Megarevo RxLNA Discussion

Here is something interesting that I have not seen before the upgrade. Been raining most all day so I have been sipping off the battery.
Think I was down to 27% SOC. Stopped raining and sun is in and out. Pulling in over 4kw and charging with 115 amps. Not sure what is going on but started to do this after I changed the charge current from 70 amps to 115.

Looks like something is limiting your PV production, could be the battery hitting its charge limit and causing the BMS to slowing charge or shutting down, which would hinder solar production. This is only a guess. It can also be solar edging from the clouds.

MNRosado said:

Looks like something is limiting your PV production, could be the battery hitting its charge limit and causing the BMS to slowing charge or shutting down, which would hinder solar production. This is only a guess. It can also be solar edging from the clouds.

Click to expand...

I do seem to have an issue with one of the arrays, but this and many other issues have surfaced with the new software. The inverter has been chugging along for a month without much issue. But I have had nothing but trouble for the last few days since the update. At this time I would say I do not recommend updating, matter of fact I would go back to the older firmware if possible. Before it was barley acceptable, but now a piece of junk. I am going to flash again to see if there is a corruption issue. Only positive I have noticed is the data has been stable sending to Solarman. This evening that has frozen up and the unit is in and out of inverting. Had an issue of "bus over volume" Not sure if that has been resolved by changing the maximum voltage because without rebooting I see no data or warnings. Rebooting is a nightmare to get it to recognize the battery, pv, and or grid.

The "bus over voltage" error is also one I experienced. I think it has something to do with the mix of PV power & battery when the load is right at the edge of PV production, or an issue with the PV string. At least that's when I experienced it.

It was early in the morning, so load was running off battery. As sun came up, PV production was right about at load consumption when the error occurred. It could be coincidence, not sure. It hasn't happened since that one time. My panel string is much below the 500V MPPT maximum so I know it isn't from that.

NCSolarGarage said:

The "bus over voltage" error is also one I experienced. I think it has something to do with the mix of PV power & battery when the load is right at the edge of PV production, or an issue with the PV string. At least that's when I experienced it.

It was early in the morning, so load was running off battery. As sun came up, PV production was right about at load consumption when the error occurred. It could be coincidence, not sure. It hasn't happened since that one time. My panel string is much below the 500V MPPT maximum so I know it isn't from that.

Click to expand...

I think I narrowed it down to a problem with array or it could even be a tracker. I shut off the suspected array yesterday morning and it ran all day without issue. Can conclude "bus over Vol" is on the DC side. I am running at 300 volts.

The array in question the other morning was at 85v while the other identical array was 275v. But later in the morning the voltage came up and it actually producing 100+ watts more then other array. Running 4s2p, and when I disconnected one set of series the voltage seemed to jump to 275v. Been putting up hay so not much time to investigate further.

So with that said I may have to revise my opinion on the new software. Maybe there is something new that makes the system more sensitive to bus voltages?

With the suspect array turned off I am still getting these quick shut downs of the PV.





The 1 minute spike of the EPS is the basement lights. I went down and disabled the "Leak Detection". It might eliminate the reboot each morning for 5 minutes as shared in another thread.

Well, I have to file this firmware upgrade as a frustrating failure. I got the versions to ARM v1.03.11 and DSP to v1.04.09 but then on restart, the unit wouldn't reconnect to the Grid - says that it's "grid low freq." So, I went from a functioning unit of PV, Battery, and Grid that was doing all the right things (except sending data properly to the Solarman site) to one that's not so useful.

I saw that @jsmetalcraft mentioned he'd had to reboot to get out of EPS mode, however, I was able to get my unit into a bricked state where the DSP firmware displayed as v0.00.00. Had to re-upload the firmware to fix it.

Anybody else done an upgrade (besides jsmetalcraft) and had it be stable?

I am upgraded and running stable. However, I am not currently using grid input, just PV and a DIY LiFePO4 battery.

I decided to poke around the firmware code and see what I could find (yes, all the typos are in there). Feeling a little weirded out that it's built on somebody's desktop using old compiler versions. The developer's nickname is xuejd and if you search on LinkedIn, you'll find him there - but as a general manager: https://www.linkedin.com/in/jiliang-wu-1707b7b0/

What I am finding is that the hardware is actually really quite good (minerva's experience notwithstanding), if you take apart the Megarevo everything is well designed and using solid technology, including the controller (seems to be a Texas Instruments platform - but at least 5-10 years old).

