13 hours ago, Coulomb said:

I think it should be OK to connect the AC-out neutrals together. You only need to parallel the batteries with Voltronic inverters when you actually parallel the outputs (lives as well as neutrals), or you want a genuine 3-phase configuration.

I'm curious as to why you want to parallel the battery negatives only.

@Coulomb  Sorry, i don't understand. So i can have two independant battery (no direct connection between them) and connect only the neutral together on the AC output ?  Yes i want to have a clean 3-phase connection (the two inverters, and maybe a third in future, create a 3-phase power output)  for powering triphased tools like a air compressor and other tools in my lab. The most of the time will use only the first phase for powering monophased equipement (lights, computers...etc)  but some days all the phases could be used. Also i don't work in my lab in the evening so most of the energy will come from solar panels during the day.

Won't this cause problems, especially with phase balancing and common neutral?

Updated schema :

Edited September 10, 20241 yr by fhocorp
add mention of concerned user

15 hours ago, fhocorp said:

@Coulomb  Sorry, i don't understand. So i can have two independant battery (no direct connection between them) and connect only the neutral together on the AC output ?  Yes i want to have a clean 3-phase connection (the two inverters, and maybe a third in future, create a 3-phase power output)  for powering triphased tools like a air compressor and other tools in my lab. The most of the time will use only the first phase for powering monophased equipement (lights, computers...etc)  but some days all the phases could be used. Also i don't work in my lab in the evening so most of the energy will come from solar panels during the day.

Won't this cause problems, especially with phase balancing and common neutral?

Updated schema :

Fhocorp, with Vevor EML3500-24L you can not do that. These types of inverters can not be synchronized. A three phase system has 120 degree phase shift between voltages and variable phase shift between current phases (depending on inductive or resistive load).

Your schematic above has one neutral and two phases with unknown and variable phase shift between them.  First read about the three phase system and understand it: https://en.wikipedia.org/wiki/Three-phase_electric_power

To build your schematic to be useful (using a load which require two or three phases ) you need an inverter suitable for synchronization.

What you may get with this schematic it's a split power network for two load sections. On EML3500-24L neutral output can be connected to earth ground as well if you do not have mounted any differential electromagnetic fuses in your installation.

7 hours ago, rowesley said:

Fhocorp, with Vevor EML3500-24L you can not do that. These types of inverters can not be synchronized. A three phase system has 120 degree phase shift between voltages and variable phase shift between current phases (depending on inductive or resistive load).

Your schematic above has one neutral and two phases with unknown and variable phase shift between them.  First read about the three phase system and understand it: https://en.wikipedia.org/wiki/Three-phase_electric_power

To build your schematic to be useful (using a load which require two or three phases ) you need an inverter suitable for synchronization.

What you may get with this schematic it's a split power network for two load sections. On EML3500-24L neutral output can be connected to earth ground as well if you do not have mounted any differential electromagnetic fuses in your installation.

Ok sir 😁 I have good knowledge in electricity but not in the triphased syncronisation part 🤐. Thanks for your help. I will keep my installation in place for two load segments and upgrade it for 3 syncronized inverters in the future. Actualy, I have received my new noctua fans for my Vevor inverters, keep you informed about the results 😁

 

On 2024/09/10 at 10:46 PM, fhocorp said:

Won't this cause problems, especially with phase balancing and common neutral?

You confused me with the two inverters. As rowesly mentioned, you need three inverters for real 3-phase loads, and they need special hardware. When you get to that stage, you will have to have effectively one battery (e.g. two 24V banks in parallel) that all 3 inverters run off.

So yes as far as I know you can common neutrals of two inverters with separate batteries, but that does you no good for real 3-phase loads like compressors. 3-phase compressors usually have very high inrush current requirements, so you will probably need 3 48V inverters. 

On 2024/09/12 at 2:42 AM, Coulomb said:

You confused me with the two inverters. As rowesly mentioned, you need three inverters for real 3-phase loads, and they need special hardware. When you get to that stage, you will have to have effectively one battery (e.g. two 24V banks in parallel) that all 3 inverters run off.

So yes as far as I know you can common neutrals of two inverters with separate batteries, but that does you no good for real 3-phase loads like compressors. 3-phase compressors usually have very high inrush current requirements, so you will probably need 3 48V inverters. 

Okay, thanks @Coulomb for your advices 👍

I will keep my installation of my two inverters in independant mode.

The very boring thing is that the auto-start of the the MPPT part of the inverter that is not working. Every day, i have to manualy power on the inverter (via it's power button) for charging the batteries.

