I am having the same issue, wont charge beyond 93%. Issue started after adding PV. Any suggestions ?

On 2022/07/26 at 10:29 PM, Dre said:

Thx, yes I think I'm going to upgrade firmware myself, they "Hubble" are not answering, emails, making empty promises they going to get mr. X y z to call me back but never happens.

 

Now I need to find the firmware, Problem is my battery only like a month old,and only started when I added pv

So your added PV is connected to your Inverter not the battery. Thus the problem logically lies with the Inverter and/or the battery interface. The Inverter controls the charge. Not sure why there is a conclusion that the Battery firmware has become a target when the Battery has no idea if and how many PV panels are hooked up to the Inverter. 
 

19 minutes ago, zsde said:

So your added PV is connected to your Inverter not the battery. Thus the problem logically lies with the Inverter and/or the battery interface. The Inverter controls the charge. Not sure why there is a conclusion that the Battery firmware has become a target when the Battery has no idea if and how many PV panels are hooked up to the Inverter. 
 

There is currently an issue with communication between Sunsynk and Hubble. Hubble is in the process of rectifying the problem. 
 

Personally, I also think it’s a Sunsynk issue as every time there is a firmware update, these issues seem to surface. 

On 2022/08/15 at 9:26 AM, Leshen said:

There is currently an issue with communication between Sunsynk and Hubble. Hubble is in the process of rectifying the problem. 
 

Personally, I also think it’s a Sunsynk issue as every time there is a firmware update, these issues seem to surface. 

Has Hubble confirmed it themselves that they are in the process of fixing it, I’m currently on the latest firmware from Sunsynk and so far it’s all good. Factory reset the inverter maybe ?

6 hours ago, SeaSickMama said:

Has Hubble confirmed it themselves that they are in the process of fixing it, I’m currently on the latest firmware from Sunsynk and so far it’s all good. Factory reset the inverter maybe ?

Their response was to use the Cloudlink to rectify the issue 

13 hours ago, Leshen said:

Their response was to use the Cloudlink to rectify the issue 

That is just crazy but yeah i have given them the Finger for other reasons. Moved all my new installs over to the SunSynk battery 

Price of the cloudlink is also crazy just to fix this issue with their own product 

Edited October 9, 20223 yr by SeaSickMama

On 2022/10/09 at 9:23 PM, SeaSickMama said:

That is just crazy but yeah i have given them the Finger for other reasons. Moved all my new installs over to the SunSynk battery 

Price of the cloudlink is also crazy just to fix this issue with their own product 

Agree, I also wont support them anymore, one of my friends wanted to buy a hubble, I convinced him to stay away. 

Send my battery in last month for repairs and still waiting.

On 2022/10/11 at 2:51 PM, JohanM said:

Agree, I also wont support them anymore, one of my friends wanted to buy a hubble, I convinced him to stay away. 

Send my battery in last month for repairs and still waiting.

I totally agree.  Hubble support is shocking.  It took my battery a month for the repair - could have been done within 7 working days or less.

I will not recommend buying a Hubble based on my experience.  The batteries are constantly problematic ... I'm lucky if my battery gets up to 98% SOC, even though there is more than enough PV available everyday.

Glad to see a few other people are also having issues with them  and agreeing that they useless 

 

just go with Sunsynk batteries they totally worth the spend otherwise Fredomwon is the alternative these days 

I've recently learned something on a different battery, but it might be relevant here as the fixes given all seem to address the same issue.

If you have a float voltage setting that is lower than the bulk/absorb and the inverter enters float stage early (for whatever reason) before the battery is at 100% SOC then the lower charge voltage prevents any individual cells reaching a high enough voltage to trigger the equalization settings in the BMS, so the battery appears to just 'sit there'.

I suspect that this is why hubble is advising people to change their float setting to the same as bulk/absorb and also why equalization is set to daily; to make sure the inverter provides a high enough voltage for long enough for the equalization process to complete and reach 100% SOC.

P.S. my BMS requires 3.5VPC before equalization kicks in

Edited October 17, 20223 yr by jumper

10 minutes ago, jumper said:

to make sure the inverter provides a high enough voltage for long enough for the equalization process to complete and reach 100% SOC.

Valid point, however what happens if your Inverter and the BMS can't even agree on what the current battery voltage is.
With SA I can monitor both and although my battery is still reporting full capacity after more than 400 cycles, I now do a monthly Voltage setting for 2 or 3 days to knock some sense into the both of them (so to speak), i.e. Inverter and BMS.

