Beta testing of v2.2

[gskjold]

Attached is a new version with changes outlined below. If anyone could test this on their setup, it would be highly appreciated.

!Warning!
This is a firmware which is still under development. Testing this firmware is on your own risk. Please only test if you have possibility to flash via TTL or USB if anything goes wrong.

!Please note!
If you upgrade to this version and then downgrade, you will loose all your data.

Changes:

• Completely rebuilt user interface
  • Can run offline
  • Does not rely on external resources
  • Faster than before
  • Hopefully more intuitive
• Added visualization of peaks in current month.
• Changed timezone configuration
• If there are HAN read errors, a error message is displayed in GUI
• Added production "income" calculation
• Various bug fixes of realtime calculations
• Energy price fetch without ENTSO-E API key
  • Fetches prices from a centralized server dedicated to this project
  • Works with both ESP32 and ESP8266
  • Note that this is a BETA feature. If it works out ok we will keep it

Please note that tariff thresholds and peaks visualization is now only visible for NO* price regions, as this is not interesting for the rest of the world.

Please let me know when you have tested this new firmware and include the following information:
Country:
Meter:
Hardware:

esp32.zip
esp32s2.zip
esp32solo.zip
esp8266.zip

• Fixed issue where first bar in day/month graph is zero
• Fixed issue with threshold peaks reported on MQTT with 100x value
• Fixed exception on Kamstrup when no load on the meter (0w)

esp32.zip
esp32s2.zip
esp32solo.zip
esp8266.zip

4e97554

• Exposed error from ENTSO-E communication in UI
• Improved HAN and MQTT error display in UI
• Added control to enable/disable auto reboot if not connected to WiFi
• Hidden upgrade link if authentication and not logged in
• Added HAN serial autodetect. Rotates between 2400 8E1, 2400 8N1 and 115200 8N1
• Fixed energyaccounting in config file download
• Fixed dayplot and monthplot in config file download for ESP32
• Fixed broken redirect after setup if user changed hostname

esp32.zip
esp32s2.zip
esp32solo.zip
esp8266.zip

1cee48e

• Increased range on datapoints for day and month graph
• Increased range on main fuse and production capacity
• Added configuration for user interface
• Added realtime month value to MQTT RAW data
• Added display for AP MAC address in status page
• Fixed issue where incorrect prices are published around midnight
• Fixed issue with unit on reactive and power factor in Home Assistant mode
• Fixed issue where uploading configuration file with non-ASCII characters corrupts configuration
  (In the real world, CRLF linechange from windows editors)
• Imroved AP selection in networks with multiple accessspoints
• Some general improvements in data refresh of graphs
• Some general improvements to bar graphs
• Disabled autocomplete on configuration forms to avoid browser replacing configuration values

esp32.zip
esp32s2.zip
esp32solo.zip
esp8266.zip

d6a8d31

• Fixed issue where HAN pin is copied to LED pin in GPIO settings when saved
• Fixed issue where initial configuration of known hardware was corrupted
• Fixed units on more HA sensors
• Some updates after moving repo to new GitHub organization
• Some updates for Landis+Gyr

esp32.zip
esp32s2.zip
esp32solo.zip
esp8266.zip

1227dff

• Fixed issue where garbage was received in cheapest6hr field in MQTT JSON payloads
• Fixed issue with Vcc boot limit on BUS powered ESP8266 devices
• Changed MQTT JSON mode's system-json to be sent every minute instead of just once
• Changed MQTT HA mode's system-json to be sent every minute instead of every 30s
• Changed color coding of voltage graph
• Changed autocomplete behavior on initial setup page
• Added sanity check for all text input in config storage
• Added default voltage 400V and main fuse 40A

esp32.zip
esp32s2.zip
esp32solo.zip
esp8266.zip

[Andreasjkoch]

I am unable to update to this version. Here is the error I am getting:

500: couldn't create file HTTP/1.1 500 Internal Server Error Content-Type: text/html Content-Length: 6349 Connection: close <htm1><body><h1>Upload already in progress!</h1> </body></html>HTTP/1.1 500 Internal Server Error Content-Type: text/html Content-Length: 6349 Connection: close <htm1><body><h1>Unable to write file!</h1></body> </html>

[ArnieO]

Which hardware and processor are you on?

