Jump to content

Sunsynk 5kW Inverter showing either 0 or 100% SOC and nothing inbetween.


Surfrat

Recommended Posts

Hi,

I have just installed a 5kw unit that is connected to a custom (discrete LiFePO4 3.2V 120Ah cells) battery pack. The battery pack is 2 banks of 16 giving 240Ah, 51.2V and 12.3kWh. This is protected with two "LBSA BMS 120A 16S 48V" BMS's. This is a dumb battery with no RS485 or CAN coms.

Question: What are the correct inverter battery settings to give more than just 0 or 100% for the SOC?

I currently have the following configured as in attached images.

The second image is my Solarman chart from yesterday with either 0 or 100% SOC.

The first image is my battery voltage in Solarman.

I also tried AGM % but this mode pulls power from the grid when it thinks the battery is 0%.

 

Thanks

Mark.

 

Screenshot - 2021-11-29 12.24.10.png

Screenshot - 2021-11-29 12.21.17.png

Screenshot - 2021-11-29 12.20.14.png

Screenshot - 2021-11-29 12.19.44.png

Screenshot - 2021-11-29 12.18.58.png

Screenshot - 2021-11-29 12.29.24.png

Edited by Surfrat
Link to comment
Share on other sites

35 minutes ago, Surfrat said:

Hi,

I have just installed a 5kw unit that is connected to a custom (discrete LiFePO4 3.2V 120Ah cells) battery pack. The battery pack is 2 banks of 16 giving 240Ah, 51.2V and 12.3kWh. This is protected with two "LBSA BMS 120A 16S 48V" BMS's. This is a dumb battery with no RS485 or CAN coms.

Question: What are the correct inverter battery settings to give more than just 0 or 100% for the SOC?

I currently have the following configured as in attached images.

The second image is my Solarman chart from yesterday with either 0 or 100% SOC.

The first image is my battery voltage in Solarman.

I also tried AGM % but this mode pulls power from the grid when it thinks the battery is 0%.

 

Thanks

Mark.

 

Screenshot - 2021-11-29 12.24.10.png

Screenshot - 2021-11-29 12.21.17.png

Screenshot - 2021-11-29 12.20.14.png

Screenshot - 2021-11-29 12.19.44.png

Screenshot - 2021-11-29 12.18.58.png

Screenshot - 2021-11-29 12.29.24.png

I don't know much about SunSynk but you have selected AGMV, yet you say you have a lithium battery.

Link to comment
Share on other sites

44 minutes ago, Antonio de Sa said:

I don't know much about SunSynk but you have selected AGMV, yet you say you have a lithium battery.

If there is no BMS comms then Sunsynk recomends using AGM voltage. Not great and I would not personally do it, but can work if he can get the right values.

@Surfrat have you thought of contacting LBSA for information?  

Link to comment
Share on other sites

Thanks for responding.

 

LBSA make intelligent BMS's with coms but I am not going to replace mine. I have done a lot of research of Lifepo4 SOC and you cannot base it on voltage alone as the voltage range from 40 to 100% is small (6 volts) on a 48V system. Even with a built in intelligent BMS it is only 20% accurate.

I have sent a query to Sunsynk and will post their answer here.

 

Thanks

Mark.

Edited by Surfrat
Link to comment
Share on other sites

 

7 hours ago, Surfrat said:

Screenshot - 2021-11-29 12.19.44.pngScreenshot - 2021-11-29 12.18.58.png

Why is your float and absorption so low?  You have 54/16 = 3.375v per cell, 98-99% charge is about 3.48v

On the flip side you have it configured to allow your battery to drain down to 2.625v per cell which is pretty low.  3v -> 2.625v is less than 1% of battery capacity so it seems strange to set such a low full charge voltage but then set such a low empty voltage.

Did LBSA give you these values?

6 hours ago, YellowTapemeasure said:

If there is no BMS comms then Sunsynk recomends using AGM voltage. Not great and I would not personally do it, but can work if he can get the right values.

@Surfrat have you thought of contacting LBSA for information?  

