Author Topic: TED Predicting Power Meter Readings  (Read 4427 times)


  • Newbie
  • *
  • Posts: 6
  • Karma: +0/-0
TED Predicting Power Meter Readings
« on: June 07, 2011, 08:50:48 PM »
I would like to develop an application the predict my power provider's meter values. The meter out on the wall has two readings; power from the grid and power to the grid. So I would like to obtain the power I take from the grid and the power I return to the grid.
For example, reading the power company's meter for the day, I get 16KWhs coming in and 24Kwhs going back to the grid. The power company figures my grid usage by subtracting the “goes out” from the “comes in”. In this case the day's usage is -8Kwh. Now for the same day I read the TED5000s. The TED attached to the power feed in the junction box reads 23 and the TED5000 attached to the solar feed in the main breaker box reads 39. Now how do I relate the TED5000s to the Power Company Meter? Note that the reading on the TED5000 was from 12 to 12PM where the power company meter is read around 7:30 AM so there are some differences in the readings.

How do I use the TED5000 readings to estimate the power company's meter?

Data From the TED5000s
Let me define GRID as the TED5000 attached to the main feed in the breaker box. I take it that this TED will read the power going both ways. So I take this to mean the absolute value of Grid IN and OUT. This should include Grid usage plus Solar that was not consumed in the house.
I'll define SOLAR as the reading at the feed from the solar inverter. Since the TED5000 reads current both in and out, this reading includes both the solar power and the consumption of the inverter. I find that my inverter consumes about 4watts. This checks out with the solar inverter's read out; less the inverter's consumption.

My reading from the Power Company meter is 16 from the grid and 24 to the grid. These readings I take to be unidirectional. So with the solar reading of 39 this indicates
my home use to be
                        16  - from the grid actual use from the grid
                        39  - solar measured at the solar inverter to house and rest to grid
                       -24 going to the grid
                         31  Home use based on the Power company meter
                                     (16 from the grid and 15 from the solar)
 I believe this number; making the home usage to be 31Kwh

So how do I relate this to just the TED5000 readings?
The TED5000 gives both the grid usage and the return from the solar. Since this reading is bi-directional I can't separate the directional values. You just can't say that the usage is the Grid – Solar because the there is some double accounting here. The actual net value to (or from) the grid is the grid in plus the power left over from the home consumption that goes out to the grid.  So in this case TED grid reading of  23  subtracting the solar reading of 39 = -16 does not reflect what I return to the grid.

So just how do I calculate the actual to and from the grid as the power company sees it?


  • Full Member
  • ***
  • Posts: 168
  • Karma: +0/-0
  • Sherlock
Re: TED Predicting Power Meter Readings
« Reply #1 on: June 13, 2011, 06:53:05 PM »
Depending on configuration your net is either actually measured or calculated. Looks like in your case your net is measured so you already have that. 

23 is your measured net return (some difference probably due to sampling timing difference between ted and utility meter)


net -23
production -39
your calculated load is then 16ish which agrees with what your meter is measuring. 

In our case we're measuring actual production. Your generation meter is only accounting for generation sent to the grid not total generation produced by your solar system.


  • Newbie
  • *
  • Posts: 6
  • Karma: +0/-0
Re: TED Predicting Power Meter Readings
« Reply #2 on: June 13, 2011, 08:16:22 PM »
Thanks. I have been snooping around and looking at the xml and the csv files I see the negative flow on my TED5000 attached to the grid. Looking at the  api/LiveData for the grid, I do see negative values; that's what I expect to see. By calculating using the hourlyhistory I come close to the Power Company readings. Where it really gets confusing, is looking at the MTU used on the grid in the Footprints display. I never see the grid going negative. This just doesn't agree with the api/LiveData. This is the source of my confusion. I just can't see the correlation between the graph for the grid MPU in Footprints. Can I have the setup wrong?