There are two technology subsystems at play here (the DSP platform and the ARM platform - which you see in the firmware versions). The DSP platform is Texas Instruments TMS320C2000 (or STMicroelectronics clone) (https://www.ti.com/sc/docs/products/dsp/c2000/24feat.htm) and seems to be what is used to monitor power and frequency and all the lower-level functions of the inverter. It also provides the firmware flash updating capability.

The other subsystem is the interface and display controller and seems to be based on the TI Concerto (Cortex-M3 platform) platform and this is using FreeRTOS (https://www.freertos.org/index.html) which has been around a very long time and is very stable.

The actual ARM subsystem code written by Megarevo for controlling the inverter only seems to be two C program files (once you filter out all the FreeRTOS includes). Having been in software for far too long, it feels like the embedded systems folks that built the inverter kinda whipped up the customer interface on a Windows desktop and update it on an as needed basis.

At the end of the day, the shit still needs to work and Megarevo really does need to step up their customer support.

I've tried rebooting a bunch of times, but I still can't get it to fire up properly on the grid. Still get an 08: grid low freq error after the upgrade. I've turned off PV and Battery.
When I navigate through the status menus, I see:

GRID: 60.00Hz
L1: 125.9V. 0.1A
L2: 125.8V 0.1A

The freq does vary from 59.99 to 60.03Hz over the course of a few minutes. I tried bumping up the FAC-MAX from 60.0 to 60.1 but no dice. FAC-MIN is 42.0 Hz (default).

Yet, the unit registers taking some power from the grid.

POWER
GRID L1: -7W
GRID L2: -7W

Note: somebody earlier asked if there is an easy way to reboot the system. In the 'stream' package for updating the firmware, if you click the 'reboot' button, it'll reboot right away.

Well disabling the "Leak Detection" did not stop the 5 minute test the Inverter goes through every morning.

Alright, some progress. I decided to navigate through the menu tree and see what I could flip to try and clear the "08: Grid Low Freq" error. There is a setting: Setup > Grid Setup and you can change the locale for the the grid. I flipped it from US_CA to CUSTOM and I heard a click and the EPS and Alert lights went out. I then flipped it back to US_CA and the lights stayed off.

Going to watch the data over the next day or so and see if all behaves.

flygeek said:

Alright, some progress. I decided to navigate through the menu tree and see what I could flip to try and clear the "08: Grid Low Freq" error. There is a setting: Setup > Grid Setup and you can change the locale for the the grid. I flipped it from US_CA to CUSTOM and I heard a click and the EPS and Alert lights went out. I then flipped it back to US_CA and the lights stayed off.

Going to watch the data over the next day or so and see if all behaves.

Click to expand...

Progress, and even with (clicking) sounds! I like that! ;-)

flygeek said:

Alright, some progress. I decided to navigate through the menu tree and see what I could flip to try and clear the "08: Grid Low Freq" error. There is a setting: Setup > Grid Setup and you can change the locale for the the grid. I flipped it from US_CA to CUSTOM and I heard a click and the EPS and Alert lights went out. I then flipped it back to US_CA and the lights stayed off.

Going to watch the data over the next day or so and see if all behaves.

Click to expand...

So beings your a bit more educated in these and electronics, what is your opinion on the processing part of this inverter? More then Up to the task, will do the task, or it has to work hard to keep up? Just seems at times it struggles with tasks. Maybe the way the software is written.

Sounds like you need a job at Megarevo or do like China does and reverse engineer and write your own. LOL

flygeek said:

Alright, some progress. I decided to navigate through the menu tree and see what I could flip to try and clear the "08: Grid Low Freq" error. There is a setting: Setup > Grid Setup and you can change the locale for the the grid. I flipped it from US_CA to CUSTOM and I heard a click and the EPS and Alert lights went out. I then flipped it back to US_CA and the lights stayed off.

Going to watch the data over the next day or so and see if all behaves.

Click to expand...

You, know Flygeek something similar happened to me but with the PV not charging. PV would not begin charging my LiFePO4, unit was in Lead acid mode since, at the time there was no Canbus on my BMS.. Finally i switch from lead acid to Lithium waited a few seconds and switch back to lead acid and all of a sudden it started charging with PV. Been working fine ever since. I guest the software is bugging and has difficulties releasing the last command.