[Andreasjkoch]

ESP32S2 - POW-K+

[ArnieO]

Strange. I have tested several previous test versions against the same hardware, and just retried now without any issue.

Sorry to ask - but since this is so unexpected: You're sure you selected the right file (firmware.bin from esp32s2.zip)?
(If Yes, we'll have to wait for @gskjold to come back and comment, maybe he has an idea.)

[Andreasjkoch]

I am selecting the firmware.bin file - yes. I just attempted the upload again, once again I got an error. However it is a little bit different than the one I mentioned earlier. It might just be because of a different browser but I don't know...

Upload already in progress!
HTTP/1.1 500 Internal Server Error Content-Type: text/plain Content-Length: 430 Connection: close 500: couldn't create fileHTTP/1.1 303 See Other Content-Type: text/html Location: /restart-wait Content-Length: 430 Connection: close HTTP/1.1 200 OK Content-Type: text/html Content-Length: 0 Connection: close

[Andreasjkoch]

Okay, I just tried rebooting it again to try and clear it of its "Upload already in progress!" status. This time it did work and upgraded to the new version...

[gskjold]

We'll write it off as a glitch for now, but it anyone else experience this, please let me know

[tronde-ams]

First impression is very good.

Norway
Aidon 6525
ESP32 (Chinese clone of something).

[tronde-ams]

Seems to be some weirdness in the graphics.

[ArnieO]

I don't see this; the graph looks as it should here:

[tronde-ams]

First bar in last 24h became OK after 01:00. Until 01:00 it was 0 as shown in previous screenshot.

[tronde-ams]

And disappeared again after 02:00. Data er gøy...
The visible bars agree with my two other units with 2.1.14.

[gskjold]

Interesting. I think I need to sit down and compare the code between the old and the new. I think it is mostly the same, but I know I made some modifications.

[ArnieO]

Nå dukket det samme opp her. Har aldri sett noe slikt før (med 2.1 eller tidligere).

[gskjold]

When I check my test-bench Pow-U, it is actually the same there, will look into this :)

[gskjold]

Found the problem :)

[TheNablaOperator]

Hey there.

Yesterday I uploaded the new firmware, it worked & looked fine so far... But I have to check again today.

• Country: Austria
• Meter: Sagemcom (DSMR-P1 5.0.2)
• Hardware: ESP8266 (AZ-Delivery D1 mini)

I am having some issues with MQTT (already with the old firmware), I want to upload to thingspeak, but AMSreader can't connect.
What do I put into "Publish topic"? Sorry, this is my first touch with MQTT ;)

[TheNablaOperator]

@ArnieO I am still trying to connect.... I moved from thingspeak to an own mosquitto broker on a raspi. I can connect via MQTTX-Client from Desktop. But the ams-reade still can't connect.

Just to start searching... Which mqtt version is used? 3.1?

If wanted I move this discussion to an own issue for discussion/help.

[Cheesebaron]

For homeassistant I use homeassistant/sensor/powk as publish topic then it takes that and appends the specific metrics as entities. It ends up looking something like this:

According to Thingspeak docs it looks like they want stuff in a different format: channels/<channelID>/publish/fields/field<fieldnumber>

You might need something in between that serves it as Thingspeak wants it, just like you are trying with your raspberry pi. As for MQTT version, I don't think it matters that much. The mosquitto broker HASS.IO uses is version 6.1.3

[ArnieO]

Agree with @Cheesebaron , MQTT version is not the issue. I recommend using Node-Red as the "something in between" that sets up a message in the format that Thingspeak wants it.

[TheNablaOperator]

@ArnieO , @Cheesebaron thanks for the reply.

Funny thing. Just updated to new beta version (from yesterday).
Works now. Nothing changed in my settings or in the broker :D

edited post.

