My smart meter data

[gar]

130104-1749 EST

Finally three months after my smart meter was installed I can now go to the DTE web site and get data from my meter. Hourly energy use from the prior day does not appear until about 3 PM. DTE has been collecting data from my meter since about 2 days after the meter was installed. Thus, all of that data is viewable.

Accessing the data is somewhat awkward, but can be useful. There is no voltage data. I have reactivated monitoring with my TED system and will do a comparison. From what I see and knowing my usage I think there will be a good correlation, but maybe an error in absolute calibration. The data can be exported, but in small batches.


There are some solar people in the area that are trying to find out if they can get data directly from the meter rather than via the website.

.

 

[Haji]

gar:
I am curious to know about the energy performance of SMPS power supply, if you have one in your residence. I want to know if its efficiency drops with decrease in its output ( i.e whether its power factor deteriorates). Thanks.

 

[gar]

130105-1746 EST

Haji:

Tested an old, maybe 30 years, PC power supply rated at 63.5 W 110 VAC.

I loaded only the 5 V output and my input was about 123 V.

6.5 W 0.39 PF no load
18 W 0.51 PF
33 W 0.56 PF
48 W 0.60 PF
61 W 0.62 PF

May not be typical for a present day PC power supply.

In the USA at Home Depot you can get a Kill-A-Watt EZ P4460 meter for about $30.

.

 

[mivey]

There are some solar people in the area that are trying to find out if they can get data directly from the meter rather than via the website.

Yes. The simple way is for the utility to provide pulses from the meter. With proper protection for the meter, it probably will set them back a few hundred but they may offer it as a service or at a reduced cost. You would be responsible for getting the cable to the meter location. You still have to wet the relay and have the software to interpret the pulses.

Without the utility pulses, you could attach a device to the meter to read pulse flashes but it will probably set you back about $500 or so before you include the other software and hardware needed to interpret the signals and get the signals back to you. Still would require utility intervention or they will think the device is some type of theft tool.

I doubt the utility is going to give you security access to access the meter directly through wireless or whatever they are using.

Perhaps the better alternative is to put in your own meter that you can communicate with directly.

 

[Hv&Lv]

Yes. The simple way is for the utility to provide pulses from the meter. With proper protection for the meter, it probably will set them back a few hundred but they may offer it as a service or at a reduced cost. You would be responsible for getting the cable to the meter location. You still have to wet the relay and have the software to interpret the pulses.

Without the utility pulses, you could attach a device to the meter to read pulse flashes but it will probably set you back about $500 or so before you include the other software and hardware needed to interpret the signals and get the signals back to you. Still would require utility intervention or they will think the device is some type of theft tool.

I doubt the utility is going to give you security access to access the meter directly through wireless or whatever they are using.

Perhaps the better alternative is to put in your own meter that you can communicate with directly.

I believe the power cost monitor reads pulses on order to interperet usage. One problem with that is that our meters (iTron with TWACS modules) only pulse for the first 24 hours or something like that when installed, and I assume after an outage.
We sold and rented these devices for awhile before we changed our meters out to the new smart meters. Imagine our surprise when we found out our rental program had to shut down.

 

[hillbilly1]

Yes. The simple way is for the utility to provide pulses from the meter. With proper protection for the meter, it probably will set them back a few hundred but they may offer it as a service or at a reduced cost. You would be responsible for getting the cable to the meter location. You still have to wet the relay and have the software to interpret the pulses.

Without the utility pulses, you could attach a device to the meter to read pulse flashes but it will probably set you back about $500 or so before you include the other software and hardware needed to interpret the signals and get the signals back to you. Still would require utility intervention or they will think the device is some type of theft tool.

I doubt the utility is going to give you security access to access the meter directly through wireless or whatever they are using.

Perhaps the better alternative is to put in your own meter that you can communicate with directly.

I think pulse meters are going the way of the dinosaur, most of the energy management systems we install now want way more information than what a pulse meter can give, but then again, they are using this information to tailor the load control for maximum savings.

 

[Hv&Lv]

I think pulse meters are going the way of the dinosaur, most of the energy management systems we install now want way more information than what a pulse meter can give, but then again, they are using this information to tailor the load control for maximum savings.

Ours will read demand, voltage, kWh...
We read all our meters everynight and collect the peak demand for each day, along with the kWh.

we can even tell when you leave home for an extended time...:lol:

 

[gar]

130105-2129 EST

mivey:

The people I am referring to are a group calling themselves solar ypsi. See solar.ypsi.com .

They have several installations around Ypsilanti, MI. One installation at the coop has a totally separate meter, not thru the main panel, for the solar connection. I believe this is the typical method for DTE on their solar currents program sort of mandated by the state. The solar meter at the coop is an electro-mechanical with pulse output. Their online display from this location has an incorrect scaling for energy because the person that wrote the program did not think. I believe the scaling should read kWh per 5 minutes, but that is from memory which may be wrong. It really should have been scaled as power. The data is quantized to 5 minutes, but drawn as a connected points plot. Thus, implying power.

This group wants to get direct RF data from smart meters in real time to avoid having to use an additional system like TED to get the data.

