Hi all

Happy New Year!

I'm new to the solar scene, and my understanding of the science of electricity and solar is vague, so please bear with me.

Had a Synapse 5 kVA inverter for the last 18 months, with Dyness 3.55 kWh Lithium battery and no PV panels (I had ordered a 5 kWh battery, but this was what was installed, and and and ...) This worked well, and battery never went under 50% during "normal"Phase 2-3 loadshedding. I made the decision to add solar panels a couple of months ago. Surprise: Synapse is not on City of Cape Town's list of approved inverters.

Sold the Synapse, bought a Deye 5 kVA with 6 x Longi 555 W solar panels - This setup was commissioned last week, and all has been well, with performance as anticipated. Plan is to also replace the battery with a 5 kWh unit early in the new year.

I had browsed this site to see what the inverter settings should be (we are a 2-adult household), and have attached screen shots of how mine is set up.

Reason for the post: Tonight at around 20h00 my wife started dinner prep, and switched on the stove and microwave. She came and called me just now, saying "There is a problem with the inverter." Went to have a look: Deye alarm was beeping, and battery showed 1%. Immediately switched over to pure Eskom power on the secondary DB (Did not want to put any load on the battery., but not sure if this would have made any difference). Battery started charging, and all the house power was being pulled off the grid.

I had a look at the graphs to try and understand what had happened:

At 20:05, the house was drawing 3.43 kW, of which 1.91 kW was  coming from the grid, and 1.61 kW from the battery. SOC was 98%.

By 20:10, consumption was 1.22 kW, all coming off the battery. SOC 99%, grid 0.02 kW. (Why??)

By 20:15, consumption of 1.14 kW, all from the battery. SOC 98%, grid 0.02 kW.

20:20, consumption 6.06 kW, battery providing 1.53 kW and grid 4.63 kW. SOC 75%.

20:25, consumption 4.64, battery into charge mode (SOC 0%), grid providing all the power (5.34 kW)

That's when I bypassed the inverter. Battery is busy charging.

I guess the first question is: Why/how did the battery go from 98% to 0% in literally 5 minutes?? And the answer is kinda obvious: With the big load, it just didn't cope.

More importantly, why did the grid supply drop to 0 kWh in this time of need?? There was no loadshedding (in which case the stove and microwave as non-essentials would not have worked), and nothing had tripped anywhere. I thought that the battery would stop supplying at 20%, as per the "Battery Setting" screen - clearly not. Also, according to my System Work Mode setup, the battery should not have dropped below 50%. Looking at the days before, it operated exactly as per the System Work Mode setup, i.e. from 15:00-20:00 maintain SOC of 80%, and then to 07:00 next morning allow it to drop to 50%, but no lower. Last night, these were ignored, and it just plummeted.

Can someone have a look, and possibly a) explain what had happened, and b) suggest changes to the configuration to prevent a repeat? I have attached info, shout if anything else is required.

Many thanks

Willem

 

 

 

 

 

Detailed Data-20231231.xlsx

4 hours ago, WdW said:

Hi all

Happy New Year!

I'm new to the solar scene, and my understanding of the science of electricity and solar is vague, so please bear with me.

Had a Synapse 5 kVA inverter for the last 18 months, with Dyness 3.55 kWh Lithium battery and no PV panels (I had ordered a 5 kWh battery, but this was what was installed, and and and ...) This worked well, and battery never went under 50% during "normal"Phase 2-3 loadshedding. I made the decision to add solar panels a couple of months ago. Surprise: Synapse is not on City of Cape Town's list of approved inverters.

Sold the Synapse, bought a Deye 5 kVA with 6 x Longi 555 W solar panels - This setup was commissioned last week, and all has been well, with performance as anticipated. Plan is to also replace the battery with a 5 kWh unit early in the new year.

I had browsed this site to see what the inverter settings should be (we are a 2-adult household), and have attached screen shots of how mine is set up.