[gskjold]

The client used in this project is 3.1.1. When you say it works, I assume you are able to connect to a newer version MQTT server? I might need to look into upgrading the client at some point.

[tronde-ams]

Seems like you have changed multiplier for {publish topic}/realtime/import/peak/[1-5] in mqtt raw?

2.1.14 was kW, now something else.
4.31 kWh in 2.1.14 has been 431 in this beta.

[gskjold]

What... That's a mistake, I will look into that!

[gskjold]

• Fixed issue where first bar in day/month graph is zero
• Fixed issue with threshold peaks reported on MQTT with 100x value
• Fixed exception on Kamstrup when no load on the meter (0w)

esp32.zip
esp32s2.zip
esp32solo.zip
esp8266.zip

[gskjold]

Note that the bar chart issue needs a cycle (1hr/1day) to fix itself

[Cheesebaron]

Just installed this, will report if I have any issues. Out of the bat, the user interface is now much more snappy, which is great!

Country: Denmark
Meter: Kamstrup
Hardware: Pow-K (ESP8266)

[Cheesebaron]

Updated to latest, still going strong! Works really well!

[Cheesebaron]

I've been keeping up with all the updates. From my point of view everything has been working very well.

My main use case is integration with Home Assistant, which has been working very well!

[GEAG]

Country: Switzerland
Meter: Kamstrup
Hardware: Pow-K (ESP8266)

Divide by zero exception when load=0 fix confirmed. Thanks

[ArnieO]

Has there been change in the handling of missing whole-hour readings?
I am running a long-term test on a card that is failing (reboots several times per day), and got this. I suspect the reason could be a missed whole-hour payload due to reboot. In firmware v2.1.x such situation was handled by averaging.
Norway / Kamstrup / Pow-K+

[tronde-ams]

Has there been change in the handling of missing whole-hour readings? I am running a long-term test on a card that is failing (reboots several times per day), and got this. I suspect the reason could be a missed whole-hour payload due to reboot. In firmware v2.1.x such situation was handled by averaging. Norway / Kamstrup / Pow-K+

My ESP32 on Aidon behaves as before. I managed to pull the wrong plug so it was without power during whole-hour. First hour it was 0, then it averaged as expected.

[gskjold]

I have actually seen this a couple of times before, have not yet been able to identify the reason...

[TheNablaOperator]

Installed. Seems to be working.
MQTT works now. But phase current and voltage is always 0.00... Anybody else?

[TheNablaOperator]

@ArnieO Thanks. I already have the hex data.

Is this still encrypted? Or the final hex data?

In the wiki blog it says: "If there are issues with interpreting the data, there should be understandable information here." But there is no link...

[ArnieO]

I think it should show data both before and after decryption, please just past the output here (text format).

(And you're right about "...understandable information here", I cannot remember what I was thinking when I wrote that; sorry! The DMLS hex data needs some tooling to interpret.)

[TheNablaOperator]

Yep... I am currently stuck in finding the right entries ;)
Hope with text you mean code format...? Otherwise let me know and I will edit the post.

