mcsMQTT 5.20.0.1 [HS3 & HS4] Sense Energy , GW100 derived sensors, Concurrency Mgmt

[Michael McSharry]

Sense Energy has been integrated in the Cloud page, Sense tab of mcsMQTT. It provides the ability to update HS devices at a user-specified rate form the Sense Energy Cloud server and to deliver graphics for historical trends. To use this feature a user needs to have Sense Energy hardware and an account with Sense Energy that is setup when the hardware is registered. The plugin also provides access to device information, device isolation status, and timeline of device changes. These additional features are not automatically available as HS devices, but are viewable for association to HS via the MQTT Page, Association tab. It also provides a conduit to know what device, status and timeline information is available from the server so future plugin enhancements can be made.



The energy consumers are presented as HS Devices



All Sense Energy information is available on the mcsMQTT Assocaition tab of the MQTT Page where it can be associated with HS deivce, if desired.



If the current energy use of a sense consumer (i.e. payload column hyperlink) is clicked then a new page is drawn with the historical energy use. The periods of time can be selected for additional views from this page. This is data from the Sense server and not collected in mcsMQTT databases. It is somewhat awkward with very little information at the beginning of the reporting cycle. The chart is presented in local time while the Snese server collects data in the UTC time representation.



This is relatively early in the Sense Energy integration so quite welcome to changes to improve the usability.

The GW1000 interface has been augmented with up to two derived sensors for dewpoint and feels like if the sensors are available to support these derived calculation.

A more significant change under the hood is the method by which concurrent access to the same data is handled. I have been running this update for a few days without issue. It seems some things may take longer such as deletion of obsolete topics. This is due to the lockout mechanisms employed to assure a dictionary of data is not being changed when it is being used.

This is version 5.20.0.1 which is an incremental change from 5.191.0 that is available from the HS Updater. The HS4 zip file has an html file that replaces the one in the HS subfolder \html\mcsMQTT. Both have the replacements for the \bin\mcsMQTT folder

HS3: http://mcsSprinklers.com/MCSMQTT_52001.zip
HS4: http://mcsSprinklers.com/MCSMQTTHS4_52001.zip

[jindala]

Thanks a lot Michael. I installed the newer version. The only thing is that I clicked on the payload and I am not able to see any chart, it says no data available.

[Michael McSharry]

Emaill the mcsMQTT Debug.txt file, assuming you have debug enabled on the MQTT Page, General tab. Note in the early morning hours on the Day feed there is no data from the Sense server. The Year or Month ones should be fine from a data perspective.

[Wade]

Michael,

With the latest update (downloaded this morning), I'm seeing Dew Point but not Feels Like derived sensor from the GW1100. As part of troubleshooting, I unassociated the topics so devices were removed, and restarted the plugin. On re-associating the topics the devices were recreated, but without status/graphics config (no icons). Any suggestions? Thank you.

[Michael McSharry]

If you want the want the plugin to create the icons it will need to be on the "first observation" of the sensor. To create this situation use the "O"bsolete rather than the "A"ssociate checkbox. When I was testing I usually removed all features with the "O" on the parent device topic, but it can be done individually. There is a Delete button at the top of the column when a "O" checkbox is checked.

A slice of the code related to the derived sensors is below. It looks like you do not have windspeedmph being reported. What is the key you get for windspeed?

[Wade]

If you want the want the plugin to create the icons it will need to be on the "first observation" of the sensor. To create this situation use the "O"bsolete rather than the "A"ssociate checkbox. When I was testing I usually removed all features with the "O" on the parent device topic, but it can be done individually. There is a Delete button at the top of the column when a "O" checkbox is checked.

Thanks for this guidance.

A slice of the code related to the derived sensors is below. It looks like you do not have windspeedmph being reported. What is the key you get for windspeed?

Ah, it makes sense now. I don't have a windspeed gauge. Would it be feasible to include the Heat Index and Wind Chill derived devices separately to account for varying equipment configurations?

[Michael McSharry]

I do not see a reason to include windchill since it is the feelslike when wind sensor is available. I did include the heatindex when windspeed sensor is not available. I did not test, but seemed like a pretty simple change.

HS3:http://mcsSprinklers.com/MCSMQTT_52003.zip
HS4:http://mcsSprinklers.com/MCSMQTTHS4_52003.zip

[Wade]

I do not see a reason to include windchill since it is the feelslike when wind sensor is available. I did include the heatindex when windspeed sensor is not available. I did not test, but seemed like a pretty simple change.

HS3:http://mcsSprinklers.com/MCSMQTT_52003.zip
HS4:http://mcsSprinklers.com/MCSMQTTHS4_52003.zip

Excellent. Thanks.

[Wade]

I do not see a reason to include windchill since it is the feelslike when wind sensor is available. I did include the heatindex when windspeed sensor is not available. I did not test, but seemed like a pretty simple change.