The issue is that my Inverter will report 53,6V battery voltage whilst the BMS is between 52,7V and 52,9V. I checked it with the Fluke  and the BMS is correct.
At this 53,6V reported Inverter voltage which stays constant from around 52,3 BMS voltage until its at 52,7V BMS I see how the capacity increases ever faster and the next moment it reports 100%.
When I set the Inverter to voltage charge (USE mode) then I have to dial in 54,1 V to get the battery to 53,6 V eventually. This then also reports 100% on the BMS. But now my cell voltages are in the 4,1 to 4,14 range whereas at the Li Protocol with 52,7V and 100% they are between 4,0 and 4,1V
Both are reported by the BMS as 100% and the charge stops.
So one has to wonder about who is in charge (Inverter or BMS) and why the Protocol (perhaps Inverter) uses it's own internal Voltage interpretation and the BMS reported voltage does not become the Master.
I have been monitoring this for months and initially I noticed that the 100% SOC was already occuring before the cell voltages even reached 4V.
Now that I do a Voltage charge setting once a month it has stabilised in the 4,1 to 4,15V cell voltage range when applying USE mode,  but the BMS will not get higher than 52,7V to 52,9V with the Li Protocol.
The Battery still reports 110A/h once at 100% whether its at 52,7V or the 53,6V with USE mode.
So what is one to believe?
My Battery lasts just as long as it did from the beginning, so from that perspective it seems to still have full capacity and I am satisfied.
Below are the screenshots of where the Battery is still charging just before 100%. Then the second set of screenshots are those taken right after the battery was at 100% and the  charge stopped and the battery went into discharge mode.
Notice how at that moment the Inverter reported Batt voltage drops to 53,1V


 

18 hours ago, zsde said:

With SA I can monitor both and although my battery is still reporting full capacity after more than 400 cycles...

I don't have SA or any BMS to inverter comms, I only work from voltage and leave the BMS to manage the charging and I monitor the battery with the BMS software directly from time to time.

18 hours ago, zsde said:

The issue is that my Inverter will report 53,6V battery voltage whilst the BMS is between 52,7V and 52,9V. I checked it with the Fluke  and the BMS is correct.

I find this intriguing, I would expect the inverter to report a lower voltage than the BMS due to losses in the cables and fuses between them. My inverter (axpert king) reports a lower voltage than the BMS as I expected, but they are never far off (I have 2m battery cable). My inverter actually gives 2 voltage readings, one at the battery and one at the SCC, but they are both lower than the BMS as in the image below. Sometimes it might go up to 0.3V difference, but not more that I've seen.

18 hours ago, zsde said:

At this 53,6V reported Inverter voltage which stays constant from around 52,3 BMS voltage until its at 52,7V BMS I see how the capacity increases ever faster and the next moment it reports 100%.

The image above is also taken right before the battery hits 100% SOC, but you can see in the graph that the BMS already started tapering the charge quite some time before that, on it's own, without comms. At around 98% SOC the BMS enters a strange state and goes into standby for a while (opens the charging circuit) and then charges for a few seconds and then goes back to standby for a minute or more and then back to charge and the cycle repeats for about an hour or so before it suddenly hits 100%, just as you report. The effect of this state of the BMS is the PV ramps right down (100W from 1.1KW available) and the charging current to the battery is drastically lowered. Once it hits 100% it will also go into discharge mode, but my battery voltage doesn't drop like yours, it stays at float voltage until the sun goes down.

I see the BMS manages itself quite well on it's own without comms and I'm not sure what the effect is of the BMS telling the inverter to change the charging current and voltage while it is in the state I described above, I suspect this might be what is causing problems and is why you and others are needing to do a USE mode charge once in a while without the comms cable.

Watching the BMS sitting in this state for so long before reaching 100% was what made me realize I needed to up my float voltage because the absorb stage would time out before 100% and then the battery would drop to float voltage which was not high enough to activate cell equalization so the few cells that were at a lower voltage could never fully charge.

Hopefully this info helps, even though it is a different battery, but I'm pretty sure it is the same BMS.

55 minutes ago, jumper said:

few seconds and then goes back to standby for a minute or more and then back to charge and the cycle repeats

That is similar to what I experience in USE mode.

The BMS comms via CAN bus with what I observe is more of an imperfect art than a reproducible technical marvel it's made out to be.

Here is a comparison of how my SOC reflects on a typical day when I use the USE mode settings and the second image of when I use the Li Protocol. And this second one is exactly what Growatt describes as their protective algorithm that uses the battery to power the Inverter as soon as it has reached 100%. It then uses the battery down to 95% whether there is solar or not, and at 94,9% it starts recharging. They claim it increases battery life and also prevents the charger switching on and off frequently. This can obviously only be validated if two identical systems (one in USE mode and one on Li Protocol) were put side by side over a 5 year or longer testing period.

 

Edited October 18, 20223 yr by zsde

32 minutes ago, zsde said:

The BMS comms via CAN bus with what I observe is more of an imperfect art than a reproducible technical marvel it's made out to be.

I tend to agree, even though I don't have personal experience, but from reading around the web it seems to still be very much a work in progress

34 minutes ago, zsde said:

And this second one is exactly what Growatt describes as their protective algorithm that uses the battery to power the Inverter as soon as it has reached 100%. It then uses the battery down to 95% whether there is solar or not, and at 94,9% it starts recharging.

