No, I configured the config.h to connect to my network WiFi so it can connect with my HA MQTT broker. Didn’t not do that step, config WiFi?Did you log into the AP? It was called DIYBMS or something similar
You have the same esp32 I do.
No, I configured the config.h to connect to my network WiFi so it can connect with my HA MQTT broker. Didn’t not do that step, config WiFi?Did you log into the AP? It was called DIYBMS or something similar
So that is different than me. I flashed, logged into AP, configured wifi. verified on router, plugged into eg4 and shunt, noticed eps overload error, logged into IP address, and configured.No, I configured the config.h to connect to my network WiFi so it can connect with my HA MQTT broker. Didn’t not do that step, config WiFi?
You have the same esp32 I do.
Tried another esp32, simply flashed without modifying any Victron-CAN code. Do not see any AP WiFi SID from phone or computer to use. Might try to do this on a PC, maybe a library or something isn’t installing properly.So that is different than me. I flashed, logged into AP, configured wifi. verified on router, plugged into eg4 and shunt, noticed eps overload error, logged into IP address, and configured.
I did install what felt like a whole bunch of shit in visual studio code.Tried another esp32, simply flashed without modifying any Victron-CAN code. Do not see any AP WiFi SID from phone or computer to use. Might try to do this on a PC, maybe a library or something isn’t installing properly.
I think it’s cool you can see that info which is why I want it. lol. Can’t believe that web UI is not working on my devices.
My 3rd esp32 is the same.Well I give up. Tired in a windows machine and took the exact steps you did with raw clone from GitHub. There is no WiFi AP to connect to via WiFi from computer or phone. Only way for me to config WiFi is in code then flash but don’t get those cool screens you showed.
I followed the instructions on the wiki. https://github.com/sijones/VE.DirectMQTTCANBUS/wiki/Install-for-testing-branch@ScrotusGobbleBottom
Please confirm if these steps are the same as yours:
I've done this on three ESPs at this point, can't figure out why there is a difference. It's controlling the inverter just fine so I'm not worried but is frustrating I can't replicate what you're seeing.
- Download and install Visual Studio Code
- Clone this repository in VS Code: https://github.com/sijones/VE.DirectMQTTCANBUS
- Once dependancies are downloaded, flash by clicking build then upload in VS Code (no tweaks to code at all, just flash)
- Builds, uploads, resets ESP
- Then you see that WiFI SSID???? - this is what I don't see
lol. Well shit, that’s why. You used the testing branch with the Web UI. Doing that now on my spare ESP32.I followed the instructions on the wiki. https://github.com/sijones/VE.DirectMQTTCANBUS/wiki/Install-for-testing-branch
I did everything as per the wiki.
I didnt hit build at all/
Just
"Run Task and you need to complete 2 tasks, the first is to flash the firmware, second is to flash the webpage data in to the LittleFS partition.The 2 tasks are calledlatformIO: Upload([device chosen])PlatformIO: Upload Filesystem Image([device chosen])"
I have no idea what Im doing lol. I didnt know that was significant.lol. Well shit, that’s why. You used the testing branch with the Web UI. Doing that now on my spare ESP32.
I think we figured it out. Thanks for chiming in tho.What is the issue you have? @ChrisG / @ScrotusGobbleBottom... Trying to understand what problem you are discussing.
I added some code to Simon's original that creates a user web interface to monitor / control the system - is that what you need?
Late pm here, so might not get back until Monday
Interesting.Had to do a ESP clean reset on chip to get it to work.
You can download VS Code for Linux for this project.Hi everyone. Nice work @sijones2010
This is exactly what I am looking for now if you could package it all together in a nice pretty box so I can just plug it in….something to consider…if not I might ask my nephew to build it, but kids these days….
My days of Microsoft are long gone and only run Linux and try not to switch my pc on more than once a year now. Let’s just say I am IT worn out. 30 years in the industry and happily retired now and prefer to watch grass grow.
Thanks @sijones2010. I’m sticking with the non-testing branch. It has worked well for over a month for me.@ChrisG @ScrotusGobbleBottom
Glad you have worked things out, am sorry that I've not been around, I had some illness and life that took over for quite a while.
As well as the web GUI the code also runs on Core and Core2 M5Stacks (https://shop.m5stack.com/products/m5stack-core2-esp32-iot-development-kit?variant=35960244109476) and has implemented the display as well.
There is much more that I want to implement, these were just some of the features that made it so far.
Please note I can't confirm how stable the testing branch code is, it's not the version I run so please be careful, happy to get feedback from you.
The error "received data not correct" means data from the Victron is corrupt and is usually caused by the resistor divider circuit being used when not required, basically the signal voltage is dropped below the minimum logic level for high.
The shunt drift should be loss, I.e. soc reports lower than actual.@ScrotusGobbleBottom and I are discussing separately the voltage/SOC drift that comes out of the shunt as once the inverter sees 100% it disables charging and I’m not getting balancing time/voltage. It’s not a code/project thing, just drift.
My charge voltage is 55.6v but SOC is hitting 100% before that so balancing doesn’t occur (luckily my pack is balanced to .005v).The shunt drift should be loss, I.e. soc reports lower than actual.
You need to tweak your shunt to reset to 100% on the following
Battery voltage over xx.x
Current under xx
For time of say 15/20 mins.
Along with the current divider you should be able to pretty much get this working well.
The only other thing is I use esphome with ha to monitor all cells, if a cell hits 3.65 I disable charging to allow for the balancer to get them back in line and then restart when it goes under 3.4.
What's your charge efficiency set in the battery part of the shunt?My charge voltage is 55.6v but SOC is hitting 100% before that so balancing doesn’t occur (luckily my pack is balanced to .005v).
Not at my cabin right now but believe it is 99%What's your charge efficiency set in the battery part of the shunt?