Hi everybody,

I'm new to this forum and found a lot of knowledge here, so I hope to find some help.

I'm from Germany and I'm running since 4 years Easun 5K/48V Inverters in 3Phase parallel mode und approx. 20+ KW 48V batteries, self-made with Daly BMS.

For monitoring I'm using Solar Assistant on a Raspberry PI connected via USB to the inverters.

A couple of weeks ago, 1 inverter smoked up and it looked like repair might be too difficult for me.

So, I quickly bought 3 new inverters from Easun ISolar SMP 5KW, as I couldn't get a replacement (same type/firmware for parallel mode) for my defect one.

I thought, I simply replace the inverters and continue operation...

Unfortunately, one inverter came up with Error 81 (No Host), indicating some issue with the parallel connection.

I checked all cable, exchanged the parallel board between the inverters, exchanged all cables, but the problem remained always with the same inverter, so it must be defect.

So, I've ordered a 4th one and contacted the Easun support  that I want to replace the defect inverter.

No response from support, but after a few days I got at least my new inverter, replaced it and happily it was working.

However, connecting my Raspberry PI via (now) micro USB cables for monitoring, it startet working for 2-3 hours and then USB connection was failing, first with one inverter, then some time later with all inverters.

The new inverters have a button on top left, which can be long pressed for 3 seconds to start a firmware update from USB drive.

Clicking this to enter firmware update mode and then go back, seamed to reset the USB connection and then monitoring was working again for a few hour until connection started to fail again.

Ok, I thought if USB connection makes trouble I ordered USB /Serial Cable with PL2303 chipset in order to use those for monitoring. Also, seamed to be working fine for 8-10 hours, then the display of one inverter turned dark and was not longer responsive for button controls also serial connection was failing. The inverter was still operating ok.

This can be recovered only by switching off the inverter and restart again.

I also figured that removing the cable from USB port while connection is established, make the inverter turn dark immediately.

I need monitoring and control as I need to charge batteries from utility when electricity price is low in winter times, which I'm controlling through home assistant.

Easun Support is not helpful at all. They also don't want to replace the defect inverter, but only send a new internal connector cable.

I've reminded them on the 30 days money back guarantee, but then the support didn't answer any more.

My questions:

1. Is there somebody having the same problems

2. Is there any firmware update fixing those issues (there are also more issues when setting the parameters)

3. Does anybody know which firmware to use for this models

4. Is there maybe any better / recommended cable (micro USB or Serial) which are connecting stable

Any help is highly appreciated.

 

Thanks,

Michael

 

 

 

 

 

 

 

 

Welcome to the forum.

On 2022/11/03 at 12:22 AM, mknecht said:

3. Does anybody know which firmware to use for this models

Sorry, I'm more and more confused about EASun's relationship with Voltronic, and why their version numbers are so different to other models' firmware version numbers. None of the firmware versions that you quote above make any sense to me.

I'm sorry to hear about your problems, and how EASun are being so unhelpful.

18 hours ago, Coulomb said:

Welcome to the forum.

Sorry, I'm more and more confused about EASun's relationship with Voltronic, and why their version numbers are so different to other models' firmware version numbers. None of the firmware versions that you quote above make any sense to me.

I'm sorry to hear about your problems, and how EASun are being so unhelpful.

The day before Yesterday, I've replaced the display of the inverter which turned dark with the display of my spare inverter.

So, at least the serial connection is running stable since then.

For me it looks like the Axpert MKS III Off-Grid Inverter, but mine has WIFI build in and also an ON-Grid feature which I'm not using.

I would use my spare inverter for testing a firmware update, when I can get at least a firmware version which is most likely to be working.

I also have a few more strange effects with the current version:

- In parallel mode, max grid charge power can only be set on inverter for phase 1, the others allow only 2A to be setup. Actually I'd rather 30A - 40A per Inverter for charging my batteries instead of 60A-80A on only 1 inverter.

- Output voltage can only be set to 220V in battery mode, while Utility is running 238V - 240V. I don't know why it's not possible to set it 240V to run it grid synchron and make the switch between battery and utility more smooth.

 

3 hours ago, mknecht said:

For me it looks like the Axpert MKS III Off-Grid Inverter, but mine has WIFI build in and also an ON-Grid feature which I'm not using.

When you replace the removable display, you get the settings for the model that the display firmware thinks it's talking to. It may be for example that you have a removable display (or at least removable display firmware) for an Infini (on-grid, grid-tie, capable of export) model.