One of the individuals in the group has a home in New Mexico with solar and he is remotely monitoring it with a TED 5000. I have been helping him get it setup to provide useful data because the TED people were unresponsive. His system backfeeds a breaker on his main panel. The solar part is easy to monitor. But the main panel has four hot wires feeding the main breaker. Each phase has two wires in parallel. The TED current transformer won't fit around two wires. Thus, only one wire of each phase is being monitored. Originally he was using the normal calibration (100%, calibration scaling of 1.0). Since the main panel input power has to have the solar subtracted from it to get the house load we need correct data from the main current transformers. This I got him to roughly correct by changing the scaling to 2.0. But observation of the data indicates this is probably not quite correct, and I really don't expect it should.

I have told him how to experimentally adjust this scaling to get correct house load power data. That has to wait until his next trip to NM.

As presently being plotted the graph does not show energy flow to and from the grid. Some time past this was shown and the house load was not.

His data can be seen at http://pvoutput.org/intraday.jsp?id=15240&sid=13106 . For today, 1-5-2013, it shows a beautiful clear sunny day result. On PVOutput you can look at many different participants around the world, and for a very large number of days.

.

 

[gar]

130105-2215 EST

I have no data from my smart meter for yesterday, Friday.

DTE is contracting to some outside organization to process the meter data and provide the graphs that I can access on the DTE web site.

Yesterday, Friday, I got data on the total usage for Thursday early in the morning, but not hourly data. The hourly data did not show up until late afternoon. Since today, Saturday, no data of any kind has appeared for yesterday I suspect that there may be some manual intervention in the data processing, or transfer of data from DTE to this other organization. It doesn't look totally automatic. It is also quite possible the data is not being provided to everyone yet, and the reason for the three months delay before I could access data.

This system is not expected to provide real time data, because normally data is only transferred once per day from the meter.

For anyone to effectively use power use data you really need real time data, but with the present DTE system design that won't happen, and the same for other power companies.

This does not mean the data is not useful, but it is hard to correlate what caused a peak in power at a certain time when that information is 24 hours later. For example, this morning I saw a large peak for a number of minutes in my TED graph. There was nothing that I would expect to produce the peak. Then I thought, did my wife use the electric heater when she got up? That was the cause. If the power information was a day later I would have had no idea of the cause.

My total daily consumption for Wed and Thur were about 5 kWh higher than the previous two days. This was because I had some long 200 F epoxy cure times running. This was easier to correlate because I knew I was the cause.

.

 

[Haji]

gar:
I think the historical energy consumption data may be useful for bench marking purpose e.g increase in energy consumption of an air-conditioner due to defective insulation may be detected by comparing with the bench mark value.

 

[Besoeker]

gar:
I am curious to know about the energy performance of SMPS power supply, if you have one in your residence. I want to know if its efficiency drops with decrease in its output ( i.e whether its power factor deteriorates). Thanks.

Since the input is a plain rectifier, the displacement power factor won't change but the distortion power will be worse/much worse at lower loads.

 

[Haji]

Since the input is a plain rectifier, the displacement power factor won't change but the distortion power will be worse/much worse at lower loads.

This brings up the question if the measuring instrument used by gar was capable of measuring both the power factors.

 

[Besoeker]

This brings up the question if the measuring instrument used by gar was capable of measuring both the power factors.

Distortion and displacement both matter when calculating total power factor. Maybe gar's instrumentation does that.

 

[Haji]

Distortion and displacement both matter when calculating total power factor. Maybe gar's instrumentation does that.

Realiy?!
The POCO measures the total power factor?

 

[Besoeker]

Realiy?!
The POCO measures the total power factor?

My response was in reference to how gar measured it.

 

[Haji]

My response was in reference to how gar measured it.

My response is to clarify it.

So please answer my question.

 

[gar]

130106-1027 EST

Haji:

I believe the Kill-A-Watt takes small time slices and essentially calculates the instantaneous power and from averaging over about 1 second gets the average "real power". Separately Vrms and Irms are measured, again about 1 second averaging. These are multiplied to obtain VA, and average real power is divided by VA. This is what is done in the TED system using the CS5461A. I will have to open a Kill-A-Watt and see if something is identifiable.

Thus, my measurements were of composite power factor, and that is what a power company meter (electro-mechanical or electronic) reads.

To separate distortion from phase shift power factor will require some definitions and then specialized measuring equipment.

.

 

[Besoeker]

My response is to clarify it.

I was responding to this from gar:

Tested an old, maybe 30 years, PC power supply rated at 63.5 W 110 VAC.

I doubt that the POCO had anything to do with that so your question is irrelevant.

 

[mivey]

This group wants to get direct RF data from smart meters in real time to avoid having to use an additional system like TED to get the data.

Maybe things will be different in the future, but I just don't see that happening.

 

[mivey]

I think pulse meters are going the way of the dinosaur, most of the energy management systems we install now want way more information than what a pulse meter can give, but then again, they are using this information to tailor the load control for maximum savings.

That's true, at least for the output pulses. The meters themselves can be quite sophisticated. We have some that practically sing and dance and will still provide a KYZ pulse.

For the discerning customer, I would recommend installing a separate metering and/or PQ point.