Replies: 8 comments 32 replies
-
|
Can you tell me the version of the integration you're using? How many panel strings do you have connected, 1 or 2? Also, can you show me the screenshot of qmodmaster with offset 122? The problem is that for some inverter models, the M160 sunspec model doesn't start at 122, so we need to find where it is in your case. For UNO-DM-* and REACT2-* models, the offset is 1104. If we can find where the M160 model starts, I can implement a fix for your model. You need to use qmodmaster to find it. |
Beta Was this translation helpful? Give feedback.
6 replies
-
|
can you set 70 as start adress (offset)? the first register should be 101 or 103. You have base address set to 1 and Little Endian configured (see the bottom status bar of qmodmaster):
configure qmodmaster settings like this (big endian):
Also, click the grid icon to show the column/row offsets, at start address 70 you should see this:
first register you should see 101 or 103: 101 means monophase inverter and 103 means 3 phase. |
Beta Was this translation helpful? Give feedback.
-
|
Here is a screenshot with start address 70: Yesterday I created a small application to scan all addresses up to 65535, but I could only read address 0 to 123. For all the others I got an "Illegal data address" error back. So it seems like my inverter does not support the M160 data? |
Beta Was this translation helpful? Give feedback.
-
|
Ok, at 70 you have value 101, which means you have a monophase inverter. There's a modpoll utility you can find, it's free, that does modbus queries via CLI. If you can't query past offset 123 it's strange. But could be that for smaller models like yours, they didn't enable M160 for some reason. Can you please show me what's available in qmodmaster starting at offset 100 for let's say 13-14 registers (the max available in your case)? What you could do is ask FIMER support for the register map of your specific model, so we can have an official document to analyze. |
Beta Was this translation helpful? Give feedback.
-
|
Sure this is everything from register 100 to 123:
I'll try to contact FIMER support to see if I can get a register map. |
Beta Was this translation helpful? Give feedback.
-
|
At 122 there's no M160 data. We need the official register map. |
Beta Was this translation helpful? Give feedback.
-
|
I will release a beta with the following code for M160, that takes into account 3 different offsets: check those 3 offsets with qmodmaster, in first register you should find the value 160. try:
# Model 160 default address: 40122 (or base address + 122)
# For UNO-DM-PLUS/REACT2/TRIO inverters it has different offset
invmodel = self.data["comm_model"].upper()
if invmodel.startswith("UNO") or invmodel.startswith("REACT2"):
offset = 1104
elif invmodel.startswith("TRIO"):
offset = 208
else:
offset = 122
read_model_160_data = self.read_holding_registers(
slave=self._slave_id, address=(self._base_addr + offset), count=42
)
_LOGGER.debug("(read_rt_160) Model: %s", invmodel)
_LOGGER.debug("(read_rt_160) Slave ID: %s", self._slave_id)
_LOGGER.debug("(read_rt_160) Base Address: %s", self._base_addr)
_LOGGER.debug("(read_rt_160) Offset: %s", offset)
except ModbusException as modbus_error:
_LOGGER.debug(f"Read M160 modbus_error: {modbus_error}")
raise ModbusError() from modbus_error |
Beta Was this translation helpful? Give feedback.
0 replies
-
|
v3.5.1-beta.4 has been released, please upgrade and let me know how it goes. Thanks. |
Beta Was this translation helpful? Give feedback.
0 replies
-
|
@pdebruine can you test latest beta for me? I introduced a new method to look for M160 data, trying ALL offsets. I think M160 data is available ONLY when using Multi-MPPTs: it doesn't depend on the model, etc. It's v3.5.5-beta.1. if you can send me a debug log you do me a favor, I want to see how it works with other inverters. Thank you. |
Beta Was this translation helpful? Give feedback.
1 reply
-
|
@alexdelprete I installed the latest beta and tried to add the integration, I got this error in the log: |
Beta Was this translation helpful? Give feedback.
-
|
That error is really weird. Never seen it before. Please test v3.5.5 final. I tested 3.5.5 with 3 other users, didn't have issues. |
Beta Was this translation helpful? Give feedback.
2 replies
-
|
@alexdelprete I have the same error with 3.5.6, and I also tried 3.5.4 but that also gives the same error. I had to go back to version 3.5.2 to get my initial error. So it seems I was running that version (or an older one) before. From version 3.5.3 to 3.5.6 I get the error. The error seems to be coming from the pymodbus module: |
Beta Was this translation helpful? Give feedback.
-
|
@alexdelprete Just noticed that is looks like the Slave ID is now a float number: when before it was an integer: Perhaps that is the issue? From the commits it looks like you did change how the input for the setting is displayed. |
Beta Was this translation helpful? Give feedback.