There are several models that look like the Axpert MKS III, e.g. Axpert VM III, Axpert King. The VM III has the display to one side, but metalwork for EASun models may be different than that for other resellers' models.

3 hours ago, mknecht said:

- In parallel mode, max grid charge power can only be set on inverter for phase 1, the others allow only 2A to be setup.

I have no idea what's going on there.

3 hours ago, mknecht said:

- Output voltage can only be set to 220V in battery mode,

With Axperts (Infinis may be different, but if so I'd be very surprised), you have to have the power switch off (hence no power to the AC-out port)  to change the output voltage. If you have a charging source (AC-in or PV), then the inverter won't switch off 30 seconds after turning the power switch off. Without a charging source, you have to be quick and finish the setting change before the inverter switches off completely. If the power switch is on, the inverter will simply NAK the command to change the output voltage, and the output voltage setting won't change. Could that be why it appears to be stuck at 220 V?

16 hours ago, Coulomb said:

When you replace the removable display, you get the settings for the model that the display firmware thinks it's talking to. It may be for example that you have a removable display (or at least removable display firmware) for an Infini (on-grid, grid-tie, capable of export) model.

I'll have a look at the infinity series to compare the specs. What is the official web-site for Axpert inverters? I found voltronicpower.com but this site  doesn't work (showing a compiler error) and I found voltronic inverters on Taiwantrade.com

16 hours ago, Coulomb said:

There are several models that look like the Axpert MKS III, e.g. Axpert VM III, Axpert King. The VM III has the display to one side, but metalwork for EASun models may be different than that for other resellers' models.

I have no idea what's going on there.

Same me. It was always working fine with the old inverters.

16 hours ago, Coulomb said:

With Axperts (Infinis may be different, but if so I'd be very surprised), you have to have the power switch off (hence no power to the AC-out port)  to change the output voltage. If you have a charging source (AC-in or PV), then the inverter won't switch off 30 seconds after turning the power switch off. Without a charging source, you have to be quick and finish the setting change before the inverter switches off completely. If the power switch is on, the inverter will simply NAK the command to change the output voltage, and the output voltage setting won't change. Could that be why it appears to be stuck at 220 V?

It's the same with my inverters, when I disconnect grid and turn power switch off, they power off after 30 sec if battery is still attached.

I've tried to change settings with power switch off, but grid still connected, which didn't work. I will try as you recommend.

Is there any download section for the firmware or do I have to search the forum for some download links?

 

8 hours ago, mknecht said:

What is the official web-site for Axpert inverters? I found voltronicpower.com

The light works fine in my office. (That is, voltronicpower.com works for me, I checked again just now).

8 hours ago, mknecht said:

Is there any download section for the firmware

It's not specifically for the firmware, but the files section of this forum is pretty good (powerforum.com/files).

There are also the firmware sections of the AEVA topics, e.g. https://forums.aeva.asn.au/viewtopic.php?f=64&t=4332

15 hours ago, Coulomb said:

The light works fine in my office. (That is, voltronicpower.com works for me, I checked again just now).

I've checked voltronicpower.com. Working now.

From the specs my inverters should be the same as the InfiniSolar V III 5KW, it's with WiFi, on grid option and 450 VDC

15 hours ago, Coulomb said:

It's not specifically for the firmware, but the files section of this forum is pretty good (powerforum.com/files).

There are also the firmware sections of the AEVA topics, e.g. https://forums.aeva.asn.au/viewtopic.php?f=64&t=4332

Great. I'll have a look.

Thank you for your help

43 minutes ago, mknecht said:

From the specs my inverters should be the same as the InfiniSolar V III 5KW,

Ah. There are far fewer firmware update files and even information of any kind for the InfiniSolar models.

On 2022/11/06 at 7:56 AM, Coulomb said:

Ah. There are far fewer firmware update files and even information of any kind for the InfiniSolar models.

Hey Coulomb,

you are a hero..., I've got so many problems and it looked like they are all related to this removable displays.

I found your post on the adapter cable to bypass the display and I've decided to try it.

So, in the morning I've build 3 adapter cables (and yes TX and RX have to be exchanged) and connected them to my solar assistant Raspberry PI.

Now I can make all settings as I'm used to, like setting each inverter individually to my preferred max grid charge current.

Also, the inverter values are now read correctly, as for example the charge source priority and output mode was showing wrong values when connecting via the displays.