For the PV charger to work, the 230V output of the inverter must be switched on, otherwise the device remains completely switched off. The frustrating thing is that if you switch it on manually in the morning and then switch it off again, the PV charger stays on until the end of the day.

I'm thinking of installing a mini MPTT in the solar input branch, which will activate a relay and switch on the inverters every morning. But still, it's crazy to have to do this for a device of this price...

For the fans, i have the Noctua NF-A8 inside, and no problems at this time (temperatures are correct and the device consume about 5watts max on iddle with a perfect silent operation).

Edited September 15, 20241 yr by fhocorp

Fhocorp: "The very boring thing is that the auto-start of the the MPPT part of the inverter that is not working. Every day, i have to manualy power on the inverter (via it's power button) for charging the batteries.

For the PV charger to work, the 230V output of the inverter must be switched on, otherwise the device remains completely switched off. The frustrating thing is that if you switch it on manually in the morning and then switch it off again, the PV charger stays on until the end of the day."

If so, this is a design mistake. However, the offgrid solar inverter is designed to have inverter ON all the time, so the mistake has an excuse. I've correct a similar  issue in two different ways, for both you need to solder two wires on the ON switch and control it from outside the inverter (meaning you turn the switch off forever).

1. use a 5w solar panel and drive a relay (preferable a DC current relay) to turn on-off the inverter (will work only if you do not need inverter when is cloudy)

2. use a 230V AC digital timer (it has 24Vdc internal supply, so you can supply it from inverter baterries) and program it ON in the morning and OFF in the evening, it had the disadvantage that you need to change programmed hours from time to time.

I've used a modified method 1 to control the charging current of my my DiY powerwall which is an independent device ( LiFePo4, BMS and inverter onboard). After two years I can say it works good, the AC charging current (from the solar inverter) it's proportional with the sun light.

 

On 2024/09/15 at 7:53 PM, rowesley said:

Fhocorp: "The very boring thing is that the auto-start of the the MPPT part of the inverter that is not working. Every day, i have to manualy power on the inverter (via it's power button) for charging the batteries.

For the PV charger to work, the 230V output of the inverter must be switched on, otherwise the device remains completely switched off. The frustrating thing is that if you switch it on manually in the morning and then switch it off again, the PV charger stays on until the end of the day."

If so, this is a design mistake. However, the offgrid solar inverter is designed to have inverter ON all the time, so the mistake has an excuse. I've correct a similar  issue in two different ways, for both you need to solder two wires on the ON switch and control it from outside the inverter (meaning you turn the switch off forever).

1. use a 5w solar panel and drive a relay (preferable a DC current relay) to turn on-off the inverter (will work only if you do not need inverter when is cloudy)

2. use a 230V AC digital timer (it has 24Vdc internal supply, so you can supply it from inverter baterries) and program it ON in the morning and OFF in the evening, it had the disadvantage that you need to change programmed hours from time to time.

I've used a modified method 1 to control the charging current of my my DiY powerwall which is an independent device ( LiFePo4, BMS and inverter onboard). After two years I can say it works good, the AC charging current (from the solar inverter) it's proportional with the sun light.

 

Nice idea👍. In my case, after some research, i have programmed the inverters batterycharger to "Solar Only" so the batteries never get charged by the utility grid. However, when then inverter is connected to the grid, it's stay in "Standby mode" so the inverter is not completely OFF (even with the power button in OFF position) so MPPT charger could start when solar panels have sunlight. In conclusion, i have just connected the utility input of each inverter to a tuya connected electric socket and programmed the socket to power on on every morning and the invers a in "standby" state every morning, ready for charging the batteries 😁

@rowesley @Coulomb I have a question for you :

For having more power on the AC output and using the same battery, can i connect the first inverter ac output on the grid input of the second inverter and the load output of the second inverter to my load devices ? Normally, the power of my load will be using the second inverter power and if too much power is used, the second inverter will go on bypass mode and i will use the first inverter power. If too much power is used, the two inverter will go to bypass mode and i will use the grid power. That's right ? 

Best Regards

 

42 minutes ago, fhocorp said:

@rowesley @Coulomb I have a question for you :

For having more power on the AC output and using the same battery, can i connect the first inverter ac output on the grid input of the second inverter and the load output of the second inverter to my load devices ? Normally, the power of my load will be using the second inverter power and if too much power is used, the second inverter will go on bypass mode and i will use the first inverter power. If too much power is used, the two inverter will go to bypass mode and i will use the grid power. That's right ? 