Ok, Reecies15 it's 6am i turned off the isolation test last night and since the grid went down at my home I turned off the grid breaker, so when the grid returned i will still be in EPS off grid mode. So far no shutdown, and PV is charging. Looks like this maybe a winner! Will monitor for a few days.

I restarted the inverter last night (first time since settings change) and I didn't get watchdog or "signal restored" alerts from Blue Iris for my shop mounted cameras this morning. I'm not saying they didn't turn off, but I'll have to pay more attention tonight to see if the test runs.

NCSolarGarage said:

I restarted the inverter last night (first time since settings change) and I didn't get watchdog or "signal restored" alerts from Blue Iris for my shop mounted cameras this morning. I'm not saying they didn't turn off, but I'll have to pay more attention tonight to see if the test runs.

Click to expand...

I am on my second day off grid with PV isolation test disabled, so far so good. No resetting of DVR, cameras, internet, clocks and out of wack smarthome connected device.

jsmetalcraft said:

So beings your a bit more educated in these and electronics, what is your opinion on the processing part of this inverter? More then Up to the task, will do the task, or it has to work hard to keep up? Just seems at times it struggles with tasks. Maybe the way the software is written.

Sounds like you need a job at Megarevo or do like China does and reverse engineer and write your own. LOL

Click to expand...

I'm slowly figuring out that the user interface part of the Megarevo is a pretty standard STM32 controller that you'll find in a lot of embedded devices, which jives with wanting to keep manufacturing costs low and standard. It definitely has enough processing power to drive the unit. Think of it like an Arduino device.

It does emit info and error messages on a console, but I haven't checking into how to read those - the 'stream' firmware updater might have the capability (with the ARM and DSP buttons - they launch a different, but blank screen).

jsmetalcraft said:

No, I mean that the BMS and inverter say that they each support both communication protocols. I don't know if either work. Guess we will find out someday.

Click to expand...

seplos BMS doen't support both CAN & RS485 only one of them depending of which version you order. don't try to upload RS485 firmware on a CAN version (or the other way) you going to brick the BMS.

flygeek said:

I'm slowly figuring out that the user interface part of the Megarevo is a pretty standard STM32 controller that you'll find in a lot of embedded devices, which jives with wanting to keep manufacturing costs low and standard. It definitely has enough processing power to drive the unit. Think of it like an Arduino device.

It does emit info and error messages on a console, but I haven't checking into how to read those - the 'stream' firmware updater might have the capability (with the ARM and DSP buttons - they launch a different, but blank screen).

Click to expand...

So the (in my case) USB C port is the serial port hookup ? Just a USB <-> USB cable, config the serial port on your OS and you should be able (famous last words) be able to read (some) console messages?

Well it ran very well off grid for 5 fully days with the PV isolation test disabled. I now turned on the Grid and will run for a few days on grid with the PV isolation test disabled. Just to make sure nothing weird happens. When i turned on the grid breaker the switch over was perfect like the Deye or sol-ark. ?

Another member asked about the boot sequence earlier. I was doing some testing today as I get Bus over vol and Bat over vol warnings. Things been running smooth for the last day. Earlier some posted that they run on lead acid setting. When I tried I had issues, so went back to lithium. I have Sepros BMS communicating via can bus to the inverter.

I made a change in lead acid to see if it, or the inverter was using or looking at both settings. Lithium has no settings. When I entered out of setting things went hay wire even though I changed back to Lithium. Setting made no difference. But as I measured the battery bus on the inverter it was like it was searching for the battery voltage. Problem was it would go from 59v to the actual battery voltage slow and then jump up to 59v again. On and on. I would get a bat over vol error and the BMS would show a warning.

Anyway I had to end up rebooting. Only way I have found so far, and the reboot sequence I follow is: Battery first. Wait for EPS mode and discharge on the battery. Then turn on PV until you see drawing watts from PV. Then turn on Grid. Your Mileage May Vary, or is there a boot sequence that others have found that work for them?

MNRosado said:

Well it ran very well off grid for 5 fully days with the PV isolation test disabled. I now turned on the Grid and will run for a few days on grid with the PV isolation test disabled. Just to make sure nothing weird happens. When i turned on the grid breaker the switch over was perfect like the Deye or sol-ark. ?

Click to expand...

Did 4 days on grid with PV isolation test disabled, and again ran very well. I think i now have a use-able inverter