Current Cost Support Forum

In the U.S. I have the EnviR and am observing ~25% error on my display unit. For example, when my utility company's Smart Meter reads ~400W, my EnviR reads ~500W. My mains line voltage reads ~235Vac at the Smart Meter. I kept the Sensable Transmitter voltage setting at the default during installation, "115V". Should I try lowering the voltage setting or raising it, or is this much measurement error (set to "115V") about right as-is?

I must say i thought that the US default voltage was 115V, that's why the EnviR is set to that.

If your mains voltage in 235V, then change the default to that figure! Should solve your problem

In the U.S. as is the case for me, standard wall outlets throughout a house are 115Vac. But, the main power source is a single phase 230Vac with a center tap neutral ground, 115V from neutral to each side of the line.

The EnviR and Sensable Transmitters sold in the U.S. have a default voltage adjustment set to "115V". The range for this adjustment at intervals of 5 volts is 100 to 130V.

Each of two CT is sensing L1 and L2. There is no CT on Neutral. There is only a SINGLE phase AC line connection to my residence. This is the household wiring standard in the U.S.

The only branch circuits that have 230Vac at the outlet are usually for an electric clothes drier and electric oven. 115Vac is also available at these outlet from either side of the line to neutral. All other branches are 115Vac, i.e. one side of the line to neutral. I think that is what the so-called EnviR "voltage adjustment" is based on, which is confusing because the location of the CT sensors is at each of the two lines, with a voltage potential across the line of 230V., NOT 115V.

One of the issues with the CTs is that they only measure the current. In a purely resistive load, the current and voltage waveforms follow each other, and it's nice and easy to calculate (power factor = 1). When you start to introduce capacitive or inductive loads (switching power supplies, AC motors), the current and voltage waveforms don't necessarily follow each other any longer. That puts the power calculations out a bit. Getting the consumption figures exactly right really means being able to put a current shunt in line with the phases, and measuring the voltage on each phase. That'll require an electrician to wire stuff in, as well as more stringent safety tests so most consumer solutions don't go to that length.

You can adjust the line voltage setting to try to match your real consumption a bit better, but you won't be able to get the exact figure - there'll still be some inaccuracy. Give it a try and see how things work out.

This inaccuracy appears to have been one of the drivers behind the Optismart, which picks up the information from a flashing light on the electricity meter (this is UK, don't know about US meters), so gives you a more exact figure.

In the U.S. there is also the difference in current in each leg between L1 and L2. With Techtoniq, the difference is very apparent, and seldom do I see anything exactly a 50-50 split.

Nevertheless, what you say is quite true regarding non-unity power factor, which is a variable, EnviR cannot address based on simply current only measurements.

Maybe it was just a coincidence, but the other day I did a side by side comparison of my EnviR Energy Now vs. Utility Meter, and when my whole house consumption was almost exactly 1000 watts, they were almost exactly the same measurements. The difference was less than +/-1%, which was quite satisfying while it lasted. At much lower whole house consumption, the EnviR readout was much higher than the utility meter, and vice versa at much higher levels of several kW; apparently a totally non-linear relationship.

What might be a better way to "adjust" for inaccuracies in your product, is for the user to accumulate sufficient data in the EnviR to obtain a measurement of "average" kWh per day (over consecutive 24 hour periods), say for a one month period, or longer. Then, look at the utility bill over the same period of time and calculate the household's average consumption rate per day (if not already given on the statement). The difference between the two averages, as a percentage of the utility meter taken to be the "actual", could then be used to (software) adjust the Envir's realtime Energy Now reading. This would provide a simple way to "calibrate" (albeit crudely) the EnviR display monitor. Of course, the caveat that must still be understood, is the display still reports an approximation of the actual power consumption at any given moment.

I think this method would be more efffective for correlation purposes in terms of COST assessments, which is what most EnviR users are most increased in using it for in the real world.

Indeed, as you say, there is likely to be a difference between the two phases, but that should be covered by the two CTs. The transmitter should transmit them separately, but the receiver seems to mix them together for its display. I don't know if the separate outputs are available in the XML stream (serial port) though. Would be good if they were! That said, you could pull out the data from the XML stream and try your correction method and see how it works out.

In the UK, the optismart product reads the flashing LED on the utility meter. The LED flashes one per x WH, and the product just counts the number of impulses to get an accurate reading.

I am not a programmer, so I don't know much about XML or that sort of thing.

However, the Techtoniq Energy Monitor (V2.05) which is a 3rd party application I use with my EnviR, does indeed separate the 240Vac single phase individual currents in to two branches. The particular app can display individually and integrate, up to 3 phases as well. That's how I found out there were two different currents on each side of the line in the same single phase mains power feed to my house (L1 and L2).

Where I live in Northern California, the utility company (Pacific Gas and Electric) recently converted all residential customers "voluntarily" to their new Smart KWh Meters. They are constantly remote monitored by the power company, eliminating the need for a so-called "meter reader" power company employee to visit everyones residence once per month. These new meters do not have a rotating wheel or flashing LED and therefore are not compatible with the OptiSmart device unfortunately.

As an aside, there seems to be a lot of resistance at Cost Current to improve the EnviR design to add new features that customers have been asking for, e.g. any and all ways to improve accuracy being one of them. A backlite display? Rechargeable battery power supply? Etc. EnviR is a nice product already, but it could be much more versatile. Why not make it even better?

That's cool, the 3rd party app is dealing with the XML output for you - hence you were able to discover the reported information. I agree on the fact that there's plenty of places that the current version of EnviR could be improved. I know from reverse engineering the protocol in use that there are issues with communication that could be made much better. Sadly I can only see that this could be done by breaking existing devices (or returning them for reprogramming - effectively a recall on every device).

Oddly enough, CurrentCost seem to be manufacturing a device for one of the UK energy companies that seems to have dealt with some of the communication issues (See here) - so with the new device, the monitor polls devices, so there's no chance of them corrupting each other's transmissions by talking at the same time. Still no backlight though. I wonder if the clock is better able to keep time?

I wish Current Cost would export their "switchable" IAM outlet monitor to the U.S. That alone would be a very desireable enhancement feature to the Envir Monitors. But, I suspect you are correct and the existing EnviR design will not support two way communications like apparently the "Eco" Monitor does.

I want to upgrade to their Internet wireless gateway product, because I need to a way to capture energy consumption data 24/7 without leaving my PC on all the time. Unfortunately the Gateway also is not yet available in the U.S. I am not even certain it is available in the UK yet either.