Using mcsTemperature to track energy usage

[aiqy65]

Still Having Problems

Still not got things going.

The story so far...

With no ini file mcs descovers all the 1 wire devices and created the counter as a water meter. Only one pair of enrties rather than 2 (one pair for each channel on the counter. Thats not a problem as both channels are connected together.
MCS then puts all the devices into Homeseer.

I then stop everything and change the DN & DC device codes to a D7 & D8 and rename the database fields the same.

Restart and MCS quickly create's a Water meter again.

Very quickly the history value of the water meter fills up to a huge number and hits overflow, stopping any data going into the database. The 'power' consumption stays as a flat line. You can watch the couner value slowly increasing on the raw data from the temp08. Clearing out the history value to zero only last a few mins as it fills up again very quickly.

When midnight came along things reset and I got a couple of mins worth of data before the water meter went off scale again. I have tried to remove the water meter device but it keeps coming back.

With the small amounts of data that have gone into the database I have never had any meaningful graphs out and the numbers do not tie up to the counter value.

Attached is the latest ini file with all the devices and the 'overflowed' water meter.

Who knows what trauma I am going to suffer when I put the main house power (this is only the garage) and rain gauge on !!! Wires and electrons I am OK with. Software is another thing (cant you tell.....)

Any pointer in the right direction would be appreciated.

Cheers
Lee

[Michael McSharry]

There is also a device type field in the .ini file and this is used to distinguish water from watts. If you edit manually did this also needs to be adjusted. Why not just change the device on the mcsTemperature page from a Water Flow to a Wattage counter? This should allow the plugin to make all the ini settings consistent.

If you are using a Temp08 then it only provides a returned value for one of the two counters unless this has changed from the earlier firmware. mcsTemperature will only create 4 devices if it is connected to the DS9097U/DS9490.

What is the rate of change of your counter? Will it often overlfow the 16 bit counter in a relatively short amount of time? I forget if the rate is a moving average or based upon the slope of two counts that are perhaps 3 samples apart. If you have a high counter change over a short period of time then the rate will be high, but the rate should not increase if utilization is steady.

[8r1an]

The newer firmware of the Temp08 from 2.07 reads both channels of the counter when set as lightning

Lightning #01[AE0000000A02E41D]=00207,23588

[Michael McSharry]

I made changes to try to accomodate the two counters. I also saw where the only support for wattage was with the DS9097U/DS9490 so I also tried to add that for the Temp05/08. Give the attachment a try to see if it helps your situation. You should also clear out the bad history so you will not get overflows.

[aiqy65]

Getting there

I have managed (dont ask me how!) to get a nice saw tooth grap out of the box for energy usage. Starting from scratch it counted up nicely from zero and reset at midnight. But as can be seen from the attached graph rather than drop back to zero it has decided that its new base line will be something else and is happy with that !!
Also the wattage rato is a zero and doesnt want to budge.

I did get a couple of overflow errors from the water counters that wont go away but that seems to have settled down now.
Any help appreciated

Thanks
Lee

[Michael McSharry]

Give V5.6.7 a try. It is posted at the top of this subforum

[aiqy65]

Where is the new version hiding?

Either I am looking in the wrong place or the board is hiding it from me somewhere.

Should it be in the first message with the revision history?

Cheers
Lee

[Michael McSharry]

Try again. I probably closed the window before it completed the upload.

[aiqy65]

It stopped at midnight

Started fresh without an ini file or database and recreated everything OK
Got a nice climbing line of "Wattage Today" upto midnight where it went back to zero and has stayed there ever since. Logging is still going on as I am getting data for other 1 wire divices OK. Wattage rate has never moved from zero and the two water counters that appear after I change things to Watts are running up huge numbers.
Thanks for your help

Cheers
Lee

[Michael McSharry]

I do not see anything by looking at the source. Enable the general debug and collect data while the Temp08 is delivering one sample interval of data. Post this and the current mcsTemperature.ini file.

[aiqy65]

For the eyes of the Expert

Log and ini file as requested

Thanks
Lee

[Michael McSharry]

Another attempt is posted. I think you should have what you need.

[aiqy65]

Things are looking good.

Things are looking good.
I am getting a very nice daily usage graph. That 100%
also the average usage is also working 99% but it has a little blip every "update interval" and drops the average down from 180 to 130 and then it goes back to the 180 average. The supply being monitored is steady at the moment so its not a usage variation. I can vary the hit in the average by changing the temp08 update reading in the settings. Currently that part of the 1 wire is now disconnected because of the below.

Also the blip seems to have crept into the temperature readings a few versions ago. It is not regular like the power average and hits all 1 wire temp devices at the same time. I have shrunk the 1 wire system down to 2 temp devices thinking its a problem with the "wires" but even with only 2 sensors it still happens. I have managed to capture the "event in question" Picture and log attached.

Does it give a clue to your expert eye or do you need some more data capturing?

I would also like to put on record for all to see my appreciation for the support you give to this very excellent package. Without it the "script incapable" HS owners like me would be up the creek without a paddle.

Cheers
Lee

[aiqy65]

Power Average Glitches

Continuing on from the previous post I have reconnected the power counter and captured one of the regular downward spikes in what would otherwise be a nice flat line average graph.
Picture, Log file (see time stamp at 12:11.27) and the current ini file attached for inspection.

Cheers
Lee

[aiqy65]

Looks like the Temp08 is having a problem

Now having captures a days worth of raw data from the temp08 I have found a few question marks burried in the data. Would this cause the average graph to drop or does it ignore those readings and continue as if nothing happened?

It also added a sensor and then took it away. That may be the one thats upset the temperature graphs.

I will send a log to the man at Midon and see is he can shed any light on it

Cheers
Lee

THU 07:13:29
Reading Sensors...
Lightning #01[5500000009EB711D]=36798,59450
Temp #01[370000010AE0C428]=26.93C
Temp #02[4A000801419AC910]=20.93C
>
THU 07:14:29
Reading Sensors...
Lightning #01[5500000009EB711D]=36822,59474
Temp #01[5500000009EB711D]= ???
Temp #02[370000010AE0C428]=26.87C
Temp #03[4A000801419AC910]=21.00C
>
THU 07:15:29
Reading Sensors...
Lightning #01[5500000009EB711D]=36840,59492
Temp #01[370000010AE0C428]=26.87C
Temp #02[4A000801419AC910]=21.00C