Best Regards

 

No, you can't. The output power of each inverter is limited to 3500w. Assuming you have two Vevor inverters and one battery, you can connect them at the same battery, but you have to split your loads in two completely separated branches. In that way you will have 2x3500W loads. The accumulator must have around 14kw to be discharged at max 50%. Useful will be a discharge down to 80% (10-12 years lifetime for Pb-carbon)

1 hour ago, rowesley said:

No, you can't. The output power of each inverter is limited to 3500w. Assuming you have two Vevor inverters and one battery, you can connect them at the same battery, but you have to split your loads in two completely separated branches. In that way you will have 2x3500W loads. The accumulator must have around 14kw to be discharged at max 50%. Useful will be a discharge down to 80% (10-12 years lifetime for Pb-carbon)

@rowesley Ok, thanks for your answer. in the inverter documentation the inverter will go on bypass mode when outpout power is over the limits so i will have only 3500W of power. If an inverter fail, it bypass output to the grid. So in my configuration i will have redondancy output if one inverter fail but no more power. 

14 hours ago, fhocorp said:

@rowesley Ok, thanks for your answer. in the inverter documentation the inverter will go on bypass mode when outpout power is over the limits so i will have only 3500W of power. If an inverter fail, it bypass output to the grid. So in my configuration i will have redondancy output if one inverter fail but no more power. 

Your creativity is good. 😉

Yes you may have redundancy for short time. If both inverters are 3500W, inverter A is supplying inverter B. Inverter B fails because the power is 3500W, inverter A will fails as well since the power through it remains 3500W...You will drawn energy from battery by two inverters, one of them just bypassing. Not very economical...

To increase your output power without external grid you need an on-grid and off-grid inverter, lets name it Daxtromn (there are other models as well). Use Vevor as offgrid inverter as normal use. Use the Daxtromn inverter programmed as on-grid and synchronized with Vevor inverter output.  Both inverters supplied from the same battery. The output power will be Vevor+Daxtromn. If Vevor will fail, Daxtromn will fail as well without any external automation.

The second option is to search carefully inside your inverter if the synchronization connector (or footprint of the connector) does not exist even it is not connected outside. This type of inverter seems to be a 2023 design. Most of the offgrid inverters have the on-grid/off-grid option. Actually, this feature must be a routine in the processor which search for the zero-cross of the sinusoidal output and gives a sinchronization output pulse, while is waiting for an input pulse. If the input pulse does not exists, the inverter is off-grid.

 

 

Edited September 22, 20241 yr by rowesley

3 hours ago, rowesley said:

Your creativity is good. 😉

Yes you may have redundancy for short time. If both inverters are 3500W, inverter A is supplying inverter B. Inverter B fails because the power is 3500W, inverter A will fails as well since the power through it remains 3500W...You will drawn energy from battery by two inverters, one of them just bypassing. Not very economical...

To increase your output power without external grid you need an on-grid and off-grid inverter, lets name it Daxtromn (there are other models as well). Use Vevor as offgrid inverter as normal use. Use the Daxtromn inverter programmed as on-grid and synchronized with Vevor inverter output.  Both inverters supplied from the same battery. The output power will be Vevor+Daxtromn. If Vevor will fail, Daxtromn will fail as well without any external automation.

@rowesley Thanks, creativity is my speciality 😁 

Yes, redondancy is mandatory for me because i will have sensitive devices behind the inverters (server bay, webcams....etc). Most of the time, the outpout devices never exceed 2500W. My first approach was for having extra power when needed. My new approach is for having redondancy when an failure occurs. Your idea of connecting an on-grid inverter in the hybrid inverter output is already in my project 😙 (i have a Solarmax 3000W grid-tie inverter ready for this 🙂).

Also, this will allow me to have more power via the panels (I have the power from the panel input of each inverter that will reach the battery).  Theoretically, I could draw power from 2 rails of panels at the same time (1 per inverter). Each inverter can handle up to 1500W of panels (160V Max). Keeping a safety margin, I could make 1450W per inverter, which would give me 2900W of total solar power. Not bad 😉

 

3 hours ago, rowesley said:

The second option is to search carefully inside your inverter if the synchronization connector (or footprint of the connector) does not exist even it is not connected outside. This type of inverter seems to be a 2023 design. Most of the offgrid inverters have the on-grid/off-grid option. Actually, this feature must be a routine in the processor which search for the zero-cross of the sinusoidal output and gives a sinchronization output pulse, while is waiting for an input pulse. If the input pulse does not exists, the inverter is off-grid.