With displays attached, I had the following problems:

- Unstable connection via USB/Micro USB cable and USB to Serial cable

- 1 of the USB to Serial cables started to get seriously hot so that I could even deform the plastics of the serial plug

- Unplugging the USB Serial cable from USB side, immediately shutdown the display

- Sporadic shutdown of the display during operations, connection became unresponsive then

- Problems to read / set inverter certain values, which didn't allow to operate the inverters as needed. 

- Sudden and unnecessary change from Battery Mode to Utility, typically every full hour. It appears to me that these timer buttons forcing some activity even though no timer is set.

So, it looks like it's all resolved. Without the adapters I would have to buy new inverters from somewhere else, as EASun is not responding to any of my requests.

Their 30 days money back guarantee is simply lie.

Hello. Where can I find information on how to make this cable. I couldn't find anything myself.

 

Quote

- Sudden and unnecessary change from Battery Mode to Utility, typically every full hour. It appears to me that these timer buttons forcing some activity even though no timer is set.

same problem with me

Edited January 15, 20233 yr by schpuntik

I have given up on EASun and bought 3 Axpert King 2 5kw.

EASun is a bunch of fraudsters

Hi Michael,

sorry I am too late, but the solution is rather easy: The SMP5K inverter displays do not all have the same potential on PG (Protective ground) due to them being covered in paint and not the same SG (Signal ground) at all - you need galvanic separation if you want to connect to more than one inverter at the same time (You might guess, how I learned that...)

However once you have this in place, they work fine and reliably.

Hi LeoP1970

Helpful information,

but can you describe exactly what you made ?

I talk to the Inverters via an ESP32, but I have verified, that the same problem exists when talking via PC/rasPi RS232 ports (oldschool or USB-Serial): They are basically all a 3-wire affair with SGND/RX/TX. The important part is, that if you have more than 1 serial port coming from the same PC/rasPi/Arduino, it is silently assumed, that all SGND have the same potential. This is not true with 3 inverters in 3-phase mode.

In essence, your serial connection pulls all 3 SGNDs together, which the inverters don't appreciate. They show that by being flaky in very subtle ways. After seeing this, I checked with a simple multimeter and found delta currents in the 10mA range.

You can fix that in different ways:

- Simplest method: Buy a Optocoupler-based RS232 insulator for each. Didn't try that in this case, but I have had good experience with those in other projects

- My method: The ESP32 or Arduino comes with TTL-Level (3.3 or 5V, doesn't matter) serials. You can get those from a PC/RasPi easily via a TTL-Level Serial to USB adapter. Now

a.. Use a Traco TEA 1-0505 to create a 5V/GND pair, that is galvanically separated from your controlling source, use it to power the inverter side of:

b.. Use an ADUM 1201 AR to galvanically separate your inverter-side TX/RX from your controller-side RX/TX - conveniently, this also handles 3.3/5V conversion, just make sure the controller-side VCC on the ADUM is correct

c.. Power a MAX3232 from the VCC/GND we created in a and use it to convert the TTL levels to RS232 levels.

If you don't like to solder SMD: Breakout boards for b and c are easily available via Amazon, the Tracos are throughwire.

Just to make it clear: You need to use a seperate Traco for each RS232 channel, because else you will insulate them from your controller, but nor from each other.

 

I have thrown together an ESP32 controller, that does all the magic with selecting a different inverter RS232, controls the on/off switch (hint: This is essential for summer homes that are uninhabited for longer periods of time in bad weather season as known here in Europe, since the self consumption of the inverters in bad weather will drain the batteries) and additionally can control a DC string switchover module to switch strings to a seperate grid-tie inverter, as the EAsuns don't have the necessary certifications to feed into the grid here in Europe. This also takes care of the yo-yo-charging problem, that plagues 3-phase plants when PV strings are attached to more than one inverter: They will regulate into a very unhealthy resonance if batteries are full, overshooting VDCmax by a lot, then dropping far too low, ...

Once this is all battle tested, I will open-source it. No SMD soldering, only TW and nothing less than 1/10", so good to go for most hobbyists.

 

14 hours ago, LeoP1970 said:

In essence, your serial connection pulls all 3 SGNDs together, which the inverters don't appreciate. They show that by being flaky in very subtle ways. After seeing this, I checked with a simple multimeter and found delta currents in the 10mA range.

Interesting.