Reason for the post: Tonight at around 20h00 my wife started dinner prep, and switched on the stove and microwave. She came and called me just now, saying "There is a problem with the inverter." Went to have a look: Deye alarm was beeping, and battery showed 1%. Immediately switched over to pure Eskom power on the secondary DB (Did not want to put any load on the battery., but not sure if this would have made any difference). Battery started charging, and all the house power was being pulled off the grid.

I had a look at the graphs to try and understand what had happened:

At 20:05, the house was drawing 3.43 kW, of which 1.91 kW was  coming from the grid, and 1.61 kW from the battery. SOC was 98%.

By 20:10, consumption was 1.22 kW, all coming off the battery. SOC 99%, grid 0.02 kW. (Why??)

By 20:15, consumption of 1.14 kW, all from the battery. SOC 98%, grid 0.02 kW.

20:20, consumption 6.06 kW, battery providing 1.53 kW and grid 4.63 kW. SOC 75%.

20:25, consumption 4.64, battery into charge mode (SOC 0%), grid providing all the power (5.34 kW)

That's when I bypassed the inverter. Battery is busy charging.

I guess the first question is: Why/how did the battery go from 98% to 0% in literally 5 minutes?? And the answer is kinda obvious: With the big load, it just didn't cope.

More importantly, why did the grid supply drop to 0 kWh in this time of need?? There was no loadshedding (in which case the stove and microwave as non-essentials would not have worked), and nothing had tripped anywhere. I thought that the battery would stop supplying at 20%, as per the "Battery Setting" screen - clearly not. Also, according to my System Work Mode setup, the battery should not have dropped below 50%. Looking at the days before, it operated exactly as per the System Work Mode setup, i.e. from 15:00-20:00 maintain SOC of 80%, and then to 07:00 next morning allow it to drop to 50%, but no lower. Last night, these were ignored, and it just plummeted.

Can someone have a look, and possibly a) explain what had happened, and b) suggest changes to the configuration to prevent a repeat? I have attached info, shout if anything else is required.

Many thanks

Willem

 

 

 

 

 

Detailed Data-20231231.xlsx 40.55 kB · 3 downloads

Is the communication to the battery's BMS set up? Can you show a shot of the Li-BMS screen?

1 hour ago, GreenFields said:

Is the communication to the battery's BMS set up? Can you show a shot of the Li-BMS screen?

Hi there

I believe it has been - the battery has been doing what it was supposed to all week - it just freaked out yesterday.

I have attached the screen shot.

Those settings look OK. Unbalanced battery cells within the battery can cause sudden SoC drops like this if one of the cell’s voltage dips below cut-off: the BMS will report zero to prevent damage to that cell.

How old is the battery (and roughly how many cycles?) 

A quick Google yields a similar issue with a similar battery:

Dyness provides contact details at the bottom of the first page, I’d contact them.

 

It makes sense that the battery was actually nearly empty at 20:00 even though it was showing 100% SOC.

I couldn't explain though how in the few hours before 20:00 you were draining the battery, yet the SOC was never decreasing, actually seeming to dip and then top itself up again, but with no injection of grid power, almost as though it were circulating the battery's own energy and somehow logging it as a charge cycle. Makes no sense.

Might be something wrong the the battery, should be checked. Also, check if there are any fault codes recorded in the Info screen.

At the same time try to change the following:

1. Un-tick the grid charge box on the battery settings screen. Just keep the grid charging ticks in the time slots of the work mode.

2. On the Work Mode timer screen, set all the "Power" fields to 500W. Peak power delivery on 35A is around 1680W, but you really don't need to try to dump all the stored energy at the highest possible rate, at dinner-time or any other time, it will get used up even at a lower discharge rate before long anyway. Rather try to run the battery at below the peak rate, even below your baseload, so you're always blending grid and battery power. Basically middle-of-the-road pro-active power management, not reacting to hitting against BMS limits, ratcheting back the power to 1.6kW when you're trying to draw up to 5kW.