 

I have completely checked the electronic board of the VEVOR hybrid inverter and except the internal USB port, i have not found any dédicated port for the syncronization board/connector ☹️

Here is a picture of the motherboard :

The only connector i have found that could be the syncronization connector is near the processor :

The connector in not soldered and no any text near him for information 🤨

 

3 hours ago, rowesley said:

You will drawn energy from battery by two inverters, one of them just bypassing. Not very economical...

Yes sir, but i have changed my battery design for having more capacity and adding an RS485 connection between the battery and the inverter (the inverter stops well before discharging the battery, but the fans remain activated and end up discharging the battery, which goes into safety mode. Obliged to reactivate the battery by restarting the inverter. Better management of the battery SOC is recommended, and requires communication between them). For more intensive use, i should migrate to a 48V battery but it will need to update the inverter to the 5500W model and this is not planned at this time. I will keep my project in 24V.

For information,my battery is lithium ion based. 

Each battery module is made up of several 18650 battery rails, each fuse-protected and replaceable without soldering. Each rail (also protected by a fuse) is connected in parallel with the others, and linked to a BMS. The BMS has an RS-485 output for the inverter and a temperature sensor. Power per module is rated at 50A maximum. The battery will consist of 2 x 4Kwh modules. I'll have a total of 8Kwh and 100A maximum draw. The battery will be installed in a thermo-ventilated industrial cabinet.

On 2024/09/22 at 1:30 PM, fhocorp said:

The connector in not soldered and no any text near him for information 🤨

Easun SMGII, Anenji and Sumry/Vevor are producing similar devices. I have seen that on 6200/5500 motherboards.  Perhaps it is a Voltronic OEM project sold to many Chinese companies for production. I do not think this firmware of EML3500-24L has synchro, however if you find some consistent documentation please post. Indeed you may process 1500W on PV input, I have tested that daily with a tracker. 🙂

1 hour ago, rowesley said:

Easun SMGII, Anenji and Sumry/Vevor are producing similar devices. I have seen that on 6200/5500 motherboards.  Perhaps it is a Voltronic OEM project sold to many Chinese companies for production. I do not think this firmware of EML3500-24L has synchro, however if you find some consistent documentation please post. Indeed you may process 1500W on PV input, I have tested that daily with a tracker. 🙂

Okay. I will try to get more information on this point. For the firmware, no chances of getting this from the supplier from analysis😔

Thanks for the information about the PV output 👍

At this time, my solar panels send too much power to the battery so it is overcharged and the inverter detect this and get in error state. I have to modify the charging rules to a lower value.

Also, the cooling of the device works better when the device is fixed on the wall (my first tests were on a table lol)

On 2024/09/15 at 2:51 PM, fhocorp said:

For the fans, i have the Noctua NF-A8 inside, and no problems at this time (temperatures are correct and the device consume about 5watts max on idle with a perfect silent operation).

I had the same idea. Did you need to change something on the connectors/pin layout or is it plug and play replacement?

On 2024/09/29 at 7:32 AM, Funstar said:

I had the same idea. Did you need to change something on the connectors/pin layout or is it plug and play replacement?

Hi @Funstar

You have to modify the connectors and wiring to work. red is +12V, black is GROUND. Blue is PWM and the yellow is RPM signal.

Edited October 8, 20241 yr by fhocorp
add confirmation for fans pinouts

One question, on my inverter i have RS485 connector. Wiring is below :

and my battery is 18650 based whith a Bluetooth/RS485 BMS :

The RS485 connection is 2 Wire based and the inverter have 4 wire connector.

After some researchs, 2 wires a half duplex connection and 4 wire are full duplex connection.

I have tried multiples connections but nothing work (the "LI" icon keeping flashing as unsucessful connection)

Any ideas ?

 

Re the four wires: my understanding is that the two RS-485 pins are in parallel, as are the other two. So only two wires required.

12 hours ago, fhocorp said:

I have tried multiples connections but nothing work (the "LI" icon keeping flashing as unsuccessful connection)

My guess is that the BMS doesn't follow one of the six or so protocols that the inverter supports. Some of those protocols ae CAN bus only. If so, you'll just have to use battery type USEr, and enter suitable float, bulk, and cut-off settings. In that case, the inverter's SoC display will be almost completely useless; hopefully the battery itself, or its BMS, gives you a reasonable SoC indication.

 

On 2024/10/09 at 7:07 AM, Coulomb said:

Re the four wires: my understanding is that the two RS-485 pins are in parallel, as are the other two. So only two wires required.

