diy solar

diy solar

Sunny Island and EG4 Powerpro help

See page 16 for BYD battery quantity to support various number of Sunny Island. That would be for AC coupled systems, able to supply power and accept charging, while doing frequency-watts management.

You could try to scale specs of the BYD battery to those of a different brand. But voltage setpoints requested by BMS would be important to leave headroom to absorb the kW for a few seconds.

What I am really interested is knowing how these "Chinese Batteries/BMS" behave with the Sunny Island in an AC Coupling setup with closed loop communication. Especially when these batteries are not in the SMA Batteries "approved" list.

BMS should be communicating with the Sunny Island at all times or the SI will switch to standby mode after a minute (or so) of missing communications with the BMS. So when the battery is full the BMS cannot go into sleep mode or stop communicating with the Sunny Island.

Also interested in knowing what the Sunny Island will do if there is an alarm (i.e overvoltage alarm). Will it stop charging and stay on or will it go into standby mode and therefore house will shut off. Stuff like that is what I am trying to find out before dumping a few thousands on two LiFePo4 batteries for my two SI6048US.
 
What brand are you trying to use?

CC had SOK server rack batteries pretty much working closed loop with Sunny Island, but had some issues and gave up. Of course doesn't want to sell a combination that wasn't perfectly reliable.

OP of this thread got PowerPro working. We don't know if issues will surface. This would seem to be the one to go for, if looking for bargains compared to approved batteries.
 
What brand are you trying to use?

CC had SOK server rack batteries pretty much working closed loop with Sunny Island, but had some issues and gave up. Of course doesn't want to sell a combination that wasn't perfectly reliable.

OP of this thread got PowerPro working. We don't know if issues will surface. This would seem to be the one to go for, if looking for bargains compared to approved batteries.

I am looking at SUNESS batteries. They say they are compatible with SMA but cannot find much information on "real-life" users review. It appears to be a fairly new battery company from China. But no specifics if they are referring to the SMA from Europe, USA or both.

 
How are you charging the LiFePo4 battery? Do you have an AC Couple setup or are you using an MPPT Charge Controller? I have two SI6048 and I am trying to find out how that battery or a similar LiFePo4 battery will work in an AC Couple setup.
I have a grid tied AC coupled system (10.5kW solar panels, Sunny Boy x2, Sunny Island 5048 x 2). The batteries charge from the grid normally. When the grid is interrupted the Sunny Islands isolate the system from the grid. This allows the Sunny Boys to continue to operate and charge the batteries from solar production
 
Update on the Sunny Island/EG4 PowerPro combination. I am using the Li-extBMS setting on the SI and the SMA CAN protocol on the battery. The two are communicating via the modified CAT5 cable as detailed in the prior posts.

Charging
Under normal conditions (grid active) the SI's have been keeping the battery at 100%. On simulated and real grid outages I have taken the SOC down to as low as 35%. The SOC on the battery display and SI display match perfectly in that range. When the external power is reconnected the batteries charge quickly and taper/end charge appropriately.

When charging off grid the solar inverters provide power to the SI's for both load requirements and battery charging.

Discharging/Off grid
When off grid (no external power) the battery supplies an emergency load panel. Loads consist of two refrigerators, two air handlers, boiler electronics, well pump, septic pump, assorted electronics (computer, router, TV, etc), a couple outlets and some lights.

Our static loads are usually less than 1kW with peak loads rarely above 2kW. Peak battery amp draws haven't exceeded 25A. (These are based on the SI inverter logs.)

The battery appears to have no problem supplying the power required for the loads. HOWEVER....

Glitch
For some reason when the external power is cut and the inverter starts drawing from the batteries there is a problem. During the switchover there is a sudden voltage drop (56V-->25V-30V) per the SI logs. When the voltage drops this low it is below the minimum voltage for the SI and the inverters shut down. They then restart and operate normally. But this takes a few seconds.