The older models are definitely opto-isolated already, and I assume that all the modern models still are. Maybe it's an issue with how you connect the grounds to carry away the pulses of current from the EMI suppression capacitors.

That sort of thing can certainly be hard to diagnose and fix.

The ones with the removable Display obviously aren't.

I didn't have the same sort of problems with the older models, but I didn't use them in 3-phase (which I experienced as not very reliable), just in parallel mode.

These newer ones (Model SMP-5K) work fine as 3-phase, but they do have this ground-potential issue. I have seen 3-phase sync fail alltogether with all 3 being connected via Serial, immediately working again when I pull one serial off.

**EDIT**

Just to make that clear: The TX/RX work fine, which might indicate optocoupling. But the moment you pull the SGND of the 3 inverters on the same potential, you get a problem.

Edited April 11, 20233 yr by LeoP1970
Clarification

2 hours ago, LeoP1970 said:

The ones with the removable Display obviously aren't.

I assume you mean aren't isolated.

I see four opto isolators right near the BMS and PC RJ-45 connectors (outlines in orange, ignore other highlights):

2 hours ago, LeoP1970 said:

I have seen 3-phase sync fail all together with all 3 being connected via Serial, immediately working again when I pull one serial off.

Interesting.

Hi, the troubles with the serial connection is solved with opto isolated rs232 converter.

I have used this one. Diyeeni RS232 auf RS232 Seriell Ports Optischer Isolator DB9 RS232 Isolierter Schnittstellenadapter: Amazon.de: Computer & Zubehör

But I have this converters in 3 phase connection, but sometimes I have a complete blackout in my house for 2-3 seconds (Maybe 1 fault in 3 days)

My inverters are in sbu-mode. No troubles in switching to grid-mode after batteries are empty and also no troubles  by switching back to sbu-mode if batteries are full.

I have now activated (enabled) parameter 25 (record fault code).

Did anyone read this data ? I must organice one micro usb adapter to put in an usb-stick. 

Did anyone have the the same troubles or maybe a solution ?

Regards from Austria Dieter

 

Edited April 12, 20233 yr by toeffi

On 2023/04/11 at 5:35 PM, Coulomb said:

I see four opto isolators right near the BMS and PC RJ-45 connectors (outlines in orange, ignore other highlights):

 

Yes, the RX and TX lines may be opto-isolated. The point is, that SGND, which is also the GND for the display unit, seems to take badly to being pulled together from different display units. I don't know why, but I see a lot of evidence for that.

On 2023/04/11 at 6:10 PM, toeffi said:

Hi, the troubles with the serial connection is solved with opto isolated rs232 converter.

I have used this one. Diyeeni RS232 auf RS232 Seriell Ports Optischer Isolator DB9 RS232 Isolierter Schnittstellenadapter: Amazon.de: Computer & Zubehör

 

Dieter seems to see the same thing.

On plants, where the display is not needed I just remove it - this gets rid of the problem immediately, the direct serial connection to the main inverter is not susceptible to this sort of problem.

On 2023/04/11 at 6:10 PM, toeffi said:

But I have this converters in 3 phase connection, but sometimes I have a complete blackout in my house for 2-3 seconds (Maybe 1 fault in 3 days)

I have seen this a lot, and I am afraid the underlying cause might be not in your plant, but in the neighbourhood. I searched for this problem for weeks and could not find any clue, when I saw it happen by accident and then could reproduce it:

- You know, that an inductive load phase-shifts AC. A strong electric motor e.g. does so quite noticeably.

- The 3-phase sync mechanism of the inverters if connected to the power grid is very sensitive: If even a slight sync mismatch between the inverter outputs in battery mode (this also includes solar mode) and the grid is detected, will make the inverters immediately disconnect and only reconnect after sync has been reestablished (1-3 seconds)

- Now assume your neighbour has a strong single phase electric motor with badly designed startup behaviour in use (I found this out when I watched the problem happen when my neighbour hat construction work done, which included a 4kW single-phase cement silo pump). In the moment this motor starts up, only the one phase is phase-shifted, enough so, that the inverters assume a loss of sync and disconnect, just to reconnect when everything is back to normal.

There are two ways around this:

- In many cases rotating the phases helps: If your inverters 1, 2, 3 are connected to phases R, S, T, then rewiring them to T, R, S will not make any problems (there is no "first" phase in 3-phase), but might move the phase, that sees phase-shifting from the second position (most sensitive) to the third position (least sensitive)

- A complete solutions is more involved, basically you need to replace soft bypass by an external hard bypass and run the inverters in blackout mode. I will give the details to that in a different thread, as I am still working on the controller  - prototype does work fine though.

