SDM630 Total harmonics percentage - default values look too high, possibly need to be scaled down

[Laarge]

Describe the bug

I wanted to experiment with the capabilities of my SDM630 and check if any of the fridges introduces harmonics etc. I was however puzzled by the default results I got.

THD (Total harmonics values) reported by the Eastron SDM630 seem unreasonably high.
I am getting constant 60-70%+ values.
According to ChatGPT, I have to divide them by 10, but I wonder if this factor should be 100 maybe?

To Reproduce
Steps to reproduce the behavior:

1. Go to Devices, Enable the SDM plugin, Select device SDM630 and get it working.
2. Click on Output configuration
3. Scroll down to select 3
   Output Configuration , select the 4 values below:
   THD (%) L1
   THD (%) L2
   THD (%) L3
   Average line THD (%)

4. See the resulting valies, either on the Devices page, or in MQTT, Grafana.

5. Then apply the suggested by ChatGPT 10x scaling down of the values "%value%/10"
   And see the reported harmonics scale go down.

But I am still not convinced that 10X is the correct factor.
And the offifial datasheet says that eastron already reports them in %.

Page 4 https://www.eastroneurope.com/images/uploads/products/protocol/SDM630_MODBUS_Protocol.pdf

HEre are the Addresses (Registers), as per the datasheet
30241 121 Phase 1 Current THD% 00 F0
30243 122 Phase 2 Current THD % 00
30245 123 Phase 3 Current THD % 00 F4
30341 171 Average line to line volts THD. ％ 01 54

Expected behavior
My expectation is that the values need to be scaled down, by some unknown factor and the default behaviour misleads the user.

Screenshots

Used platform (please complete the following information):

• ESP type: ESP32
• Build version: mega-20241222
• Build set energy,
  Platform Specifics (when applicable, please complete the following information):
• Flash size: [e.g. 4M]
• Brand/Model: NodeMCU,
• Browser + OS Brave

Additional context
I am not that savvy, nor that I have other equipment to cross-check, but 60% values look odd to me, therefore I want the issue to be addressed.

[tonhuisman]

A first remark: You haven't provided a usable title for this issue, can you please provide a fitting description? (Use the Edit button, next to the title)

We can discuss the working of the plugin, but not the data provided by the device connected. As that's whats happening here, the plugin reads the register(s) of the SDMxxx, and provides the values as is, without further processing. So if the device documentation states the result should be a percentage, then that's what it is, a percentage.

NB: I can't verify your measurements, as I don't have access to a SDM630, and I also don't know what a reasonable result for the THD percentages would be, but please be aware that the results that you are seeing can be correct, as it's depending on the signal provided.
Have you tried to use the L to N THD (%) L1 .. L3 values? I don't know what values to expect there either, but maybe they are more in line with your expectation?

NB2: The register for the Average Line THD (%) you have selected in your screenshots is actually 00 FA, register 01 54 provides Average Line to Line THD (%) (Description is nearly the same)

[Laarge]

A first remark: You haven't provided a usable title for this issue, can you please provide a fitting description? (Use the Edit button, next to the title)

We can discuss the working of the plugin, but not the data provided by the device connected. As that's whats happening here, the plugin reads the register(s) of the SDMxxx, and provides the values as is, without further processing. So if the device documentation states the result should be a percentage, then that's what it is, a percentage.

NB: I can't verify your measurements, as I don't have access to a SDM630, and I also don't know what a reasonable result for the THD percentages would be, but please be aware that the results that you are seeing can be correct, as it's depending on the signal provided. Have you tried to use the L to N THD (%) L1 .. L3 values? I don't know what values to expect there either, but maybe they are more in line with your expectation?

NB2: The register for the Average Line THD (%) you have selected in your screenshots is actually 00 FA, register 01 54 provides Average Line to Line THD (%) (Description is nearly the same)

You might be right, I just do not know what I am looking at.
Just tried the other values and I am getting much smaller percentages, raw values, no scaling down required.

I will look into it for a few days, in the meantime will try to delete this issue not to mislead anyone else (presumably the AIs, crawling around :) )

[TD-er]

If a conversion is needed, then I would expect indeed maybe a factor of 10 or 100 could make sense.
But only when the reported values are int-types.
If the values are float, then I see no reason why they need a correction.

Right now I am not in the position to check the plugin code, nor the datasheet as the internet connection today is extremely poor (hotel in Bhutan) but when I get back home, I can have a look.

[TD-er]

Oh and if the discussion leads to a conclusion, please do not delete the issue.
It can be closed if resolved, but I find constructive discussions way more valuable than taking into account what some AI may 'think' of it.
AI bots are one of the reasons our forum gets DDoS'ed every now and then and I personally hate the way how they tend to make people stop thinking and just accept what an AI bot returns.