This transient voltage drop and shutdown doesn't happen every time. Maybe one out of three. But it has also occurred once or twice during normal operation. With my previous lead acid batteries this NEVER happened. Switchover was so instantaneous that sometimes we didn't even know the power had gone out.

I'm in the process of trying to figure out the problem and have sent an email to support at Signature Solar (@SignatureSolarJames ). Waiting to hear back.
 
Update on the Sunny Island/EG4 PowerPro combination. I am using the Li-extBMS setting on the SI and the SMA CAN protocol on the battery. The two are communicating via the modified CAT5 cable as detailed in the prior posts.

Charging
Under normal conditions (grid active) the SI's have been keeping the battery at 100%. On simulated and real grid outages I have taken the SOC down to as low as 35%. The SOC on the battery display and SI display match perfectly in that range. When the external power is reconnected the batteries charge quickly and taper/end charge appropriately.

When charging off grid the solar inverters provide power to the SI's for both load requirements and battery charging.

Discharging/Off grid
When off grid (no external power) the battery supplies an emergency load panel. Loads consist of two refrigerators, two air handlers, boiler electronics, well pump, septic pump, assorted electronics (computer, router, TV, etc), a couple outlets and some lights.

Our static loads are usually less than 1kW with peak loads rarely above 2kW. Peak battery amp draws haven't exceeded 25A. (These are based on the SI inverter logs.)

The battery appears to have no problem supplying the power required for the loads. HOWEVER....

Glitch
For some reason when the external power is cut and the inverter starts drawing from the batteries there is a problem. During the switchover there is a sudden voltage drop (56V-->25V-30V) per the SI logs. When the voltage drops this low it is below the minimum voltage for the SI and the inverters shut down. They then restart and operate normally. But this takes a few seconds.

This transient voltage drop and shutdown doesn't happen every time. Maybe one out of three. But it has also occurred once or twice during normal operation. With my previous lead acid batteries this NEVER happened. Switchover was so instantaneous that sometimes we didn't even know the power had gone out.

I'm in the process of trying to figure out the problem and have sent an email to support at Signature Solar (@SignatureSolarJames ). Waiting to hear back.
Please let us know what you find out
 
What do the BMS logs report the voltage as being at the same time the SI sees it as 25v to 30v.
 
I don't suppose that "resistive load" fix is documented in, well, a "document"?

A 12 minute video is not documentation for how to configure equipment.

"Startup", to me is the single time you turn on switches and bring up the system. Does that fix this issue?
Or do you need to leave a constant resistive load on the battery 24/7 so your inverter is able to pick up your household loads when grid fails?

Adding a resistive load (e.g. 120V light bulb) on battery was my suggestion to idoco for debug.

OK, here's a potential kludge fix: SI has an option for SPDT signaling relays, "on grid". Use that to apply a resistive load to battery any time grid is connected. (Disconnected when offgrid, avoids discharging battery.) Don't connect light bulb (tungsten load) through the relay, because current at turn-on would be 10x as much. Just select a suitable resistor. And fuse the circuit, of course.
 
Not sure that will fix the issue, the firmware upgrade is to resolve the inrush when you connect the battery and the inverter capacitors are empty want to fill quickly. Your issue is after that point, the battery is charged and the inverter is on and the capacitors are full, all that is tripping the inverter is a voltage drop (real or possibly not) when the battery is going from charge to discharge.
 
Just got confusing. From the Sunny Island CAN Protocol technical documents:

"Required for the optimal operation of the product

If there is a pre-charge resistor required by the battery system, in order to prevent high inrush current, the value of this resistor shall be less than 50 Ohms because of required voltage rise of the internal power supply when powering up Sunny Island. Otherwise the inverter may not start (enters Low Battery Mode (LBM)"

Now I'm wondering if these low voltages in the Sunny Island log are "real" battery voltage drops related to load/current or just "phantom" low voltage readings as the inverter shuts down.