To check, if this actually is your problem you can just switch off grid input to the inverters when your batteries are full enough, that SBU never falls through to "U". If the problem then disappears chances are, that this actually is what you are seeing.

BTW: I am from Austria too, seeing this problem mostly on EVN grid nodes, less so on Wienstrom nodes.

All the best,

Leo

 

Edited April 14, 20233 yr by LeoP1970
Fixed bad english grammar (not my first language)

Hallo Leo danke für deine Antwort.

Ich dachte, dass es ein Problem mit dem Gitter gibt.

Ich habe nachts im Grid-Modus ein weiteres interessantes Phänomen.

Ein Wechselrichter versucht, die Batterien zu laden, aber ich habe alle Wechselrichter, um die Batterien nur im Solarmodus zu laden  🤒.

Von wo aus bist du in Österreich?

Ich bin in der Nähe von Klagenfurt 

Gruß Dieter

 

Edited April 14, 20233 yr by toeffi

1 minute ago, toeffi said:

I thought that there is a problem with grid.

Try switching off the grid if you have enough Solar / Battery and see, if the problem goes away to confirm.

2 minutes ago, toeffi said:

I have another interesting phenomena at night on grid mode.

One Inverter tries to load the batteries but I have all inverters to load the batteries only in  solar mode 🤒.

Again I know this problem: Sometimes in 3P mode one of the inverters will not accept OSO charging mode - it will show no errors, but will silently ignore it.

Only known possibility:

- switch off all inverters (button not pressed)

- remove all power inputs - Grid, PV and Battery and wait 30s

- Apply battery and AC power only to the one inverter, keeping it in soft off (button not pressed), then configure it to standalone (not 3phase), wait 30s, then configure  OSO charging, wait 30s, then configure back to 3p, wait 30s

- Power down this inverter again

- Power up all inverters

This has helped with the mentioned problem on2 different plants now.

 

I am in Vienna and the plants I work with are all inside 150km of Vienna

All the best,

Leo

On 2022/11/04 at 10:56 AM, mknecht said:

The day before Yesterday, I've replaced the display of the inverter which turned dark with the display of my spare inverter.

So, at least the serial connection is running stable since then.

For me it looks like the Axpert MKS III Off-Grid Inverter, but mine has WIFI build in and also an ON-Grid feature which I'm not using.

I would use my spare inverter for testing a firmware update, when I can get at least a firmware version which is most likely to be working.

I also have a few more strange effects with the current version:

- In parallel mode, max grid charge power can only be set on inverter for phase 1, the others allow only 2A to be setup. Actually I'd rather 30A - 40A per Inverter for charging my batteries instead of 60A-80A on only 1 inverter.

- Output voltage can only be set to 220V in battery mode, while Utility is running 238V - 240V. I don't know why it's not possible to set it 240V to run it grid synchron and make the switch between battery and utility more smooth.

 

Hi die Einstellungen die bei ihnen sich nicht einstellen lassen sind die die sie nur einstellen können wenn sie den parallel Betrieb rausnehmen. Dann einstellen und wieder zurück stellen in 3phasen Betrieb.

On 2023/04/14 at 7:48 PM, LeoP1970 said:

Yes, the RX and TX lines may be opto-isolated. The point is, that SGND, which is also the GND for the display unit, seems to take badly to being pulled together from different display units. I don't know why, but I see a lot of evidence for that.

Hi. I manage several plants with Easun inverters (mostly SMP5K, SV IV5.6 and SMW11K). Almost every plant is monitored by some ARM (mainly OP3 or NanoPi M1+) boards.

On plants with 2, 3 or 4 inverters, I had many issues : usb link unstable, rs232 adapters burning, or USB hub burnings after few days or weeks. I came to the conclusion that there is a GND issue. Your conclusion looks almost the same. To avoid this issue, on SV IV5.6K, I now connect only 1 inverter, and use PGS command to query the other ones. But with SMP5K, there is no equivalent command, and I sometimes ends with 1 arm board per inverter (and sometimes, it works fine with 1 board for 3 inverters).

But as you advocate to isolate more, I wonder if the inverse option could work : did you try to connect GND of displays together ? I think I'll give a try next time I'm in a plant with unstable links, but as I understand, it should make things goes better. 
 

kinds regards from Corsica,

Greg.