I would probably not use AGM voltages, I would do as you say and reach out to the "manufacturer" and ask them what they want your float and bulk charge voltages to be.  For my own lithium battery I use the same value for both float and bulk and I have mine targeting 3.48v per cell which is a good compromise between life and near full charge.  My cut-off is 3v per cell, again good compromise.  Using those values you should get pretty much 96% of the battery capacity.

4 hours ago, Surfrat said:

Even with a built in intelligent BMS it is only 20% accurate.

Patently false.  A quality BMS has a shunt internally so it can count the amount of power that the battery has provided down to the watt.  Based on the battery capacity it is perfectly able to determine how many watts is left in the battery.  They are super accurate.  Any decent pre-built battery can accurately determine the capacity left and will not rely on voltage alone.

There is a YouTube channel where a guy reviews Lithium batteries and all the high quality Lithium pre-built batteries come with BMS that have shunts and thus measure the Wh capacity the battery has.

On the flip side, why would you buy a dumb Lithium battery?  Was it a lot cheaper?  From what I can tell Hubble costs pretty much the same as LBSA and they have really high quality BMS in their batteries.

Edited by Gnome
Link to comment
Share on other sites

I don't know if this will help, but put some hysteresis in the battery settings of shutdown and restart.

It's not logical to have shutdown and restart both at 42V, what must the machine do? 🤔

Try ShutDown = 43V, Low Bat = 45V, and Restart = 48V.

Like others have advised, speak to LBSA about the correct settings.

Link to comment
Share on other sites

Thanks guys for all your input.

I am the end user and these settings were set by the installer.

 

Quote

It's not logical to have shutdown and restart both at 42V, what must the machine do?

Agree, I have changed this to ShutDown = 43V, Low Bat = 45V, and Restart = 48V

Quote

@Surfrat have you thought of contacting LBSA for information?  

I will contact them.

Quote

Patently false.  A quality BMS has a shunt internally so it can count the amount of power that the battery has provided down to the watt.

Apologies, I am on a learning path with this setup...

 

Quote

On the flip side, why would you buy a dumb Lithium battery?

I had a smaller system that was supplied with discrete cells and a BMS to make up the battery pack. I have just added to them for my bigger system.

 

I will chat to my installer and show him this thread.

 

Edited by Surfrat
Link to comment
Share on other sites

  • 1 year later...
On 2021/11/29 at 3:53 PM, Gnome said:

 

Why is your float and absorption so low?  You have 54/16 = 3.375v per cell, 98-99% charge is about 3.48v

On the flip side you have it configured to allow your battery to drain down to 2.625v per cell which is pretty low.  3v -> 2.625v is less than 1% of battery capacity so it seems strange to set such a low full charge voltage but then set such a low empty voltage.

Did LBSA give you these values?

I would probably not use AGM voltages, I would do as you say and reach out to the "manufacturer" and ask them what they want your float and bulk charge voltages to be.  For my own lithium battery I use the same value for both float and bulk and I have mine targeting 3.48v per cell which is a good compromise between life and near full charge.  My cut-off is 3v per cell, again good compromise.  Using those values you should get pretty much 96% of the battery capacity.

Patently false.  A quality BMS has a shunt internally so it can count the amount of power that the battery has provided down to the watt.  Based on the battery capacity it is perfectly able to determine how many watts is left in the battery.  They are super accurate.  Any decent pre-built battery can accurately determine the capacity left and will not rely on voltage alone.

There is a YouTube channel where a guy reviews Lithium batteries and all the high quality Lithium pre-built batteries come with BMS that have shunts and thus measure the Wh capacity the battery has.

On the flip side, why would you buy a dumb Lithium battery?  Was it a lot cheaper?  From what I can tell Hubble costs pretty much the same as LBSA and they have really high quality BMS in their batteries.

to be frank bullshit, shunts can drift so although they are excelent at counting current in and out they can't account for losses

they do calibrate themselves at the bottom and top of the charge cycle , but since many believe/practice the 20-80 thing if your battery doesn't go to those states the shunt doesn't recalibrate

but don't worry thanks to the drift at some point in time you will land up with a low voltage cut-off eventually and it will recalibrate 

doesn't LBSA use daly BMSs they are crap imo if it's the smart one where you can at least set the parrameters they are less crap