At this point I'll have to wait for the BMS communication cable in order to troubleshoot. Then I can compare the battery logs to the inverter logs. And update battery firmware if needed.
 
The shutting down of SI would not be from something phantom.

PowerPro it seems either disconnects load or disconnects all but precharge function (either resistor or FET gate voltage reduced.)
SI had been powering its 4W idle draw from AC, but when AC went away it had to deliver power and pulled down the voltage on its capacitors.

Battery needs to keep its BMS output FETs on.

Come to think of it, a few watts load on battery may not help. SI would be told to maintain float voltage. A resistive load would draw current at that voltage, which SI would supply. Battery would not see any difference, so would still turn off its output. Large enough load so there is battery voltage ripple when AC crosses zero, that would get its attention.

Firmware fix needed.
 
Update on the Sunny Island/EG4 PowerPro combination. I am using the Li-extBMS setting on the SI and the SMA CAN protocol on the battery. The two are communicating via the modified CAT5 cable as detailed in the prior posts.

Charging
Under normal conditions (grid active) the SI's have been keeping the battery at 100%. On simulated and real grid outages I have taken the SOC down to as low as 35%. The SOC on the battery display and SI display match perfectly in that range. When the external power is reconnected the batteries charge quickly and taper/end charge appropriately.

When charging off grid the solar inverters provide power to the SI's for both load requirements and battery charging.

Discharging/Off grid
When off grid (no external power) the battery supplies an emergency load panel. Loads consist of two refrigerators, two air handlers, boiler electronics, well pump, septic pump, assorted electronics (computer, router, TV, etc), a couple outlets and some lights.

Our static loads are usually less than 1kW with peak loads rarely above 2kW. Peak battery amp draws haven't exceeded 25A. (These are based on the SI inverter logs.)

The battery appears to have no problem supplying the power required for the loads. HOWEVER....

Glitch
For some reason when the external power is cut and the inverter starts drawing from the batteries there is a problem. During the switchover there is a sudden voltage drop (56V-->25V-30V) per the SI logs. When the voltage drops this low it is below the minimum voltage for the SI and the inverters shut down. They then restart and operate normally. But this takes a few seconds.

This transient voltage drop and shutdown doesn't happen every time. Maybe one out of three. But it has also occurred once or twice during normal operation. With my previous lead acid batteries this NEVER happened. Switchover was so instantaneous that sometimes we didn't even know the power had gone out.

I'm in the process of trying to figure out the problem and have sent an email to support at Signature Solar (@SignatureSolarJames ). Waiting to hear back.
Hey @idoco Please don't hesitate to reach out if you're having trouble! We're here to help!
 
I've been running the Sunny Island-5048 x 2 operating in split phase as master/slave for the last 10+ years and just upgraded to EG4 Powerpro batteries. With a single Sunny Island master connected to the Powerpro BMS the Sunny Island and battery communicate via the CAN bus and the battery charges fine.

The problem I'm having is that when running the system with the slave connected to the master, the slave is throwing a CAN bus error and goes into standby mode. The master continues to work fine. The master and slave are otherwise communicating over the CAN bus as the error on slave shows up on the master log.

The Poweredge supports SMA CAN protocol and is set to use the SMA CAN protocol. The Sunny Island is set to use Li-extBMS.

Anyone with experience with the Sunny Island have any suggestions?
Following as I have been looking into other batteries to work with SMA sunny island 60/48
 
... and that solved the problem!!!

The terminator for the CAN bus is built into the battery. But the terminator for the Sync bus needs to be added between pins 3 and 6 of the Cat 5 cable (see below).

Now I just need to get out the solder and heat shrink?
Hello I am very interested in your connections for your Sunny Islands. I have two 60/48 set as master and slave, did you have any other issues you had to rectify with the EG4 Pro. I have been looking at many other Battery manufacturers with the ability to communicate in closed loop.
Been looking to purchase three ,Seems like EG4 Pro are able to do it.
Is all your connections in the thread previous.
Thanks in advance
 
Back
Top