REC BMS Sunny Island, Sunny Boy question: What happens when a heavy load turns off, the battery is nearly charged, and the sun is out?

[Maitake]

Under this scenario the REC can be calling for a charge current as low as .8 amps, the Sunny Boys are ripping at 5+kw to supply the load ( oven, dryer etc). What happens when the load switches off? Where does the power go? Actually I don't care where the power goes as long as nothing breaks. Does anyone know if this situation is handled gracefully by the SI SB REC combination?
Sunny boys will be likely set for CA rule 21.
Oh it's normally a grid tie system, but the SI is manually backfed during outages.
Thx

 

[toms]

The Sunny Island raises line frequency and the Sunnyboy(s) reduce output to zero. Battery voltage momentarily raises for the few seconds this takes. This is why SMA spec a certain size battery bank / kw of PV.

I have used this system for many years with no noticeable issues.

With my system for example, maximum battery charge voltage in the REC BMS is set to 54.5V. If the system is already at this voltage and the above scenario occurs, the battery voltage can raise to 55V with no issues. There is plenty of available headroom.

I have set up many LiFePO4 whole house systems, the AC coupled SMA with the REC is the best i’ve seen for many reasons.

 

[Maitake]

Great. Thanks toms.
Seems I'm on the right path ?

 

[timselectric]

Solar power isn't being pushed, it's being drawn.
If the load goes away. No more power is drawn.
No different from unplugging a vacuum cleaner, without turning it off. Current just stops flowing.

 

[Maitake]

Solar power isn't being pushed, it's being drawn.
If the load goes away. No more power is drawn.
No different from unplugging a vacuum cleaner, without turning it off. Current just stops flowing.

SMA put out a bulletin, apparently some SB/SI hardware software combinations were experiencing catastrophic hardware failure when used as grid backup. Seems settled now. Just wanted to get some real world input if the REC complicated this.

 

[Maitake]

The Sunny Island raises line frequency and the Sunnyboy(s) reduce output to zero. Battery voltage momentarily raises for the few seconds this takes. This is why SMA spec a certain size battery bank / kw of PV.

I have used this system for many years with no noticeable issues.

With my system for example, maximum battery charge voltage in the REC BMS is set to 54.5V. If the system is already at this voltage and the above scenario occurs, the battery voltage can raise to 55V with no issues. There is plenty of available headroom.

I have set up many LiFePO4 whole house systems, the AC coupled SMA with the REC is the best i’ve seen for many reasons.

You say the battery voltage momentarily rises to accommodate the power surge until the SI responds to the frequency shift. This makes total sense to me. Are any special settings required in the BMS to prevent the BMS contactor from opening in this scenario?

 

[toms]

You say the battery voltage momentarily rises to accommodate the power surge until the SI responds to the frequency shift. This makes total sense to me. Are any special settings required in the BMS to prevent the BMS contactor from opening in this scenario?

Just allow a little headroom between your absorb voltage and your cell high disconnect voltage. My balance voltage is set between 3.40V and 3.45V, cell high disconnect is set to 3.55V.

Due to the size of the battery (400ah) the voltage rise is often unnoticed, but i have seen it rise a fraction, not enough to put a cell over the balance voltage. Certainly not close to putting a cell over the high disconnect voltage.

 

[toms]

Solar power isn't being pushed, it's being drawn.
If the load goes away. No more power is drawn.
No different from unplugging a vacuum cleaner, without turning it off. Current just stops flowing.

That isn’t how the SMA system works at all. (or any grid tie inverter for that matter)

They will attempt to output maximum power (generally into the grid - done by raising their voltage a fraction over grid voltage)

In the case of the SMA, unless the Sunny Island commands the Sunny Boy to reduce output - it will overcharge the batteries until they fail.

 

[timselectric]

That isn’t how the SMA system works at all. (or any grid tie inverter for that matter)

They will attempt to output maximum power (generally into the grid - done by raising their voltage a fraction over grid voltage)

In the case of the SMA, unless the Sunny Island commands the Sunny Boy to reduce output - it will overcharge the batteries until they fail.

The question was, "what happens if a large load is disconnected ". Are you saying that these inverters are going to self destruct? Because they weren't designed for normal daily situations?
I was under the impression that these were known as high quality equipment.
I personally don't have any experience with them. But, most people talk very highly about them.

 

[Ampster]

The question was, "what happens if a large load is disconnected ". Are you saying that these inverters are going to self destruct? Because they weren't designed for normal daily situations?

A grid tie inverter does push power because it is designed to be constantly connected to the grid which presents an infinite load. Your assumption about GT inverters is not correct.
However when in an AC coupling scenario, as explained above, when the load drops, the grid forming inverter has to direct that power to the batteries until they are full and then increase frequency to cause the GT inverter to ramp down or shut down. That is the mechanism that prevents them from self destructing.

 

[timselectric]

Short answer: nothing bad happens.

 

[toms]

Short answer: nothing bad happens.

Nothing bad happens when set up correctly, which was what the original question was trying to determine. If you set the system up incorrectly you can damage your battery.

Your responses indicate you have no experience setting up a frequency controlled AC coupled off grid system.

 

[timselectric]

Nothing bad happens when set up correctly, which was what the original question was trying to determine. If you set the system up incorrectly you can damage your battery.

Your responses indicate you have no experience setting up a frequency controlled AC coupled off grid system.

Any system set up incorrectly, can do damage to itself or anything connected to it.
And as I previously stated, I have zero experience with this product. But, it's not that complicated.
If current has nowhere to flow (open circuit). It just stops flowing.