Might make no difference whatsoever, but it's what I'd try anyway.

 

6 minutes ago, GreenFields said:

It makes sense that the battery was actually nearly empty at 20:00 even though it was showing 100% SOC.

I couldn't explain though how in the few hours before 20:00 you were draining the battery, yet the SOC was never decreasing, actually seeming to dip and then top itself up again, but with no injection of grid power, almost as though it were circulating the battery's own energy and somehow logging it as a charge cycle. Makes no sense.

Might be something wrong the the battery, should be checked. Also, check if there are any fault codes recorded in the Info screen.

At the same time try to change the following:

1. Un-tick the grid charge box on the battery settings screen. Just keep the grid charging ticks in the time slots of the work mode.

2. On the Work Mode timer screen, set all the "Power" fields to 500W. Peak power delivery on 35A is around 1680W, but you really don't need to try to dump all the stored energy at the highest possible rate, at dinner-time or any other time, it will get used up even at a lower discharge rate before long anyway. Rather try to run the battery at below the peak rate, even below your baseload, so you're always blending grid and battery power. Basically middle-of-the-road pro-active power management, not reacting to hitting against BMS limits, ratcheting back the power to 1.6kW when you're trying to draw up to 5kW.

Might make no difference whatsoever, but it's what I'd try anyway.

 

Thank you for this. Will apply the changes you suggest, and see what it does.

Since I am a Philistine with these matters: What does the "Power" field do? This system was installed 10 days ago, and the installed set these fields, except that he kept the battery at 100% - "It's a small battery, and I don't want to stress it." I changed the values, to see if the system did what it was supposed to.

45 minutes ago, JayMardern said:

Those settings look OK. Unbalanced battery cells within the battery can cause sudden SoC drops like this if one of the cell’s voltage dips below cut-off: the BMS will report zero to prevent damage to that cell.

How old is the battery (and roughly how many cycles?) 

A quick Google yields a similar issue with a similar battery:

Dyness provides contact details at the bottom of the first page, I’d contact them.

 

Thanks - this is helpful!

The battery and original inverter were installed April 2021. So couple of months short of three years. We replaced the original Synapse inverter with the Deye, and added 6 x 555 Longi panels. Plan is to replace this battery with a 5 kW unit in the next couple of months.

Cheers

W

 

Edited January 1, 20242 yr by WdW

22 hours ago, WdW said:

Thanks - this is helpful!

The battery and original inverter were installed April 2021. So couple of months short of three years. We replaced the original Synapse inverter with the Deye, and added 6 x 555 Longi panels. Plan is to replace this battery with a 5 kW unit in the next couple of months.

Cheers

W

 

Hi.

An mistake so many make when there inverter are setting to % is only focus on the % and never on the volt so when something does change or go wrong there's confusion .

So a bit of advice learn to keep an close eye on both % and volts so if the system does freak out you have an better idea of what's going on 

Got at least 2 clients that lithium bms gives false % , at times they show 100% full but low volts othere time they show 80% full but volts indicate the battery is full . 

Both lithium are getting software upgrade when I'm back again  to see if new firmware will fix the issue 

Edited January 2, 20242 yr by GMAC

20 hours ago, GMAC said:

Hi.

An mistake so many make when there inverter are setting to % is only focus on the % and never on the volt so when something does change or go wrong there's confusion .

So a bit of advice learn to keep an close eye on both % and volts so if the system does freak out you have an better idea of what's going on 

Got at least 2 clients that lithium bms gives false % , at times they show 100% full but low volts othere time they show 80% full but volts indicate the battery is full . 

Both lithium are getting software upgrade when I'm back again  to see if new firmware will fix the issue 

Thanks, GMAC - to a novice like me, a percentage is just easier to understand! I will have to go and do some reading to understand what these should be.

Cheers

W