6000XP Not going to float. Just me?

[photons2e-s]

My 2 new 6000XPs are set to charge at 56.2V and 54.0V for float. No battery comm. Firmware is the latest.
They get to 56.2V and then stay there until the sun goes down. They never go to 54V.
It is damaging to the batteries to have them stay at the high V once full. That's why we switch to a lower float V.

Does anyone else with 6000XPs have this problem or is it just my 2 units?

Sig Solar is looking into this.

Thanks.

 

[EG4TechSolutionsTeam]

Feel free to send me a DM of the inverters serial number and I can look into this.

 

[photons2e-s]

Has anyone else with a 6000XP seen this? Or even checked their system for it?

 

[ScrodusGargleRotten]

Has anyone else with a 6000XP seen this? Or even checked their system for it?

I have never seen float either. I am using the JK inverter BMS with coms.
The BMS has a requested float and charge voltage setting and it still doesnt work.

 

[RyGuy]

If you get this figured out please post it up here. I have a luxpowertek sna 6000(6000xp sans breakers) with 2 non coms batteries. I have yet to hook up PV and would love to hear what becomes of this. Thanks.

 

[AshleyL]

I have never seen float either. I am using the JK inverter BMS with coms.
The BMS has a requested float and charge voltage setting and it still doesnt work.

Might be late in replying, but the reason is, all inverter based on Voltronic design ignores the float voltage command in BMS communication mode.

There is only CV charging mode (no float mode) plus the 95% - 100% SOC charging rinse repeat.

In voltage based control mode, you need to have 2Volt differential between the CV and Float.

 

[photons2e-s]

Might be late in replying, but the reason is, all inverter based on Voltronic design ignores the float voltage command in BMS communication mode.

There is only CV charging mode (no float mode) plus the 95% - 100% SOC charging rinse repeat.

In voltage based control mode, you need to have 2Volt differential between the CV and Float.

With my 6000XPs, I am using voltage control (not BMS comm) with 54 and 56.2V float and charge. Still, the units never go to float (54).
Is there anyone else out there that owns 6000XPs that can confirm this behavior ?

 

[ScrodusGargleRotten]

Never got float with or without comms

 

[photons2e-s]

Might be late in replying, but the reason is, all inverter based on Voltronic design ignores the float voltage command in BMS communication mode.

There is only CV charging mode (no float mode) plus the 95% - 100% SOC charging rinse repeat.

In voltage based control mode, you need to have 2Volt differential between the CV and Float.

I tried the 2V and 3V difference and that did not enable float.

 

[AshleyL]

I tried the 2V and 3V difference and that did not enable float.

Quick question, does the current still going in at your bulk voltage or the current already taper off to a very low level at that voltage?

 

[photons2e-s]

Quick question, does the current still going in at your bulk voltage or the current already taper off to a very low level at that voltage?

The current tapers off as the set charge V is reached. As expected.

 

[photons2e-s]

Update -- I heard back from EG4 on this. To summarize, they say that LiFePO4 batteries do not need to have a float V and are fine to keep at the ~56V charge voltage once they get to 100% SOC. The justification for this by EG4 was given here . I know that Victron at least does not agree.

So, the question now becomes, what is the best way to charge LiFePO4 for longevity? I thought that this had been figured out and tested extensively by now. Why is there such a strong disparity in the industry? Where is the test data by reputable companies?

 

[AshleyL]

Update -- I heard back from EG4 on this. To summarize, they say that LiFePO4 batteries do not need to have a float V and are fine to keep at the ~56V charge voltage once they get to 100% SOC. The justification for this by EG4 was given here . I know that Victron at least does not agree.

So, the question now becomes, what is the best way to charge LiFePO4 for longevity? I thought that this had been figured out and tested extensively by now. Why is there such a strong disparity in the industry? Where is the test data by reputable companies?

Standard reply from Voltronic OEM sales/customer service booklet. They are not wrong. Just they didn't tell you the complete story. Hell, even Growatt replied me with the same script.

