Dyness B4874 Lithium battery failure

[John37]

I recently had installed a Growwatt 5kva inverter with 6x355watt panels and a Dyness B4874 battery. The main load was four fridges which draw approx 800w when running but up to 4000w briefly on start up as the fridges are wired to a single changeover switch to change between utility and inverter souces. We only get utility power from roughly 10pm to 5am so the procedure was for the battery to charge overnight and when the sun was up at 8am to switch the fridges to the inverter. This worked fine initially, with no sign of stress or overload from the inverter but after about a week the inverter would trip out immediately when the changeover was made, and in fact any load over a few watts would cause the inverter fail and try to reset. The battery SOC lights showed it as fully charged but whenever a load was applied the voltage would drop. The installers had a look at the problem and changed the inverter to a Mecer 5 Kva model SOL-1-AX-5P64. This had no effect and the problem remained. After more consultation, they then changed the battery for another Dyness B4874 battery and the installation worked fine again with only a minor niggle in that the inverter would stop charging the battery from the panels when it was almost full, but not restart charging until the battery was physically disconnected and then reconnected. However, after about a week and a half the old problem resurfaced and the inverter crashed on changeover and could not take any load. The battery SOC lights show it is fully charged, even the inverter charging light shows it is fully charged but as soon as the sun goes down the inverter goes dark until the utility comes back.

Obviously we have a problem with the battery, but two of them in a row?  Has the load from the fridges killed the battery in some way even though when discharging the SOC lights have always been at least one or more. A cable had been connected  fron the CAN/485 port on the battery to the RS232 port on the inverter but I'm told by another source that this will have no effect and perhaps the fact that there is no communication method between the battery and the inverter may have contributed to the failures.

I would be grateful for any advice or explanation i could get from your members to help me decide the way forward.

 

[SonOne]

Hi @John37 did you ever get to the bottom of this issue? I am supplying solar components, including batteries, to clients in southern Africa and looking at potentially adding Dyness units alongside Pylontechs, so am curious about this thread.

[John37]

Good morning, unfortunatly there was no response to my post and the supplier maintained that the battery had failed due to being overloaded. I have now replaced the Dyness with gel batteries and the system is now working fine. That's not to say that there is anything wrong with the Dyness batteries but if fitting one to a 5Kva system causes it to fail then I have my doubts.

 

[Gerrie]

On 2019/11/17 at 12:33 PM, John37 said:

The main load was four fridges which draw approx 800w when running but up to 4000w briefly on start up

It might be better to start only two fridges at first, and than after they are running then only start the other two. That would be a bit easier on your battery’s and will avoid that 4000W startup.

The Dyness B4874 can only handle 74A maximum load. So in your case the 4000W  load at startup that will draw about 90A is way over the capacity of the battery. Did you get bigger gel batteries otherwise they might also fail pre-mature?

 

 

[chris81964]

Dyness does not recommend the battery you have with a 5 kw inverter. They recommend at least 2 of them. Sorry about your experience

[John37]

Thanks chris81964, guess we never stop paying school fees!