 

[toms]

So far off the mark it’s laughable.

Do you think a SCC open circuits a PV panel to limit output?

If you have no idea on how something works, it’s OK to ask.

 

[timselectric]

The question has nothing to do with a solar charge controller.

But, this conversation isn't helpful for the OP.
I wish them luck.

 

[toms]

The question has nothing to do with a solar charge controller.

But, this conversation isn't helpful for the OP.
I wish them luck.

The grid connect inverter IS the solar charge controller in an AC coupled system.

I agree that your information isn’t helpful to the OP.

 

[Hedges]

SMA put out a bulletin, apparently some SB/SI hardware software combinations were experiencing catastrophic hardware failure when used as grid backup. Seems settled now. Just wanted to get some real world input if the REC complicated this.

Have a link? I'd like to read it.

Solar power isn't being pushed, it's being drawn.
If the load goes away. No more power is drawn.
No different from unplugging a vacuum cleaner, without turning it off. Current just stops flowing.

Voltage is "push".
If load has a non-infinite impedance, current flows due to that push. If there is source impedance, voltage is split across source and load impedance.

In a GT PV inverter, current through an inductor helps it deliver a sine wave from PWM switching.
So long as it is connected to a bottomless grid, it pushes sine wave current into sine wave voltage.
In an island grid, the power goes into loads and battery inverter. If load suddenly shuts off, current through inductor causes voltage rise because it sees higher impedance.
Battery inverter can run its own switching power supply and start pulling down current to keep island grid voltage regulated. It (Sunny Island in this case) should do that for a few seconds until it has raised frequency to readjust GT PV output to match reduced load.

If battery can't accept that current (say 100A for 3 seconds), voltage would rise and GT PV inverter might drop offline. But voltage could spike rather high, not sure what inductor would do.

I have AGM so expect it to accept almost anything. Lithium BMS might not. I would think setting target max SoC a bit lower would help.

The Sunny Island raises line frequency and the Sunnyboy(s) reduce output to zero. Battery voltage momentarily raises for the few seconds this takes. This is why SMA spec a certain size battery bank / kw of PV.

100 Ah per kW of PV. Mine is about 1/3 that size.

Sunny boys will be likely set for CA rule 21.
Oh it's normally a grid tie system, but the SI is manually backfed during outages.

Newer Sunny Boys, either Rule 21 for on-grid and grid-backup, or off-grid for strictly off-grid.
Older ones with RS-485 switch between on and off grid parameters. Except, at least some fail to implement all off-grid correctly in this scheme.

 

[Maitake]

Have a link? I'd like to read it.



Voltage is "push".
If load has a non-infinite impedance, current flows due to that push. If there is source impedance, voltage is split across source and load impedance.

In a GT PV inverter, current through an inductor helps it deliver a sine wave from PWM switching.
So long as it is connected to a bottomless grid, it pushes sine wave current into sine wave voltage.
In an island grid, the power goes into loads and battery inverter. If load suddenly shuts off, current through inductor causes voltage rise because it sees higher impedance.
Battery inverter can run its own switching power supply and start pulling down current to keep island grid voltage regulated. It (Sunny Island in this case) should do that for a few seconds until it has raised frequency to readjust GT PV output to match reduced load.

If battery can't accept that current (say 100A for 3 seconds), voltage would rise and GT PV inverter might drop offline. But voltage could spike rather high, not sure what inductor would do.

I have AGM so expect it to accept almost anything. Lithium BMS might not. I would think setting target max SoC a bit lower would help.



100 Ah per kW of PV. Mine is about 1/3 that size.



Newer Sunny Boys, either Rule 21 for on-grid and grid-backup, or off-grid for strictly off-grid.
Older ones with RS-485 switch between on and off grid parameters. Except, at least some fail to implement all off-grid correctly in this scheme.

SB-OffGrid-TI-US-en-22.pdf
As a hardware engineer I read "system failure" as something broke that needed replacing. In my world faults can be reset, failures need replacing. Don't know if that holds in SMA land.

 

[toms]

SB-OffGrid-TI-US-en-22.pdf
As a hardware engineer I read "system failure" as something broke that needed replacing. In my world faults can be reset, failures need replacing. Don't know if that holds in SMA land.

It does hold true with respect to SMA. If you have too much PV for your system, or too little battery capacity - there is risk of permanent hardware damage.

With LiFePO4, the Sunny Island charge/discharge parameters are controlled by the BMS. So back to your original question, it is critical to allow headroom in your BMS charge settings for frequency shift power management.

In practice - there are thousands of AC coupled SMA / LiFePO4 systems that have been happily operating for many years. Unless you try really hard - or have zero understanding of your system requirements, you will be fine.

 

[RCinFLA]

Biggest issue with freq shift control on GT inverter output is it is relatively slow reacting.

Worst case situation is lots of PV generation, lots of direct house consumption, with little to no back feed to hybrid inverter, then someone turns off a heavy house load leaving a sudden large overproduction to deal with.

Should have an automatic load dump or ability to 'pull the plug' on GT to handle the sudden overproduction issue. A relay to disconnect the GT inverter based on max back feed current and battery voltage is simplest.

SolArk inverters recommend feeding GT input through Generator AC input so they can quickly open the generator connect relay to immediately get out of an overproduction situation that may damage inverter or battery. They also have load dump control option.

I believe the recently introduced low freq All-In-One hybrid inverters essentially are a normal LF hybrid inverter and GT AC output inverter in one box with PV feed to hybrid inverter AC output. Their advantage is everything is hard wired in the same box so they have direct and immediate control over the GT inverter output.