HS3:http://mcsSprinklers.com/MCSMQTT_52003.zip
HS4:http://mcsSprinklers.com/MCSMQTTHS4_52003.zip

Heat index feature was created, but calculation seems amiss.

tempf = 90.3
humidity = 48
heat index = 75.2

[Michael McSharry]

When I plug in your numbers with the debugger I get 94.3 for heat index. The same temp and humidity values are passed to the dew point calculation. If you continue to get wrong result then I can add some debug if you tell me if HS3 or HS4 plugin is being used.

[Wade]

When I plug in your numbers with the debugger I get 94.3 for heat index. The same temp and humidity values are passed to the dew point calculation. If you continue to get wrong result then I can add some debug if you tell me if HS3 or HS4 plugin is being used.

I'm using the HS4 plugin. Watching the associations page, the heat index is updating regularly. Temperature and humidity readings are correct for current conditions.

[Michael McSharry]

Could it be that you have an Expression function on the Edit tab for HeatIndex?

The attached has debug going to the HS log that shows the parameters during the HeatIndex calc and then shows the JSON that becomes the payload as a simulated MQTT message.

[Wade]

edit: As I was about to post this I realized my Outdoor Temp (tempf) and Heat Index (heatindex) have the same device ref 6466. This may be the problem as it can't be good? If not, please read on below.


---------------------


Michael, apologies if I'm just being thick but having problems getting the debug in the HS log. I'm attaching the mcsMQTT debug. Does it give you what you need?

[ATTACH]n1496961[/ATTACH]


Here's what I tried.
1. Shut down the PI
2. Copied the 2 new files into ...HomeSeer HS4/bin/mcsMQTT
3. Restarted the PI
----No debug in the HS log
4. Turned on Enable General Debug in the PI
---No debug in the HS log
5. Shut down HS4
6. Redownloaded and recopied the 2 files into .../bin/mcsMQTT
7. Restarted HS4
---No debug in the HS log

Here's what was logged from mcsMQTT from the 2 tries above.


BTW, nothing in the expression field.

[Michael McSharry]

I do not know why the HS Log debug was not showing up. The normal debug for the last reading is shown below does show the issue. Inside rather than outside sensors were being used. This is fixed in 5.20.0.4

HS4: http://mcsSprinklers.com/MCSMQTTHS4_52004.zip
HS3: http://mcsSprinklers.com/MCSMQTT_52004.zip

9/18/2021 9:17:07 AM 599785 | ActOnMessageForTrigger QueueSize=0 ,Topic 16E2C6E430C718E745B8BE46AF12DE1D,Payload
{"stationtype":"GW1100B_V2.0.3","dateutc":"2021-09-18 14:16:06",[B][COLOR=#c0392b]"tempinf":76.1[/COLOR][/B],[COLOR=#c0392b]"[B]humidityin":43[/B][/COLOR],
"baromrelin":29.353,"baromabsin":29.353,[B]"tempf":79.7,"humidity":82[/B],"rainratein":0.000,"eventrainin":0.000,
"hourlyrainin":0.000,"dailyrainin":0.000,"weeklyrainin":0.15 0,"monthlyrainin":2.756,"yearlyrainin":49.858,
"totalrainin":49.858,"temp1f":75.7,"humidity1":50,"soilmoist ure1":29,"soilmoisture2":56,"wh40batt":1.6,"wh26batt":0,
"batt1":0,"soilbatt1":1.5,"soilbatt2":1.8,"freq":"915M","model":"GW1100B","dewpoint":59.8,[COLOR=#c0392b]"[B]heatindex":75.4[/B][/COLOR]}

[Wade]

I do not know why the HS Log debug was not showing up. The normal debug for the last reading is shown below does show the issue. Inside rather than outside sensors were being used. This is fixed in 5.20.0.4

HS4: http://mcsSprinklers.com/MCSMQTTHS4_52004.zip
HS3: http://mcsSprinklers.com/MCSMQTT_52004.zip

9/18/2021 9:17:07 AM 599785 | ActOnMessageForTrigger QueueSize=0 ,Topic 16E2C6E430C718E745B8BE46AF12DE1D,Payload
{"stationtype":"GW1100B_V2.0.3","dateutc":"2021-09-18 14:16:06",[B][COLOR=#c0392b]"tempinf":76.1[/COLOR][/B],[COLOR=#c0392b]"[B]humidityin":43[/B][/COLOR],
"baromrelin":29.353,"baromabsin":29.353,[B]"tempf":79.7,"humidity":82[/B],"rainratein":0.000,"eventrainin":0.000,
"hourlyrainin":0.000,"dailyrainin":0.000,"weeklyrainin":0.15 0,"monthlyrainin":2.756,"yearlyrainin":49.858,
"totalrainin":49.858,"temp1f":75.7,"humidity1":50,"soilmoist ure1":29,"soilmoisture2":56,"wh40batt":1.6,"wh26batt":0,
"batt1":0,"soilbatt1":1.5,"soilbatt2":1.8,"freq":"915M","model":"GW1100B","dewpoint":59.8,[COLOR=#c0392b]"[B]heatindex":75.4[/B][/COLOR]}

Appears to be working as expected now. Device reference duplication is also fixed. Thanks again for implementing these features.