-
|
I noticed that, and I modified the code: class ABBPowerOneFimerCoordinator(DataUpdateCoordinator):
"""Class to manage fetching data from the API."""
config_entry: ConfigEntry
def __init__(self, hass: HomeAssistant, config_entry: ConfigEntry) -> None:
"""Initialize data update coordinator."""
# get parameters from user config
self.conf_name = config_entry.data.get(CONF_NAME)
self.conf_host = config_entry.data.get(CONF_HOST)
self.conf_port = int(config_entry.data.get(CONF_PORT))
self.conf_slave_id = int(config_entry.data.get(CONF_SLAVE_ID))
self.conf_base_addr = int(config_entry.data.get(CONF_BASE_ADDR))
self.scan_interval = int(
config_entry.data.get(CONF_SCAN_INTERVAL, DEFAULT_SCAN_INTERVAL)
)It's in latest release. Did you test it? I don't have your error, and 3 other users neither. So I'm a little bit confused about your issue. What HA version are you using? Please post the FULL debug log, filtering with "abb" in the HA search input and post it here. I'll compare it to mine. |
Beta Was this translation helpful? Give feedback.
6 replies
-
|
If you can, delete the integration, restart HA, and reinstall the integration from scratch. |
Beta Was this translation helpful? Give feedback.
-
|
@alexdelprete I did everything you mentioned, I deleted the integration, rebooted, installed the latest 3.5.6 version, rebooted again and just tried to add the integration, but still the same error. This is the full log filtered by abb: and this is the full log without filtering from after trying to add the integration: Still seeing the slave id as a float in the logs. I'm running this version of HA: Core 2025.1.0 Are you testing from an integration that is already setup? Maybe that is the difference? I'm still trying to add the integration from the devices menu. |
Beta Was this translation helpful? Give feedback.
-
|
I think the type system of Python doesn't work as you expect it, if I run this code: on python-fiddle.com I get this output: So even though I defined a variable as an int it's still possible to assign it a float value which is accepted without a problem. So my guess is that the HA config slider returns a float value, which is passed to pymodbus and that gives the error. |
Beta Was this translation helpful? Give feedback.
-
|
I will test this tomorrow, and I'll remove the slider (which I don't like anyways) but I wonder why I have no other reports of this problem from other users. I have 3 other users on latest release that I'm in contact with on Discord, and they didn't report any issue. Let alone that on my system I see no issue, obviously, but the dev should never rely on his system to chase bugs. :) Also, the other thing I modified in latest release is that I converted those values to int() in coordinator, that fetches data calling class ABBPowerOneFimerCoordinator(DataUpdateCoordinator):
"""Class to manage fetching data from the API."""
config_entry: ConfigEntry
def __init__(self, hass: HomeAssistant, config_entry: ConfigEntry) -> None:
"""Initialize data update coordinator."""
# get parameters from user config
self.conf_name = config_entry.data.get(CONF_NAME)
self.conf_host = config_entry.data.get(CONF_HOST)
self.conf_port = int(config_entry.data.get(CONF_PORT))
self.conf_slave_id = int(config_entry.data.get(CONF_SLAVE_ID))
self.conf_base_addr = int(config_entry.data.get(CONF_BASE_ADDR))
self.scan_interval = int(
config_entry.data.get(CONF_SCAN_INTERVAL, DEFAULT_SCAN_INTERVAL)
)so in your test code, you should try myMethod(int(slave_id)) and check the result. |
Beta Was this translation helpful? Give feedback.
-
Bingo! That's the difference, and I think I know where the bug is: get_unique_id() in configflow. I just made a few changes: removed the selectors (slider etc.) in config_flow and made sure all variables/parameters are converted to int. Maybe I exaggerated, but it's better to be safe than sorry. :) I'll release a beta, please test it and let me know. |
Beta Was this translation helpful? Give feedback.
-
|
I tested with an int conversion, like I do in the integration, and it works: slave_id: int = 247.0
def myMethod(slave_id1: int):
return slave_id1.to_bytes(1, 'big')
value = myMethod(int(slave_id))
print(value)result: b'\xf7'
** Process exited - Return Code: 0 **
Press Enter to exit terminal |
Beta Was this translation helpful? Give feedback.
5 replies
-
|
I have installed version 3.5.7-beta2, when I try to add the integration I still get an unknown error, this is in the log: I think that when checking the response from the modbus client you also need to check if the read succeeded, when it fails then an ExceptionResponse is returned instead of the registers. |
Beta Was this translation helpful? Give feedback.
-
|
Ok, I added other two exceptions to be managed. Test latest beta.3 |
Beta Was this translation helpful? Give feedback.
