1 month newer weeks compared to the old stable file.

hybrid_appLNAV1.03.24B01D15_20230824.axf
The new version is almost 6 months newer then the previous stable.

As I stated before, I had an issue with my unit suddenly not charging from the grid anymore.
Yesterday I first did a powercycle, Battery SoC (state of charge) went from 80% -> 40%.
Got an error of no BMS communication.
After a few hours I upgraded to this latest version.
Still wouldn't communicate.
That is when I power cycled my 4 batteries and finally got communication back online.
Thanks to @alleydog77 for providing the new firmware image.

I need help, I was updating my inverter with the latest version sent by Megarevo. When I connect the cable the computer detects the cable, but when I go to do the update it stays loading and the update is not performed. I changed the option from COM3 to COM4 to see if it works, and when the process begins the inverter turned off, and did not turn on again. When I open the top cover the plates have lights on. Any help, could it be the cable that is not doing its job?

Chrizbr said:

I need help, I was updating my inverter with the latest version sent by Megarevo. When I connect the cable the computer detects the cable, but when I go to do the update it stays loading and the update is not performed. I changed the option from COM3 to COM4 to see if it works, and when the process begins the inverter turned off, and did not turn on again. When I open the top cover the plates have lights on. Any help, could it be the cable that is not doing its job?

Click to expand...

On the bottom side of the control board, there are three LEDs. The one on the right is power. The left one (middle of the board) is for the DSP processor. The middle-right LED is for the ARM processor. Both of these flash to indicate operation. If either is not on or on solid, that processor is locked up. The .out file updates the DSP processor, and the .axf file updates the ARM processor.

Scroll up to my January 9th post in this thread for recovering from a failed firmware update. When you do a power-cycle, you may need to disconnect the battery in addition to solar and grid power. Make sure those LEDs go out before turning it back on.

As for the com port, it's always COM3 for me. The update can take 30 seconds or so before it starts. Total time should be under two minutes or so per file. Also, you may want to try copying the firmware files to your C:\temp folder so that we don't have problems with special characters in the file name.

Hello, I was able to fix and install the latest version. The problem, if I can tell you it is a problem, is that I see that the inverter is sending more current from the utility than what the loads need? Any recommendation?

Could it be because of the Power factor? or simply function of the inverter to improve load consumption?

Chrizbr said:

Hello, I was able to fix and install the latest version. The problem, if I can tell you it is a problem, is that I see that the inverter is sending more current from the utility than what the loads need? Any recommendation?

Could it be because of the Power factor? or simply function of the inverter to improve load consumption?

Click to expand...

Could it be that the setting for zero export power is changed? Sometimes when new features are added to the firmware, the new setting can be kind of crazy. It'll use a new memory location for the setting and use whatever's there instead of a default setting value. I'd recommend going through all the settings and checking if they make sense.

"Zero export power" adjusts the amount of power the inverter generates to target zero sell-back during times when solar isn't enough to cover more than the load demand. I have a question on the graph. Is negative grid power selling back to the grid, and positive grid power buying? I'm pretty sure the MR uses negative for buying and positive for selling grid power. Try verifying the readings with a clamp-on ammeter, although it won't tell you the direction. If the CTs are backwards or not installed correctly, that could be a problem. Arrow toward grid. L1 to L1 port; L2 to L2 port.