(D) Serving /data.json over http... (D) Serving /data.json over http... (unwrapData) GCM frame: DB 08 53 41 47 35 00 02 62 5D 82 01 F2 30 00 00 ED A3 2F 45 53 54 35 5C 32 35 33 37 34 30 39 37 36 5F 41 0D 0A 0D 0A 31 2D 33 3A 30 2E 32 2E 38 28 35 30 29 0D 0A 30 2D 30 3A 31 2E 30 2E 30 28 32 32 31 32 31 31 32 30 32 37 34 38 57 29 0D 0A 31 2D 30 3A 31 2E 38 2E 30 28 30 30 32 36 38 36 32 37 33 2A 57 68 29 0D 0A 31 2D 30 3A 31 2E 38 2E 31 28 30 30 32 30 38 33 31 36 30 2A 57 68 29 0D 0A 31 2D 30 3A 31 2E 38 2E 32 28 30 30 30 36 30 33 31 31 33 2A 57 68 29 0D 0A 31 2D 30 3A 31 2E 37 2E 30 28 30 30 30 30 30 30 30 37 36 2A 57 29 0D 0A 31 2D 30 3A 32 2E 38 2E 30 28 30 30 30 30 35 35 31 35 37 2A 57 68 29 0D 0A 31 2D 30 3A 32 2E 38 2E 31 28 30 30 30 30 35 35 30 39 39 2A 57 68 29 0D 0A 31 2D 30 3A 32 2E 38 2E 32 28 30 30 30 30 30 30 30 35 38 2A 57 68 29 0D 0A 31 2D 30 3A 32 2E 37 2E 30 28 30 30 30 30 30 30 30 30 30 2A 57 29 0D 0A 31 2D 30 3A 33 2E 38 2E 30 28 30 30 30 31 33 37 37 32 32 2A 76 61 72 68 29 0D 0A 31 2D 30 3A 33 2E 38 2E 31 28 30 30 30 31 32 39 31 34 34 2A 76 61 72 68 29 0D 0A 31 2D 30 3A 33 2E 38 2E 32 28 30 30 30 30 30 38 35 37 38 2A 76 61 72 68 29 0D 0A 31 2D 30 3A 33 2E 37 2E 30 28 30 30 30 30 30 30 30 31 33 2A 76 61 72 29 0D 0A 31 2D 30 3A 34 2E 38 2E 30 28 30 30 31 31 31 35 38 37 38 2A 76 61 72 68 29 0D 0A 31 2D 30 3A 34 2E 38 2E 31 28 30 30 30 37 33 30 36 32 31 2A 76 61 72 68 29 0D 0A 31 2D 30 3A 34 2E 38 2E 32 28 30 30 30 33 38 35 32 35 37 2A 76 61 72 68 29 0D 0A 31 2D 30 3A 34 2E 37 2E 30 28 30 30 30 30 30 30 30 35 38 2A 76 61 72 29 0D 0A 21 46 32 44 42 0D 0A FC 38 89 B5 66 D3 C7 F3 B4 A6 79 DC (unwrapData) DSMR frame: 2F 45 53 54 35 5C 32 35 33 37 34 30 39 37 36 5F 41 0D 0A 0D 0A 31 2D 33 3A 30 2E 32 2E 38 28 35 30 29 0D 0A 30 2D 30 3A 31 2E 30 2E 30 28 32 32 31 32 31 31 32 30 32 37 34 38 57 29 0D 0A 31 2D 30 3A 31 2E 38 2E 30 28 30 30 32 36 38 36 32 37 33 2A 57 68 29 0D 0A 31 2D 30 3A 31 2E 38 2E 31 28 30 30 32 30 38 33 31 36 30 2A 57 68 29 0D 0A 31 2D 30 3A 31 2E 38 2E 32 28 30 30 30 36 30 33 31 31 33 2A 57 68 29 0D 0A 31 2D 30 3A 31 2E 37 2E 30 28 30 30 30 30 30 30 30 37 36 2A 57 29 0D 0A 31 2D 30 3A 32 2E 38 2E 30 28 30 30 30 30 35 35 31 35 37 2A 57 68 29 0D 0A 31 2D 30 3A 32 2E 38 2E 31 28 30 30 30 30 35 35 30 39 39 2A 57 68 29 0D 0A 31 2D 30 3A 32 2E 38 2E 32 28 30 30 30 30 30 30 30 35 38 2A 57 68 29 0D 0A 31 2D 30 3A 32 2E 37 2E 30 28 30 30 30 30 30 30 30 30 30 2A 57 29 0D 0A 31 2D 30 3A 33 2E 38 2E 30 28 30 30 30 31 33 37 37 32 32 2A 76 61 72 68 29 0D 0A 31 2D 30 3A 33 2E 38 2E 31 28 30 30 30 31 32 39 31 34 34 2A 76 61 72 68 29 0D 0A 31 2D 30 3A 33 2E 38 2E 32 28 30 30 30 30 30 38 35 37 38 2A 76 61 72 68 29 0D 0A 31 2D 30 3A 33 2E 37 2E 30 28 30 30 30 30 30 30 30 31 33 2A 76 61 72 29 0D 0A 31 2D 30 3A 34 2E 38 2E 30 28 30 30 31 31 31 35 38 37 38 2A 76 61 72 68 29 0D 0A 31 2D 30 3A 34 2E 38 2E 31 28 30 30 30 37 33 30 36 32 31 2A 76 61 72 68 29 0D 0A 31 2D 30 3A 34 2E 38 2E 32 28 30 30 30 33 38 35 32 35 37 2A 76 61 72 68 29 0D 0A 31 2D 30 3A 34 2E 37 2E 30 28 30 30 30 30 30 30 30 35 38 2A 76 61 72 29 0D 0A 21 46 32 44 42 0D 0A (readHanPort) Received val