-
|
Still getting the same error: I saw you added some exception handling, but I think this call: Is not throwing an exception, but is returning an ExceptionResponse object. So my guess is that you would have to check if the response is an ExceptionResponse and handle that. |
Beta Was this translation helpful? Give feedback.
-
|
Try beta.4 and let me know. |
Beta Was this translation helpful? Give feedback.
-
|
Yes, I was able to add the integration, here is the log from adding the integration: |
Beta Was this translation helpful? Give feedback.
-
|
Sorry for getting back at you late on this. Good to know now we manage errors for these inverters correctly, it helped making the code more resilient. But I see a weird issue: your inverter reports it has 2 MPPTs, but I see data only of one. How many strings you have, 1 or 2? 2025-01-22 14:50:33.860 DEBUG (SyncWorker_6) [custom_components.abb_powerone_pvi_sunspec.api] (read_rt_160) Model: UNO-3.0-TL-OUTD
2025-01-22 14:50:33.860 DEBUG (SyncWorker_6) [custom_components.abb_powerone_pvi_sunspec.api] (read_rt_160) Slave ID: 247
2025-01-22 14:50:33.860 DEBUG (SyncWorker_6) [custom_components.abb_powerone_pvi_sunspec.api] (read_rt_160) Base Address: 0
2025-01-22 14:50:33.860 DEBUG (SyncWorker_6) [custom_components.abb_powerone_pvi_sunspec.api] (read_rt_160) Offset: 122
2025-01-22 14:50:33.871 DEBUG (SyncWorker_6) [custom_components.abb_powerone_pvi_sunspec.api] (read_rt_160) mppt_nr 2
2025-01-22 14:50:33.871 DEBUG (SyncWorker_6) [custom_components.abb_powerone_pvi_sunspec.api] (read_rt_160) dc1curr: 0.17 Round: 0.17 SF: -2
2025-01-22 14:50:33.871 DEBUG (SyncWorker_6) [custom_components.abb_powerone_pvi_sunspec.api] (read_rt_160) dc1volt 237.0
2025-01-22 14:50:33.872 DEBUG (SyncWorker_6) [custom_components.abb_powerone_pvi_sunspec.api] (read_rt_160) dc1power 40
2025-01-22 14:50:33.872 DEBUG (SyncWorker_6) [custom_components.abb_powerone_pvi_sunspec.api] (read_rt_160) dc2curr: 0.0 Round: 0.0 SF: -2
2025-01-22 14:50:33.872 DEBUG (SyncWorker_6) [custom_components.abb_powerone_pvi_sunspec.api] (read_rt_160) dc2volt 0.0
2025-01-22 14:50:33.872 DEBUG (SyncWorker_6) [custom_components.abb_powerone_pvi_sunspec.api] (read_rt_160) dc2power 0
2025-01-22 14:50:33.872 DEBUG (SyncWorker_6) [custom_components.abb_powerone_pvi_sunspec.api] (read_rt_160) Completed |
Beta Was this translation helpful? Give feedback.
12 replies
-
|
I want to see logs when you added and also after 5 minutes, to see if data is being read correctly every polling cycle. |
Beta Was this translation helpful? Give feedback.
-
|
Sure, no problem. Here is the log from setting up the integration: and here is the latest sync log: Seems to be working correctly now. |
Beta Was this translation helpful? Give feedback.
-
|
So, basically, it doesn't find M160 in your inverter. But I remember that at some point we did find it at offset 122. Now it doesn't. It's weird... Can you check with qmodmaster at offset 122 if you see value 160? I remember we did find it in previous tests, not sure it was at offset 122 but we did find it. Or maybe I'm just confused with other users issues. :) |
Beta Was this translation helpful? Give feedback.
-
|
We had a look at that earlier: #205 (reply in thread) At offset 122 there is no 160, the value is 65535 |
Beta Was this translation helpful? Give feedback.
-
|
My goodness, I must have confused your case with someone else's. :) Sorry, and thanks again for the help in finding this problem. |
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hi, I'm trying to setup this integration with my ABB UNO-3.0-TL-OUTD inverter with the VSN300 WiFi logger card. I have installed HACS and the custom component, but when I try to add the device in HA I get an "Unknown error".
I see this in the logs:
As you can see it's reading some data from the inverter, but if fails in the read_sunspec_modbus_model_160 method. I had a look at the code and saw that it's trying to read data with an offset of 122 and a count of 42. I tried to do that myself but then I also get an read error in the Modbus Master tool.
After some experimenting I noticed that I can only read 2 register from offset 122, more than 2 registers will give a read error in the tool.
Do you have any clue why this happens? Should this component work with my inverter?
Beta Was this translation helpful? Give feedback.
All reactions