Hi All, 

My Hubble AM2 is about 2-3 months old and paired with a DEYE 5kW inverter. The SOC% has started to creep from the start. 

I've send numerous mails to Hubble. They reply and say I should get a Cloudlink for easy trouble shooting. But paying R2.5k is a lot of money if they should fix the problem. They said they could send me loan Cloudlink, but two weeks and three emails later and still no answer.

Did anyone have a similar issue?

Please see my graphs attached. The problem should be self explanatory on the graph. The voltage stays the same but the SOC creeps lower by the day. My fully charged battery is now at about 60%.

Are you using the BMS CANbus as per the Hubble Inverter guide? https://www.hubblelithium.co.za/deye
If so, have you tried to do manual settings via Voltage control to observe if this changes the SOC profile?
Also scan through this thread which has accumulated vast amounts of user experience with this battery with different Inverters. Some were able to solve their issues (similar to yours) and others not.  

 

It doesn't look like you are reaching the bulk/absorb voltage of 54V suggested for the AM-2 in this doc: https://www.fullcirclesolar.co.za/wp-content/uploads/2021/02/Hubble-Lithium-charge-settings.pdf

The older manual suggests 53.6V, which it looks like you are also not reaching on your graphs, I think you need to increase your charge voltage.

11 hours ago, zsde said:

Are you using the BMS CANbus as per the Hubble Inverter guide? https://www.hubblelithium.co.za/deye
If so, have you tried to do manual settings via Voltage control to observe if this changes the SOC profile?
Also scan through this thread which has accumulated vast amounts of user experience with this battery with different Inverters. Some were able to solve their issues (similar to yours) and others not.  

 

Thanks for the response. 

From the day I installed it it was plug and play, then the SOC, battery temperature everything showed up. I tried manual but then I assume it only monitors the voltage, so it's not accurate. 

The hubble manual doesn't say much about the CAN bus setup (nothing like baud rates, IP addresses etc.) it was just plug and play and enabling lithium on the inverter.

Hi all, any other suggestions? I've fiddled with the charge voltage. Even ran it for a day without CANBUS to see if charging at 54V would help. But everything is back to defualt settings. 

My current state of charge when fully charged is at 38% and it keeps creeping lower. See graph attached.

This is concerning since the battery is about 3 months old. 

4 minutes ago, Gerrit123 said:

Hi all, any other suggestions?

Do have a look at this thread . @swazz99jhb ran into a similar problem and Hubble assisted with some smoke and mirror Firmware and capacity change in my view.
Perhaps his experience so far can assist you to go the same route he did. 
 

 

Edited December 3, 20223 yr by zsde

7 minutes ago, zsde said:

Do have a look at this thread . @swazz99jhb ran into a similar problem and Hubble assisted with some smoke and mirror Firmware and capacity change in my view.
Perhaps his experience so far can assist you to go the same route he did. 
 

 

Thanks for giving me a good lead. It looks to be quite complicated in that thread. I hoped the Hubble would be plug and play. But seems like there are a lot of BMS issues on these batteries. 

I have 2 Hubble AM 2 and my oldest one is 5 months old.

I had been getting some odd data from it and some weird  behavior.

2 days ago it caught fire.The other hubble was the slave and it correctly shut down.The stuffed one was glowing inside and nasty smoke was coming out of the vents.

I switched both off and isolated them and my installer has sent it to the Hubble suppplier here in Dbn.

I have been trying to connect the riot cloud to my system but with no joy as its not working.

Hubble seem to be good and am sure they will replace it as its only 5 months old.

Wow that's scary. Thanks for the warning. Then I won't run mine based on voltage measurement, but rather keep the BMS plugged in and controlling the battery. 

45 minutes ago, Gerrit123 said:

Hi all, any other suggestions? I've fiddled with the charge voltage. Even ran it for a day without CANBUS to see if charging at 54V would help. But everything is back to defualt settings. 