Did they mention about that 100% SOC only works if BMS communication is utilized? It does not float at 56v in BMS comm mode. EG4 forgot to tell you that in BMS comm mode, it will bulk charge to 56V (or until the BMS reports 100%), then the inverter will stop charging the battery and operate with solar (priority) to power the load and if the solar is insufficient, the battery will cover the remainder.

It will not attempt to bulk charge again until the SOC drops below 95%. The usual 95% to 100% cycling.
Remember, there is no float in BMS comm mode. The battery will not be kept at 56v after the initial "bulk" is done and the voltage will naturally settle down to nominal voltage.

Disagree on the continuous >56v part. It will cause degradation of the electrolytes in the cell. Most people here just keep the bulk charge less than 1 hour for cell balancing to take place before switching to lower float voltage.

You can actually charge the LFP to full with just 3.4v percell (54.4v for 16s), but it will take a long time for the current to taper off. Around extra 3 hours in "lead acid charging mode"?

 

[photons2e-s]

I have not used the batt comm closed loop mode since I use different battery makers. I use open loop V-mode (which they call lead acid mode) which does not drop to float. Whatever battery is attached, in lead acid mode it will not drop to float. I assume that is not good for lead acid either.

EG4 also said that they are working on making a LFP mode for the 6000XP.

 

[AshleyL]

I have not used the batt comm closed loop mode since I use different battery makers. I use open loop V-mode (which they call lead acid mode) which does not drop to float. Whatever battery is attached, in lead acid mode it will not drop to float. I assume that is not good for lead acid either.

EG4 also said that they are working on making a LFP mode for the 6000XP.

Technically, they already have it, just not being "copy + paste" into their custom firmware yet.

In Growatt + Luxpower, there is a charging mode known as US2 (not BMS Comm mode, it is a charging mode supposely special made for lithium pack without BMS). Let say if you set the voltage to 56v, the solar inverter will charge it up to 56v and stop right after battery reached that voltage. Problem with this mode, it relies on voltage drops for the inverter to trigger bulk charge again. It is -2V for Growatt. In my case, whenever the LFP battery cell naturally settles down to its nominal fully charged voltage (3.350v x 16 = 53.6v), the inverter starts to bulk charge again to 56v as the trigger point for the bulk recharge is 54v (-2V). You can imagine my grip with this particular issue.

At the end, the proper charging method is BMS comm mode, but it relies on the BMS algorithm. I wil say this in advance, do not get the JK Inverter BMS. Its BMS algoritm is not suitable for Voltronic based design.

 

[Crowz]

Or you can do what I did and buy a victron 135/50 and some panels for it and let it handle the float job

 

[AshleyL]

Or you can do what I did and buy a victron 135/50 and some panels for it and let it handle the float job

Hoho.....previously I was considering using separate charge controllers and using my growatt as pure inverter........until I saw the price of Victron, ouch.......Bought one 100/20 48v for experimental purpose (maxed out 55.2x20 =1104w, overpaneled,parallel the panels in 2S2P and clipped due to the controller limit) and it works pretty well. Note, the Victron charge controller is really hot at 1000w..........haha.....

Unfortunately, probably I will never buy another Victron, it is simply too expensive.

 

[Crowz]

Well any mppt will do for topping the batteries off. Its not to recharge the batteries after they have been actually used. I just use it to handle the idle consumption. I only have 500 watts of panels on it. My main panels are on the sungold 10kw itself for handling loads and recharging the batteries after they have been actually used to power the house from power outages.

 

[photons2e-s]

I just switched from Victron to EG4 based on $/Watt.
The Victron MPPT was versatile, being able to charge to a set V, then have a set time for absorption there in order for the balancing to take place. Then, dropping to a set V for float. That's the way to do it I think.
Why the heck dont these other brands do this? Its just some simple programming and testing.