[gskjold]

The DSMR frame above (starting with 2F) you can drop straight into a hex to ascii converter

[TheNablaOperator]

Seems that the meter only sends Power and Energy only. Thanks for the help!

[tronde-ams]

Can you check monthly accumulated price?
I have a feeling it is very low.

My two other units (FW 2.1.14) shows the same accumulated energy (643 kWh) and 2052 NOK, and we know they only counts 23h. This makes more sense when the average price until now in NO1 is 3.45 NOK ex. VAT. I have 1 as multiplier for price on all units.

[gskjold]

OK, I'll keep an eye on mine and see what happens

[tronde-ams]

I can't see it, but Is this value hidden in the saved config file? I have saved config from 2.12 to first beta and from first to to second beta.

[gskjold]

Yes, realtime data is stored like this:
energyaccounting [version] [month] 0.00 [costYesterday] [costThisMonth] [costLastMonth] [peaks...day and value]

[tronde-ams]

Same time. I can't figure out where to find 1662,70?

Edit:
Short up-time is because me, not the interface.

[gskjold]

1377 is the number you are looking for. It is only updated once per day. It adds day cost when it is passing midnight.

And I forgot I changed the format a bit in 2.2

[tronde-ams]

Something became lost when I updated from 2.1.12?

from 2.1.12 to 8a809ec

from 8a809ec to 6c3ddc5

[gskjold]

You are absolutely right, there is a problem with energyaccounting in the export in v2.2 where some data is lost. Will fix this for next patch

[TheNablaOperator]

Hey guys.

A suggestion... I am using the AMS reader on a Sagemcom with open collector output. Would it be possible to add in the hardware settings a check box to enable the internal pull-up resistor of the used communication port?

Regards from Austria!

[ArnieO]

The internal pullup resistor is too weak for that. They are of (as far as I know) undocumented value - but around 30 kohm. For a reliable pullup strength from meter with OC port you need a pullup of around 3 kohm. Some meters brands may work with weaker pullups, but I do not want to encourage using the internal pullups by implementing it in the code.

[ArnieO]

To be more specific: You need to pull around 1 mA. If you pull to 5V, use around 4k7. If you pull to 3,3V, use around 3k.
If you pull to 5V, use a buffer that translates to 0/3,3V as interface to the ESP.

[TheNablaOperator]

hmmm ok... Currently I pull with 10k to 3v3, which works so far fine... I agree 100k doesn't work. I don't know if it's working with internal pull-ups or not...

I agree, if one pulls to 5v, a level converter would be necessary!

[ArnieO]

We discovered during beta testing of Pow-P1 that pulling to 3.3V with 10k works only on some meter brands. So we decreased it to around 3k. Even if 10k works for you, I think you will improve your design margins if you select a smaller resistor - and the extra current consumption is really marginal.