My current state of charge when fully charged is at 38% and it keeps creeping lower. See graph attached.

This is concerning since the battery is about 3 months old. 

I would send those graphs to Hubble, there is clearly something wrong.

Are you able to see individual cell voltages on the inverter, maybe there are some clues there?

1 minute ago, jumper said:

I would send those graphs to Hubble, there is clearly something wrong.

Are you able to see individual cell voltages on the inverter, maybe there are some clues there?

I've sent them a mail for the 4th time just now. Someone has called me once about 3 weeks ago and said they will send me CloudLink to trouble shoot with, but nothing after that. I'll give them a call on Monday. 

It's just frustrating to buy the more expensive battery with local support so it's plug-and-play with the DEYE inverter. But then the cheaper imported options would have been easier.

They've sent me a mail to do the following, but that didn't work:
1) Disconnect the CAN comms cable.

2) Change the battery type o the inverter to AGM V.

3) Please input the Float, Bulk and equalization Voltage at 53,6Vdc.

4) Please set the Low Voltage to 44Vdc.

5) Kindly charge up battery until the inverter stops charging the battery at 53.6Vdc, this may take some time as the battery will do a cell balance at full voltage.

6) Once inverter has stopped charging, kindly reconnect your CAN cable and go back into your inverter battery menu and activate the Lithium settings and CAN communications.

59 minutes ago, Gerrit123 said:

Wow that's scary. Thanks for the warning. Then I won't run mine based on voltage measurement, but rather keep the BMS plugged in and controlling the battery. 

Yeah the one that bombed out had been playing up for a couple of weeks and we had tried everything.Hubble support were fine but until a cloudlink is in place then they cannot see whats happening.It was crazy stuff.

4 hours ago, Gerrit123 said:

It's just frustrating to buy the more expensive battery with local support so it's plug-and-play with the DEYE inverter.

Totally understand your frustration, these things are expensive so one expects them to work without issues, at least for a few years.

One thing I noticed in the instructions from Hubble is that on other occasions for users here on the forum there was another point between 4 & 5 instructing them to first discharge the battery until cutoff before charging it up again.

19 minutes ago, jumper said:

Totally understand your frustration, these things are expensive so one expects them to work without issues, at least for a few years.

One thing I noticed in the instructions from Hubble is that on other occasions for users here on the forum there was another point between 4 & 5 instructing them to first discharge the battery until cutoff before charging it up again.

I'll give it a try. Thanks for the tip.

On 2022/12/03 at 12:10 PM, Nicholas Strachan said:

2 days ago it caught fire.The other hubble was the slave and it correctly shut down.The stuffed one was glowing inside and nasty smoke was coming out of the vents.

Sorry about the frightening experience. Looks like the internal fire suppression system contained the fire inside the battery casing. Did you take any photos or videos?

I didnt but it appears the bms was the part that caught fire and hubble will replace the battery.Crazy stuff ha ha

Hi All,

To close out this thread. Hubble assisted me a lot by sending me a Hubble Cloud Link module and remotely logging into my battery and updating the firmware. Now my battery is back at 100% SOC and working as it should. 

Thanks a lot to Hubble and their great service. I would highly recommend their products. 

Regards 

On 2022/12/03 at 10:59 AM, Gerrit123 said:

Thanks for giving me a good lead. It looks to be quite complicated in that thread. I hoped the Hubble would be plug and play. But seems like there are a lot of BMS issues on these batteries. 

My bms caught fire on 6 month old am 2

On 2022/12/27 at 3:57 PM, Nicholas Strachan said:

My bms caught fire on 6 month old am 2

Hi Nicholas, 

Any warning signs to look out for?

38 minutes ago, Gerrit123 said:

Hi Nicholas, 

Any warning signs to look out for?

My soc kept on dropping daily no matter how my pv or ac input.i changed the setting from soc to voltage on the ac charge setting and it got back to 100% for a few days and then conked.The final event was it burning out.