This is also what my BMS does, it will charge to 100% and then only allow discharge until 96% and will then charge again. My inverter still powers the loads from PV though and doesn't use the battery much, it usually won't go below 98% by end of day like your first pic and doesn't hit 100% multiple times a day like yours does in Li. I have deactivated this once because I thought it was preventing the battery reaching 100% and it had the effect I expected; charged to 100% and stayed at 100%. I turned it back on after I figured out it wasn't the problem because I suspect it could cause the battery to get quite hot as it doesn't have active cooling.

Interesting to see the difference in the 2 graphs. It looks like, if the inverter doesn't get the SOC from the battery via comms then it doesn't use the battery as much as it does in Li mode. It seems the growatt algorithm is only active when it has comms (or accurate SOC). Looking at those 2 images my intuition tells me the second one would age the battery sooner as it is using it and cycling it more when it could be using PV.

33 minutes ago, jumper said:

Looking at those 2 images my intuition tells me the second one would age the battery sooner as it is using it and cycling it more when it could be using PV

The typical Li Protocol only powers the Inverter from the battery for as long as sufficient Solar is available to power the demand.
I have to disconnect the CAN comms when using USE mode else it will shutdown with an error.
USE2 mode is meant to be voltage control with BMS comms, but I found that behaviour identical to what the Li Protocol does in terms of discharge to 95% and the start to recharge. The USE2 mode is basically the same as Li except that user can dial in the voltages which is inhibited in Li Protocols.

There a couple of schools of thoughts around the discharge and charge cycles vs keeping Lithium at 100% and one has to read and decide which one of the theories to accept as gospel. I have not seen any long term real life tests where the two have been run side by side to prove that one is better for the battery than the other one. There are plenty of articles where tests were done to see what SOC level  produces the best longevity for Lithium. Whether any of those tests included regular discharge and recharge by 5% cycles I have not seen.

I only observe and analyse my system data. Once I see the cell voltages reporting around 4V or lower at the BMS reported 100% SOC, then I apply the USE mode for a day or two and then the cell voltages are more in line with the intended full charge battery voltage again. Till now my battery has delivered exactly what it promised. Others on this forum have not been as fortunate as can be seen from some other Inverter/Hubble combo posts and reports.

It's as much a learning and experimenting exercise for me as for others. Sometimes one needs to use common sense or gut feel and of course interact with others to gain more knowledge or experience.

Edited October 18, 20223 yr by zsde

21 hours ago, jumper said:

I've recently learned something on a different battery, but it might be relevant here as the fixes given all seem to address the same issue.

If you have a float voltage setting that is lower than the bulk/absorb and the inverter enters float stage early (for whatever reason) before the battery is at 100% SOC then the lower charge voltage prevents any individual cells reaching a high enough voltage to trigger the equalization settings in the BMS, so the battery appears to just 'sit there'.

I suspect that this is why hubble is advising people to change their float setting to the same as bulk/absorb and also why equalization is set to daily; to make sure the inverter provides a high enough voltage for long enough for the equalization process to complete and reach 100% SOC.

P.S. my BMS requires 3.5VPC before equalization kicks in

You are probably referring to cell balancing, equalization (or an equalization charge) is used on LA batteries to charge each cell to a higher voltage/ deliberate overcharge to remove sulfates from the plates.

Applying an equalization charge (over voltage) to a lithium battery you are likely to do more damage than what you are trying to solve.

3.5v per cell is rather high for the cell balancing to start, unless offcoarse you are referring to NMC chemistry. 

23 minutes ago, I84RiS said:

You are probably referring to cell balancing

Yes, I am referring to balancing, in my BMS settings it is called equalization though... I guess they mean cell equalization, but yes it can be confused with equalization charge of LA batteries.

23 minutes ago, I84RiS said:

3.5v per cell is rather high for the cell balancing to start

I also thought so after a lot of reading online as this was the default setting on my battery and 3.5V is also the cell over voltage alarm threshold which had me worried. I have tried lowering it to 3.4V (and 3.45V), but then the battery won't get to 100% SOC. I've played around quite a lot with the BMS settings over the last couple of months and 3.5V works best for me. With the correct charge voltage settings none of the cells ever go into protection and they balance perfectly. Here's another image, soon after completing 100% charge, in standby with only 2mV difference, I'm happy with my settings now

3 hours ago, jumper said:

At around 98% SOC the BMS enters a strange state and goes into standby for a while (opens the charging circuit) and then charges for a few seconds and then goes back to standby for a minute or more and then back to charge and the cycle repeats for about an hour or ...

The BMS is likely just doing what it has been designed to do. It is likely that one cell has reached the upper cell voltage limit. The BMS stops the absorption charge to try and balance the cells (reduce the voltage of the cell with the high voltage) and will allow a charge again once the voltage drops far enough. 

Not many BMS'es do cell balancing while still in an absorption charge (although some of the more clever BMS'es will do this by cutting of the charge to the highest cell while allowing the other cells to continue with their absorption charge).

 

Edited October 18, 20223 yr by I84RiS