[TheNablaOperator]

aah ok thanks for that. I'll give it a try. It's currently a breadboard design. ;)

[gskjold]

Another update:

• Exposed error from ENTSO-E communication in UI
• Improved HAN and MQTT error display in UI
• Added control to enable/disable auto reboot if not connected to WiFi
• Hidden upgrade link if authentication and not logged in
• Added HAN serial autodetect. Rotates between 2400 8E1, 2400 8N1 and 115200 8N1
• Fixed energyaccounting in config file download
• Fixed dayplot and monthplot in config file download for ESP32
• Fixed broken redirect after setup if user changed hostname

esp32.zip
esp32s2.zip
esp32solo.zip
esp8266.zip

[ArnieO]

Ouch...
Tested the same on a Pow-U now; it restarted as it should. No issues.

[Ontrackx]

I got it working again. Will retry tomorrow

[Ontrackx]

Mornig gents/gentleladies, it started up with 2.1.16 after resetting. All the config and history gone. Ive only had the device for 4 days so not much history to loose anyways. But is it normal behavioure that from 15 to 16 we need to reconfigure? or is it just "bad luck"

[ArnieO]

Since you're in the discussion page for beta testing of v2.2: Did you test it in between? If so loosing history is as expected.
Upgrade from 2.1.15 to 2.1.16 should not cause lost history or need for reconfiguration.
You mentioned "Will try and whipe and reinstall tomorrow". Wiping means you need to reconfigure.

[Ontrackx]

Yes, i whiped it . To get it to boot again. It no problem, ive learned some stuff :), nice job on the firmware 2.1.16 by the way. Im pretty new to the esp world so trying to learn as much as possible. Thank you for the help

[dbeinder]

Is there a public development branch?

I'm currently tinkering with my own hardware - my Austrian Kaifa meter provides more than 100mA at 27V on MBUS, so I'm working on a ESP32-S2 vampire hardware that doesn't have to worry about depleting a supercap. So far, the only change I needed was to disable the pullup on the HAN pin, since my level shifter design is pretty high impedance. Decoding worked right away, thank you for this great piece of software!

[gskjold]

Currently I have not yet shared this publicly, but will be pushed in the not too far future :)

[GEAG]

Great work. Some observations with v2.2

Country: Switzerland
Meter: L&G E450
HW: Pow-U (ESP8266)
Installed version: 4e97554

1. The HAN icon (top of UI) sometimes turns YELLOW or RED when there is no load on the SM. If the load is non zero, the HAN icon stays green. There is a verbose debug message: "No HAN data received last 30s, single blink" despite that power data (load = 0) is received correctly and displayed in verbose debug mode.
2. This SM pushs power measurements with 1/5 Hz and energy with 1/60 Hz. The current import power displayed in the webUI drops to zero once every minute.
3. If the SM does not deliver reactive power or energy, the respective block in the webUI should better not be shown.
4. The reactive energy counters delivered by the SM at 1/60 Hz are not yet interpreted and displayed.
5. It would be nice, if the SM's serial number (OBIS 96.1.1) would also be displayed under the info-display of the webUI
6. The "future energy price" for Switzerland displayed in the webUI does not match the price information displayed in graph by transparency.entsoe.eu for Switzerland

[gskjold]

Thanks for the feedback!
1: I cannot see any reason why HAN would turn red if the load is 0. Could you send me a debug frame of the data received?
4: If this is present in the data received, I would very much like a debug of this as well :)
5: 96.1.1 should be visible under /status page
6: Do you have a screenshot of this or other form of comparison? Code is using area code 10YCH-SWISSGRIDZ, which I think is correct

[ArnieO]

Thank you @GEAG; great feedback.

1. No load is an unusual situation, so a potentially undiscovered bug.
2. Strange bug, never seen or heard that before. Reporting interval ranges from 1/2.5 to 1/10 Hz, so yours is not extreme. For how long does it display zero? Could you make a video of your screen and email to [email protected]?
3. Maybe a better option is an "opt-out" from reactive energy - which is meaningless and unimportant to most users.
4. Sounds like a bug. Could you post a "whole hour" debug frame to see what data you receive?
5. I believe the serial number of my meter is the ID? Do you not have it presented?
   
6. Apparently a bug, thank you for reporting it!

[gskjold]

Happy new year everyone! Here is yet another update, maybe we are getting close to the end?