My guess is that the BMS doesn't follow one of the six or so protocols that the inverter supports. Some of those protocols ae CAN bus only. If so, you'll just have to use battery type USEr, and enter suitable float, bulk, and cut-off settings. In that case, the inverter's SoC display will be almost completely useless; hopefully the battery itself, or its BMS, gives you a reasonable SoC indication.

 

Perhaps: a) check if indeed A and B are tied together in the connector, b) use a slave MODBUS software on PC, and an RS485/USB converter.  See what is sending the inverter using a PC software like https://www.modbustools.com/modbus_slave.html

A BMS protocol configuration implemented in some inverters is described here: https://workdrive.zohopublic.com.cn/file/vrz5k03066cc0152f4740b9bfff16dd8afe47

or, do nothing and let BMS to do it's job. 🙃

On 2024/10/09 at 6:07 AM, Coulomb said:

Re the four wires: my understanding is that the two RS-485 pins are in parallel, as are the other two. So only two wires required.

My guess is that the BMS doesn't follow one of the six or so protocols that the inverter supports. Some of those protocols ae CAN bus only. If so, you'll just have to use battery type USEr, and enter suitable float, bulk, and cut-off settings. In that case, the inverter's SoC display will be almost completely useless; hopefully the battery itself, or its BMS, gives you a reasonable SoC indication.

 

@Coulomb Yes that i have suspected. Actually, i have tested the 3 parameters for lithium :

- LI 2 : Lithium BMS for US2000C Pylontech (Not working in my case)

- LI 4 : Standard BMS Protocol (Not working, LI icon flashing without communication)

- LI 6 : Lithium profile without BMS communication (Working without problems but no SoC or temperatures measured).

In the next month i will switch to US2000C Pylontech batteries for safety and monitoring purposes. The Pylontech batteries will be connected to the inverter via RS485 and the Solar Assistant will receive the battery status via the inverter.

 

My 18650 Powerbank will remains in my testing part of my installation, with a link to another Solar Assistant via an USB-RS485 dongle. The second Vevor inverter will be used for my testing installation.

@rowesley  I will check this 😁  BMS communication is mandatory for me for monytoring purposes because i have to control the system outside of my home (and i am regulary away for work)

 

Edited October 21, 20241 yr by fhocorp

2 hours ago, fhocorp said:

fhocorp: I will check this 😁  BMS communication is mandatory for me for monytoring purposes because i have to control the system outside of my home (and i am regulary away for work)

We all are at work... This "need" of seeing the BMS parameters all the time it is a false one. The same as "being connected" via phones. Stupid things  learned by our minds with the true purpose that other people behind corporations to earn money on our back.  You do not need solar assistant as well. A Raspberry pi is more expensive than a powerful 6w Thinkcentre ($30-$50) running windows by a Celeron processor without any fan.

I also have a BMS and I check it in the morning and in the evening on my phone. You can do nothing if your BMS stops working in the middle of day... that's it if white puff (smoke) was already coming outside from it....😝. Knowing is worse than didn't. 😆

 

 

 

4 hours ago, rowesley said:

 

Hi @rowesley, you are right, but in my case i already have a raspberry and it is powered directly on the batteries via an DC-DC regulator (as you said i could use an mini PC but this is too much wiring).

Yes of course i can't do anything if a fire happends at my home due to a BMS failure but i can be warned of that and calling the fire brigade. I my case, the final installation will be installed i a secure metal case (for security purposes) and a fire detector will be installed in the case. I case of fire, i can alert my neighbours about the fire (sound alarm).

Yes, monitoring is mot mandatory, but i can be very useful 😁

Edited October 21, 20241 yr by fhocorp

On 2024/09/29 at 4:48 PM, fhocorp said:

Hi @Funstar

You have to modify the connectors and wiring to work. red is +12V, black is GROUND. Blue is PWM and the yellow is RPM signal.

Did it now, but get an error 14 (fan blocked), even I can see it spinning (but with very low airflow). Any suggestions?

Edited November 30, 20241 yr by Funstar

3 hours ago, Funstar said:

Did it now, but get an error 14 (fan blocked), even I can see it spinning (but with very low airflow). Any suggestions?

Hi @Funstar i had also this error on my inverter. This error does not prevent operation of the device.

The fans spins measured are too low for the inverter so it create this error. The inverter should have a minimum RPM value hardcoded in the firmware.

On 2024/11/30 at 2:54 PM, Funstar said:

Did it now, but get an error 14 (fan blocked), even I can see it spinning (but with very low airflow). Any suggestions?

I do not get it: (i)you have a brand new inverter and you have error 14, or (ii) the inverter worked ok for a while, then error 14 occur?  Those two situations may have two different approaches.