 

[dtbaker61]

My 2 new 6000XPs are set to charge at 56.2V and 54.0V for float. No battery comm. Firmware is the latest.
They get to 56.2V and then stay there until the sun goes down. They never go to 54V.
It is damaging to the batteries to have them stay at the high V once full. That's why we switch to a lower float V.

Does anyone else with 6000XPs have this problem or is it just my 2 units?

Sig Solar is looking into this.

Thanks.

I can confirm this behavior, and will attach snapshots.....

I sent this information to EG4, and signature Solar, about a month ago and the response was that they would tell their programming team.... but I have never heard anything back, and they do not have a formal 'bug-tracker' or trouble ticket system so I dunno how we will be informed if there IS a fix ?!

I like the 6000xp otherwise, but I do NOT like the behavior of the charge controller.

I have installed two 6000xp's recently, and run open loop with batt-type=lead-acid because I have found closed loop comms 'problematic'. I like to be able to use Voltage setpoints, not calculated SOC%. It *should* be possible for the 6000xp charge controller to use a CC-CV charge curve that is good for Lithium, but it is NOT behaving as programmed. (firmware version ccaa-C0D)

I have set:
Vbulk = 56.1
Vfloat = 54.0
Veq = 54.0
Eq days = 0
Eq time = 0


this should result in battery voltage charging up to 56.1v once a day, *not* holding at that voltage, and settling to Vfloat.
If loads exceed pv and bring the batt down, charge should return to Float voltage (not Bulk/Boost/Charge voltage).
In most charge controllers set for Lithium, the controller won't bring batts up to 'end of charge' voltage again until the 'next day', as detected by more than x time with PV voltage < 100v

what the 6000xp does is float at the Bulk voltage (56).... this is NOT good for Li battery life and increases the risk of overvoltage cells and thermal runaway.

 

[dtbaker61]

Standard reply from Voltronic OEM sales/customer service booklet. They are not wrong. Just they didn't tell you the complete story. Hell, even Growatt replied me with the same script.

Did they mention about that 100% SOC only works if BMS communication is utilized? It does not float at 56v in BMS comm mode. EG4 forgot to tell you that in BMS comm mode, it will bulk charge to 56V (or until the BMS reports 100%), then the inverter will stop charging the battery and operate with solar (priority) to power the load and if the solar is insufficient, the battery will cover the remainder.

It will not attempt to bulk charge again until the SOC drops below 95%. The usual 95% to 100% cycling.
Remember, there is no float in BMS comm mode. The battery will not be kept at 56v after the initial "bulk" is done and the voltage will naturally settle down to nominal voltage.

Disagree on the continuous >56v part. It will cause degradation of the electrolytes in the cell. Most people here just keep the bulk charge less than 1 hour for cell balancing to take place before switching to lower float voltage.

You can actually charge the LFP to full with just 3.4v percell (54.4v for 16s), but it will take a long time for the current to taper off. Around extra 3 hours in "lead acid charging mode"?

--------

exactly right....

The point is that if I am running open loop, and using Voltage setpoints to program a charge curve, you *should* be able to get what you want!

I have worked with LFP batteries for years, and there is no way to obtain a 'proper' charge curve with the current firmware as far as I can tell.

 

[LVObserver]

Important things to keep in mind:

LiFePO4 batteries (really any Lithium batteries) do not self-discharge like lead acid batteries

If you keep pumping power into LFP batteries, they will start a fire

I suggest charging them to 90-95% and then leaving them alone

 

[dtbaker61]

Important things to keep in mind:

LiFePO4 batteries (really any Lithium batteries) do not self-discharge like lead acid batteries

If you keep pumping power into LFP batteries, they will start a fire

I suggest charging them to 90-95% and then leaving them alone

yes, this is why you want LFP batteries to 'float' at 54.0v, not at 56.0

What I am confirming is that the eg4-6000xp, running open-loop (with voltage setpoints), it is not possible to program the optimal charge curve for LFP. This will require a programming change from EG4, made available in a firmware update to the 6000xp, hopefully sooner than later if enough people here chime in.