04/15/2023

Hello All About Circuits!

I need help with a simple question that any electrical engineer (I'm not) should be able to answer.

As we are all aware, smart meters are being installed world-wide as part of AMI (Advanced Metering Infrastructure) networking.

The manufacturer of the smart meters being installed by my local public utility is Itron, headquartered in Washington State.

The smart meter name is Gen5 Riva Meter.
The specification sheet can be found here: https://www.itron.com/na/solutions/product-catalog/gen5-riva-meter

I've attached the spec sheet to this post if anyone cares to look at it.

The meter is essentially a computer with a CPU, OS (Linux), RAM, SSD and a RF radio transmitter.

The meter will transmit and receive data multiple times every day, 365 days a year.

Itron offers its customers 5 different intervals of time in minutes (5, 10, 15, 30, 60).

I assume transmission and reception are "active" bidirectionally within the chosen interval of time.

I tried to contact Itron and my public utility company asking this question but received no response:

"How many watts of power does the Gen5 Riva meter use when it is connected to the supply side of the utility line?"

The meter uses energy in watt-hours to continuously capture and record kWh (customer demand side of the utility line) in addition to when it transmits the data (Radio Output Power 1W) and receives data within the chosen interval of time.

Does anyone know the answer to that question?

I would greatly appreciate if you can answer this question or if not, point me to another person who could.

Thank you in advance,

PokeTheTruth

 

I am only guessing, but this meter would be a completely insignificant load compared to your normal living requirements. Don't worry about it.

 

There's nothing in the specification sheet as you say so without measuring the current it's hard to say although I'd be surprised if it's not similar to a smart phone and therefore not worth worrying about?

The meter may derive its power from the power input side and therefore cost the customer nothing?

 

...therefore cost the customer nothing?

Nothing costs the customer nothing.

Customers pay for everything.

Edit: Including transmission losses.

 

Approximately one watt average continuous, depending on the particular meter. As noted above it is not taken from the consumer side so not metered to the end power consumer but treated as a transmission loss.

 

There's nothing in the specification sheet as you say so without measuring the current it's hard to say although I'd be surprised if it's not similar to a smart phone and therefore not worth worrying about?

The meter may derive its power from the power input side and therefore cost the customer nothing?

They may not be being charged directly, but it is certainly included in the rate that they pay for the energy they are being charged for.

It's not going to be very much, and what matters is how much MORE energy it needs over the lowest-energy means of measuring energy usage.

But it doesn't take much energy usage to add up to a noticeable amount given how many such meters there are going to be. It would need to be put into some meaningful perspective, such as how much it is relative to the other parasitic losses, such as transmission line losses) or perhaps how many feet of additional power line between the transformer and the service panel that it is equivalent to.

The big question is whether the additional data obtained is actually used to make decisions that save considerably more energy that could not have been saved without that data.

 

Even back in the days of the electromechanical meter, the voltage coil consumed at most a few watts, and connecting it ahead of the current coils in the meter ensures that the energy consumed by the meter itself does not register on the meter. The power supply in smart meters is still going to be hooked up the same way and there are industry standards that set the limits for how much energy can be consumed by the meter regardless of technology.

Like Joey says, we DO pay for that, but it's gonna be an insignificant fraction of the monthly overhead charge we pay before the energy usage enters the equation.

 

Even back in the days of the electromechanical meter, the voltage coil consumed at most a few watts, and connecting it ahead of the current coils in the meter ensures that the energy consumed by the meter itself does not register on the meter. The power supply in smart meters is still going to be hooked up the same way and there are industry standards that set the limits for how much energy can be consumed by the meter regardless of technology.

Like Joey says, we DO pay for that, but it's gonna be an insignificant fraction of the monthly overhead charge we pay before the energy usage enters the equation.

You mention that there are industry standards that set limits on how much energy the meter can consume.

Do you have any kind of a reference for that? That number would at least give the TS an upper limit that they can use.

 

An electronic meter draws some amount of power (possibly less than the electric spinning disk type), and as others have said, whether it draws from "their" side or yours, ultimately you end up paying for it. It's a cost of doing business.

But before y'all get your conspiracy theories in a knot, how does it compare to a meter reader driving down the street, getting out of his truck and reading your meter every month or two? Electronic meters have been installed to reduce costs of determining usage. They may be used for demand charges (and possibly crappy power-factors) but this just means if your usage costs more, YOU pay for it.

 

1 KWH seems to cost between 10¢ and 20¢ most places in the world. One month is about 720 hours if a power meter takes 1 watt, how much does that cost the customer? Why is that worth worrying about when it is less than the cost of mail a letter in most countries?

 

1 KWH seems to cost between 10¢ and 20¢

More than twice that here (.48) Which is why Massachusetts, despite being far north and not particularly sunny, is one if the best states to install solar.

 

More than twice that here (.48) Which is why Massachusetts, despite being far north and not particularly sunny, is one if the best states to install solar.

With that, I will stop complaining about my electric bill. Checking, it is still less than a first class letter in the U.S. Now starting to sound like my grandmother "When I was a child it was 4¢ to mail a first class letter..."

 

a good way to estimate things like this without actually putting them into a circuit is to check teardown info. should be also available on the FCC website, such as:

https://fccid.io/SK9ACT2/Internal-Photos/Internal-Photos-3946762

besides seeing inside circuitry, documentation itself does include more info like in this photo

 

looks like a typical meter for 200Amp service, 240V, power consumption 3W

 

With that, I will stop complaining about my electric bill. Checking, it is still less than a first class letter in the U.S. Now starting to sound like my grandmother "When I was a child it was 4¢ to mail a first class letter..."

Oh, God. It was 6 cents when I was a kid. In kindergarten I mailed a letter from my grandmother's house to a TV station to enter a drawing contest. I few days later I got a letter from them at my home -- I was so excited because I was sure I won. Until I realized that I had swapped the mailing and return addresses on the envelope. So how much younger than your grandmother could I possibly be?

 

39p per kWh for electricity, and 54p per day standing charge.
That means that the standing charge represents a 57W parasitic load (1.38kWh per day). A lot more than the meter takes!

 

Did everyone miss the battery spec?

It radio transmits data at 1W and then this

Power Outage Notification (PON) » Standard – 25 second hold up (10 second momentary filter + 15 second transmit window). Meter transmits only its own PONs during 15s window. » Extended Last Gasp – 75 second hold up (60 second momentary filter + 15 second receive/transmit window). Meter transmits its own PON and receives/transmits neighboring PONs during 15s window.

Which indicates to me that it doesn't continuously transmit and I would expect the same communication protocol transmitting usage data monthly. Plus

Display duration: 1-15 seconds

Which indicates it is saving output power driving the display as well as the radio. I would expect that even without battery power that any power used for metering would come from before metering at no cost on the customer billing (other than that hidden in the company operating cost per kWh). Lots of assumtions...

 

OP raises a good question. I actually got my utility to request the info for that very meter from Itron as part of a regulatory hearing. Itron gave only an average figure (which I forget, was 3-5 Watts, similar to other smart meters). But under full CPU load, I assume it could be much more. And, since it can run 3rd party apps, it's conceivable someone (hacker, rogue former employee?) could try to use the proposed half-million nodes as a supercomputer.

 

Welcome to AAC!
3.5W per day, week, month, year? It should be on their dime although it is the user who ultimately pays for it. Even then, it is miniscule compare to paying meter readers salaries, benefits and retirement costs. I doubt that anyone would want to use its tiny uP for nefarious reasons with all the RF communications overhead included as there are infinitely better solutions if it was even possible.

 

It runs linux! The speed is low, but half a million nodes!