Link to comment
Share on other sites

On 2021/11/29 at 1:02 PM, Antonio de Sa said:

I don't know much about SunSynk but you have selected AGMV, yet you say you have a lithium battery.

lithium iRon phosphate batteries can be happily swopped out and charged with the same curves as lead acid ( if you have a 16s battery - if you have 15s the per cell voltage will be higher the bms will cut-off charging so not a biggy)

that is the one plus from the lifep04 batteries that uses cells of 3.2v lithium batteries built with these cells match lead acid voltages very closely

i as far as i have it on agm it's recommended to limt charging  to 0.2c so if the agm selection uses the same logic and charges at max 0.2c of battery capacity 

the inverter will go lighter on the lithium on agm setting

Link to comment
Share on other sites

On 2023/02/18 at 3:26 PM, Leondavibe said:

to be frank bullshit, shunts can drift so although they are excelent at counting current in and out they can't account for losses

they do calibrate themselves at the bottom and top of the charge cycle , but since many believe/practice the 20-80 thing if your battery doesn't go to those states the shunt doesn't recalibrate

but don't worry thanks to the drift at some point in time you will land up with a low voltage cut-off eventually and it will recalibrate 

doesn't LBSA use daly BMSs they are crap imo if it's the smart one where you can at least set the parrameters they are less crap

A shunt is a low value resistor, the thermal and age drift will be in the parts per thousand at worst.  As far as the ADC is concerned, a shunt circuit uses a differential amplifier and voltage reference.

Drifts in shunts, ADCs and amplifiers are things people in metrology are interested in, in other words a 0.0001v change is considered significant.  For a shunt in a BMS that kind of drift will make 0 difference.

The kind of drift you are conflating this with is the batteries.  They are a chemical reaction and they aren't perfect buckets of water.  Some energy is wasted as heat and they lose efficiency over time.  That is where your so called "drift" is coming from.  But that has nothing to do with the shunt drifting, it's accuracy and repeatability will be far beyond what a battery requires.

These differential amplifier ADCs aren't unique to battery systems they are used in a LOT of applications, cars, planes, machines, lab equipment, hell your TV probably has some.  And the drift there is measured in years.  A car for example is expected to drift some amount in 10 years but not enough to materially cause a difference in the operation of the engine management system.  That is how little these things drift.

A ongoing chemical reaction in a battery on the other hand significantly drifts even from one month to the next.  But to say that is "shunt drift" is invalid conclusion.  The software is simply unable to determine accurately the true state of the battery without being reset because the battery has chemical changed.

Edited by Gnome
Link to comment
Share on other sites

 so for a laymen

What that says is a shunt isn't a useful means of measuring SOC if it doesn't calibrate from time to time

with what is happening in the battery

 

so though its measurement is exctremely accurate

 its measurement becomes irrelvant as what it thinks is the state, isn't the actual state

hence "drifts"  from what is happening in reality

 

so a rose by any other name is still a rose

 

Edited by Leondavibe
Link to comment
Share on other sites

On 2021/11/29 at 3:53 PM, Gnome said:

 

Why is your float and absorption so low?  You have 54/16 = 3.375v per cell, 98-99% charge is about 3.48v

On the flip side you have it configured to allow your battery to drain down to 2.625v per cell which is pretty low.  3v -> 2.625v is less than 1% of battery capacity so it seems strange to set such a low full charge voltage but then set such a low empty voltage.

Did LBSA give you these values?

I would probably not use AGM voltages, I would do as you say and reach out to the "manufacturer" and ask them what they want your float and bulk charge voltages to be.  For my own lithium battery I use the same value for both float and bulk and I have mine targeting 3.48v per cell which is a good compromise between life and near full charge.  My cut-off is 3v per cell, again good compromise.  Using those values you should get pretty much 96% of the battery capacity.

Patently false.  A quality BMS has a shunt internally so it can count the amount of power that the battery has provided down to the watt.  Based on the battery capacity it is perfectly able to determine how many watts is left in the battery.  They are super accurate.  Any decent pre-built battery can accurately determine the capacity left and will not rely on voltage alone.