• Increased range on datapoints for day and month graph
• Increased range on main fuse and production capacity
• Added configuration for user interface
• Added realtime month value to MQTT RAW data
• Added display for AP MAC address in status page
• Fixed issue where incorrect prices are published around midnight
• Fixed issue with unit on reactive and power factor in Home Assistant mode
• Fixed issue where uploading configuration file with non-ASCII characters corrupts configuration
  (In the real world, CRLF linechange from windows editors)
• Imroved AP selection in networks with multiple accessspoints
• Some general improvements in data refresh of graphs
• Some general improvements to bar graphs
• Disabled autocomplete on configuration forms to avoid browser replacing configuration values

esp32.zip
esp32s2.zip
esp32solo.zip
esp8266.zip

[dbeinder]

Hardware: ESP32-S2 "Wemos S2 mini"
Setting the HAN GPIO pin seems to also change the LED pin to the same GPIO. In the settings page at least, the signal input works. Maybe just a display issue of the settings page?

[gskjold]

Nice catch! It does not actually change, but the LED field is populated with HAN pin on return

[ElMyggo]

Hardware ESP8266 D1 Mini
Cant save L3 idx value in Domoticz, shows only 0
but if i export config file and change manually then import its show correct idx value

[gskjold]

Confirmed, thanks. Bit of a copy/paste error :) Will fix

[tronde-ams]

1cee48e
Aidon 3p IT Norway
ESP32

When I change settings for MQTT the HAN port freeze and turns red. A reboot solves this.
I would have expected no problem, or an automatic reboot?

[tronde-ams]

UART2 and LED GPIO overlaps :) I think it is because of a issue in the earlier 2.2 builds that interchanged the values in those two fields.

Is this something that does not update the values shown?
HAN = UART2 = GPIO16, and LED is 2 and 15.
I have the same setting on my 2.1.16 which has never been 2.2.

[gskjold]

In your screenshot LED=16, Red=2, Green=15. Set the LED pin blank and you should be good.

[tronde-ams]

In your screenshot LED=16, Red=2, Green=15. Set the LED pin blank and you should be good.

I see now.

[tronde-ams]

In your screenshot LED=16, Red=2, Green=15. Set the LED pin blank and you should be good.

I set the pin blank and save, but it does still show 16 even after re-boot. ???

[gskjold]

Ah, you are correct. This is actually a bug in the current testing version, sorry about that. I have fixed it, and just assumed this was before I published the latest update..

[tronde-ams]

Same problem if I change other settings as well.

[gskjold]

Another update

• Fixed issue where HAN pin is copied to LED pin in GPIO settings when saved
• Fixed issue where initial configuration of known hardware was corrupted
• Fixed units on more HA sensors
• Some updates after moving repo to new GitHub organization
• Some updates for Landis+Gyr

esp32.zip
esp32s2.zip
esp32solo.zip
esp8266.zip

[gskjold]

Final update! Aiming for release this weekend

• Fixed issue where garbage was received in cheapest6hr field in MQTT JSON payloads
• Fixed issue with Vcc boot limit on BUS powered ESP8266 devices
• Changed MQTT JSON mode's system-json to be sent every minute instead of just once
• Changed MQTT HA mode's system-json to be sent every minute instead of every 30s
• Changed color coding of voltage graph
• Changed autocomplete behavior on initial setup page
• Added sanity check for all text input in config storage
• Added default voltage 400V and main fuse 40A

esp32.zip
esp32s2.zip
esp32solo.zip
esp8266.zip

[tronde-ams]

I have lost track...
Will the final 2.2 require a specific 2.x to upgrade from?

[gskjold]

v2.1 og v2.2-beta builds are all compatible with the release

[ArnieO]

A big Thank you to all contributors to this firmware upgrade!

An updated user manual is now available for download on the product pages in our webshop https://amsleser.no
Direct download link: https://amsleser.no/index.php?controller=attachment&id_attachment=45

If anyone see improvement points I'd love to hear from you!