Communication Error's

[techdotrom]

Hello,

Around every 8 - 12 hours after the ESP has been working I get an battery communication error.
When I look into the details on the inverter, it seems the " Battery charge voltage, Charge current limit and Discharge current limit" are all unpopulated... If I reset the ESP32, they are populated again.

Is there anyway to add a refresh for these values to be sent to the inverter every few hours or so?

I have tried this on SunSynk and a Solis inverter and the same thing happens.

Or does anyone else have any other ideas?

Thanks

[Uksa007]

It updates the inverter every 1 second.
Is the ESP32 connected to Wifi?
It gets unhappy it can't connect to wifi and reboots to try to connect.

If no wifi, comment out the Wifi and API components like below:

#wifi:
#  ssid: !secret wifi_ssid
#  password: !secret wifi_password
#  domain: !secret domain

#api

Mine has been up for 54 days without issue connected to wifi.

If you still have issues post logs showing when the issue occurs.

Consider supporting the project https://www.patreon.com/Uksa007Codedevelopment

[Uksa007]

Please consider supporting my work so I can continue to develop new feature and provide help and assistance.
https://www.patreon.com/Uksa007Codedevelopment

[techdotrom]

Hi, I still had issues. It seemed to be in the "if" loops around lines 211. It seems under certain circumstances the " battery charge voltage, charge current limit and discharge current limit" values do not get sent to the inverter which causes a BMS loss error. Anyway, the sunsynk/deye/gsl inverter have low voltage protection built in to the inverter so you don't need the BMS to request any charging. I just deleted all the " if " statements and it's been working fine for a few days now. Hopefully this information is helpful. Attached is the code I've been using for my test setup. Best regards, Carl Code removed not safe see below for solution

[Uksa007]

I still had issues. It seemed to be in the "if" loops around lines 211. It seems under certain circumstances the " battery charge voltage, charge current limit and discharge current limit" values do not get sent to the inverter which causes a BMS loss error.

Carl this is not a good solution, your code has no charging logic and will request charging the battery constantly, resulting in the battery being kept at the absorption voltage for long periods of time. This is bad for the battery health and safety.

Please use the Latest released code it has improvements for deye and other inverters, which should resolve your issues.
V1.13.3 Improve compatibility with Deye and other inverters
Latest released version can be found here

Regards

[techdotrom]

Yes that is the code I was using. I'll have a play with that code again. Maybe it doesn't like the values being set to zero... I don't know. Are the " battery charge voltage, charge current limit and discharge current limit" values being set constantly? As in, there is no possibility condition where these values are not being sent? Thanks.

…

On Sat, 13 May 2023, 23:11 Uksa007, ***@***.***> wrote: I still had issues. It seemed to be in the "if" loops around lines 211. It seems under certain circumstances the " battery charge voltage, charge current limit and discharge current limit" values do not get sent to the inverter which causes a BMS loss error. Carl this is not a good solution, your code has no charing logic and will request charging the battery constantly, resulting in the battery being kept at the absorption voltage for long periods of time. This is bad for the battery health and safety. Please use the original code with this, it should resolve your issues: You could try this code, it has better compatibility with Deye. #1 (reply in thread) <#1 (reply in thread)> I have removed your code as it is unsafe. Regards — Reply to this email directly, view it on GitHub <#18 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AYGMTHPQ32OYCANM2DBJHZ3XGABIPANCNFSM6AAAAAAWIDLDJ4> . You are receiving this because you authored the thread.Message ID: ***@***.*** com>

[Uksa007]

It’s new code, always sends charge voltage, which solves similar problems with Deye inverter.

The Charge Current is zero when no charging is required.
This how the inverter knows when to turn off charging.

Give it a try.

[techdotrom]

I don't get the BMS error but it is stuck in some sort of a loop?I've set the inverter to charge the battery to 50% from the grid and it just sits here.If I change inverter to discharge, put a load on, turn the load off, set it to charge then it starts working again.If I connect a BMS with coms to the inverter, the 3 parameters (BCV,CCL,DCL) on the right if the picture never changes even at 100 SOC - that was my thinking in removing the IF conditions in the code previously.-------- Original message --------From: Uksa007 ***@***.***>Date: Sun, 14 May 2023, 07:50To: Uksa007/esphome-jk-bms-can ***@***.***>Cc: Carl ***@***.***>, Author ***@***.***>Subject: Re: [Uksa007/esphome-jk-bms-can] Communication Error's (Discussion #18) It’s new code, always sends charge voltage, which solves similar problems in Deye. Give it a try. —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you authored the thread.Message ID: ***@***.***>

[Uksa007]

What do mean it’s in a loop?
Can you attach a log file from the esp32?
What is your battery chemistry and configuration, eg 16s lifep04
What is the charging status in the esp32.

[techdotrom]

Initially if I set it to charge to 20%, once it reaches 20% the Charge Current Limit gets set to 0. If I then up the charge to 30% it doesn't do anything unless I discharge the battery first. It's in "Wait" mode so I assume it'll get itself out of this in the next TOU window. [logs.txt](https://github.com/Uksa007/esphome-jk-bms-can/files/11472210/logs.txt)

[Uksa007]

It is in "Wait" which means it not requesting charge, as the voltage is not below the absorption_voltage - rebulk_offset
The BMS is reporting a pack voltage of 52.86v
It is waiting for the battery voltage to get below 52.7v then it will request charging, this is stop micro cycling the battery.

[Uksa007]

It will never request charging when the Battery voltage is above the absorption_voltage(charge voltage) - rebulk_offset
It is waiting for the battery voltage to get below 52.7v then it will request charging, this is stop micro cycling the battery.
Seems like it is working how it is intended, this is how the charging logic works, to prevent the battery being constantly charged.

Discharge the battery a little and it request charging.

[techdotrom]

The code is exactly the same as your example except I changed the max charge current to 40A. It's just one of my 16s LifePo4 test packs. The absorption voltage is set to 55.2, for 30mins. Charge to Absorption Voltage 55.2v Once 55.2v is reached it will continue charging for 30mins After 30mins the Battery Charge Voltage is set to 55.2 - bulk = 52.7v the battery will then not charge unless the inverter starts discharging the battery first. Then the whole sequence repeats? I'll start again - I must have done something wrong to get it to think it was at 100% SOC which is why it was in wait mode. Thanks

…

On Sun, 14 May 2023 at 15:54, Uksa007 ***@***.***> wrote: It will never request charging when the Battery voltage is above the absorption_voltage(charge voltage) - rebulk_offset For Lifep04 set the absorption voltage to 55.2v and rebult_offset to 2.5 and test. — Reply to this email directly, view it on GitHub <#18 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AYGMTHLP6KMCR3NWPCNXGQLXGDWZVANCNFSM6AAAAAAWIDLDJ4> . You are receiving this because you authored the thread.Message ID: ***@***.*** com>

[Uksa007]

The charging logic is designed for the inverter to supply charge current for a long amount of time, typically from solar which is at least 8hrs.

It's is best to set the inverter to charge to 100% and let the battery control when it is finished charging.

If the inverter is turning of the charge midway through the absorption phase the charging logic will not know and so the battery may not be 100% charged, though it would be pretty close.

If you don't want to charge to 100% set the absorption voltage lower.
eg Setting 54V Absorption voltage and 30min absorption time, is around 90% SOC

Thanks for the donation it is appreciated.

Regards