There is a YouTube channel where a guy reviews Lithium batteries and all the high quality Lithium pre-built batteries come with BMS that have shunts and thus measure the Wh capacity the battery has.

On the flip side, why would you buy a dumb Lithium battery?  Was it a lot cheaper?  From what I can tell Hubble costs pretty much the same as LBSA and they have really high quality BMS in their batteries.

lithium will rest at 3.35v if fully charged so the notion of 98-99 is 3.48 is bullshit quite frankly

sure if you want to talk about measurements while charging maybe as with enough amps the voltage will lift 

 if you breach 3.45 you are most likely fully charged

so it depends how much amps you push try it set high amps then drop it when you hit that voltage point set low amps and see the voltage drop

so though the voltage is reading 3.48 does not mean it is at 3.48v

3.48 charging isn't the sane thing as 3.48v resting

But this is irrelevant as you are full not 99% and just pushing the steep part of the curve

 

so for float 3.35v per cell is perfectly fine  if you know they are full cause that is will a full cell will rest

Edited by Leondavibe
Link to comment
Share on other sites

I don't have a sunsynk

Have a cheapie synerji 

But also use a dumb battery 

Half shoto half diy 15s

ie no coms between inverter and battery

 

On a 16s i would set charge at  55.2-58.4

 

Charges a bit quicker at the last part of charging cycle if voltage is a bit higher ie above 3.45

But a bit lower voltage will

Be better health wise for battery 

Opinions vary

 So anything between 3.45-3.65 per cell

Edit :My shoto reported 99%soc on 3.43v on cells

And by 100% cells was 3.45

Settled resting  voltage 1.25hrs after charging in sleep mode  100%soc 

Is 3.35v per cell

 

I use 3.53 per cells

To compensate for a bit of a voltage drop  so 3.48v per cell effectively on the battery

3.4 means hours of floating to get full  

Sudden burst of sunshine that pushes voltage up can fool you

ie thinking battery is full meantime it isn't

You need a bit higher voltage to be able to get the amps into the battery

So theoretically it would be possible to get a battery 100soc at 3.35v per cell but you would probably take more than a day

While if you get it over 3.45v

You now it is full just removes the guess work

 ie at 3.35 you can end up with 70% soc or 100% with enough time

That is why i upped the voltage for voltage drop

For my inverter

Cause i have found that you can end at 70%soc as my inverter obviously has time linked to it's lithium charging algorithm Ie CC until it hits the voltage it will charge for x amount of time at constant voltage 

 And since at that lower voltage it just simply sees it as full after x amount of charging time 

While i can see the battery isn't full

I force a charging cycle then it charges full

Having the voltage i have it charges full every time 

 

Float at 53.6v

As a full lifep04 will come rest at a 3.35-3.4 can't recall will check when mine is fully charged in a bit

Ok so after  1.25hrs it has settled to 3.35v per cell with 100soc 

Sure You can keep it at a higher voltage , but for what it is full

 

Shutdown at 44v-48v 

ie 2.75v-3v per cell

dependant on where you want to shutdown 

There is so little power in a lifep04 under 3v that it won't buy you a big time difference anyway 3v is probably 9% soc but estimating it via voltage is problematic as sag can get you there  when closer to empty

 

Startup is up to you i would go 47-48 ,you just don't want it to close to the shutdown one

ie under use the voltage will sag especially under 3v per cell

And if the load is gone the voltage will settle if restart voktage is to close

 

 you will land in a start stop cycle ie

Voltage settles from the sag

Switches on load sags it shuts down settles start up sag rinse repeat

So if you get this up it a bit

 

 you want it to switch on when it starts charging again

Either from solat or grid

I would try 48v but maybe others with more experience with having their batteries at that state regularly can give input

I have not really paid attention to what the difference is once settled

 

Low battery i would say is up to you how soon you want a warning , dont know how sunsynk handles this 

ie what does it do only give a signal 

 

i would say anything under 49.6v as with sag you might be in the flat portion of the graph and the low battery warning will sound even though you aren't close to low yet

If you get false signal thanks to sag you can go lower anything above shutdown voltage to give you enough warning in your opinion

Edited by Leondavibe
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...