Get Smart about Energy™


by Bruce Ackerman – Resident of Fairfax

There is considerable opposition to PG&E’s roll out of smart meters, with many calling for a moratorium on the program until better information is available and concerns are addressed. I would support a moratorium, and considerably better communication, followed by improvements to the technology where possible. Unfortunately, this controversy has become polarized to the point where some citizens believe smart meters are of no public value, and this is emphatically not true: we must find our way through this impasse because the smart grid is critical to the building of a sustainable energy infrastructure.

This discussion is going badly for several reasons, in my view. The proceedings of the CPUC, which regulates the electricity grid, are daunting to participate in. These questions are also very technical: the grid itself is quite complex, the proposed “smart grid” even more so, as well as the technologies involved including wireless mesh networks. Further, the health effects of electromagnetic radiation from wireless communication are not fully understood, and so excite a high level of worry.

Possibly the biggest reason that this discussion is going so badly, however, is the polarization around PG&E. PG&E has made a fool of themselves by opposing Marin Clean Energy and then by trying to pass Prop 16. Among active citizens, we are exhausted and disgusted by these long, costly and unnecessary battles. The public has little faith in the accuracy of PG&Es statements at this point.

And on PG&E's part, they seem to be in a bunker mentality. I have not succeeded in getting replies to emails from people in PG&E who I thought I'd be able to ask questions of. I’ve just learned that Andy Tang of PG&E, one of the people I initially tried to contact, did in fact make himself available to the San Anselmo Council, which is much appreciated. On the corporate level, PG&E’s explanations of the technology, purpose, and future plans for smart meters have been extremely poor, leading to a craving for an explanation as to why so much money would be spent on this roll out for so little apparent reason.

Again may I ask that all involved respect the importance of factual information, even though we know it is difficult to obtain – when we must repeat conjecture or hearsay, please acknowledge so. And please consider the larger context of this discussion. I will try here to describe that context briefly, but first I’ll try (this isn’t easy) to summarize the concerns.

Sorting out the Concerns
Based upon what I've heard, the concerns raised about smart meters seem to include:

1. Exposure to EMF from wireless communication and its health effects.

2. Possibility that a criminal or hacker could intercept these signals.

3. Privacy concerns as to what PG&E or others might do with this detailed usage information.

4. Worries about the power of PG&E to control your appliances, or (I believe this is an example of misinformation) even to shut off or “brown out” your house.

5. Support for PG&E meter readers who would be out of work as a result of smart meters.

6. Reports of billing errors in some earlier smart meter roll outs, leading to fear of huge erroneous bills or of more subtle increases in utility bills.

7. A general feeling that, even if we don't understand it, it's probably nefarious.

I hope that covers the concerns, and it's a daunting list. For what it’s worth I'll try to comment on them one by one.

1. Exposure to EMF
This seems to be the largest concern, and the hardest to address given the lack of consensus on the health effects and the complexity of the science. I think this is a valid concern, though wildly overstated at times, and agree that the precautionary principle should be applied, i.e. that we should work hard to minimize exposure from the outset. Though it's disappointing that a wired communication method was rejected (and I had not known of that decision until this controversy came up), I can imagine that there are good reasons for it (extreme cost difference), though I'd like to have a conversation with someone about that decision. It's most likely too late to turn the ship that radically at this point, though I'm only guessing.

However, the effect of EMF is indisputably linked to the power level of the communication, as well as the amount of time the signal is being broadcast (how often bursts are sent and duration of bursts). These are controllable, and I think there would be considerable room for good engineering to lower all three of these factors. Moreover, these should (in a well-engineered modern product) be controlled by the embedded software (“firmware”) inside the meter. This is important because if so, it would mean that loading updated firmware into the meters could improve them, rather than needing to toss millions of the devices into the landfill and make new ones.

It is important to understand another indisputable fact, which is that the strength of the exposure drops off dramatically with distance from the source. In fact, it drops off proportional to the square of the distance. The importance of this cannot be overstated: it means that unless you are standing very, very close to the meter, your exposure is likely to be a minuscule fraction of that from, say, a cell phone call.

Part of the complexity is that the meters are using a mesh network. What this means is that, rather than every meter on every house needing to be in range to communicate directly with the nearest PG&E neighborhood communication station, the meters could relay messages from one to another until they were received by the station. Thus the meter on your house could be transmitting more often than only the amount required to tell PG&E about your household: it would also be transmitting as it relays your neighbor's information. Again, understanding the parameters of the mesh network being proposed would be helpful, and in all of this it would be so valuable were someone well-informed and in a decision-making power within PG&E or the network designers (a separate company) to directly communicate with someone with adequate technical understanding and the trust of both sides from the community, so we could resolve the huge unknowns in this discussion and talk about mitigation strategies.

Finally, we need accurate and agreed-upon measurements on meters in the field. I’ll suggest a strategy for this.

2. Potential for hacking
I have heard from concerned citizens that PG&E has no encryption on the signals, or that they are easy to hack. I haven’t heard that anyone has even tried, so I’d like to rephrase this concern that “the public needs to be assured that state of the art encryption is being used, as verified by impartial experts”. There are in fact solid ways of encrypting signals, as well as other ways to avoid a criminal being able to know your activities by listening in on your meter. This needs to be talked about openly and done to the state of the art, and then the concern can be laid to rest.

3. Privacy concerns
We have a lot of these in our complex society, requiring involvement and vigilance. PG&E is presently very, very protective of utility usage information getting out their doors (so much so that it impedes legitimate research on energy conservation!), so I feel that this is one we can handle through CPUC. Given the detail of personal information potentially collected by smart meters, CPUC needs to vigorously and clearly prohibit sale of such information for commercial uses, or really any uses not related to electricity delivery and conservation.

4. PG&E controlling your appliances
This is, within limited parameters, one of the capabilities of a smart grid, but not actually of the smart meters. However, everything written to date about the future grid has assumed (1) that we as customers would need to give our agreement before any appliance in our house were controlled in any way, (2) we would be financially compensated for providing this service to the utility, and (3) this is in-arguably good for energy conservation and greenhouse gas reduction. I would like to verify all this, but my guess is that these are still true and we need not worry about some of the scenarios that I have heard talked about in meetings, e.g. “a cell phone call accidentally might turn on your stove and burn your house down”, or “PG&E could at any time turn off electricity to your house”. The Smart Meters PG&E is installing will not be able (by CPUC ruling) to control appliances. An entirely separate Home Area Network would be required for that, and CPUC has ruled that the HAN may not be done by PG&E.

5. PG&E meter readers
A retired PG&E employee told our Town Council that the meter readers were all being either moved to other jobs or given generous severance pay. I don't know if this is the case. I do know that many modern innovations reduce jobs in some areas while creating jobs in others, and I think that if the other concerns were not present this would not have come up. I hope the meter readers do well. I hope we all do well, given the ecological mess we're creating for ourselves.

6. Reports of billing errors
If there are actual malfunctions resulting in billing errors, they'll get ironed out. CPUC regulates billing very strictly. I have heard that some of the reports might be due to a change in rate schedule that PG&E implemented in Bakersfield at the same time as the meters were installed, but unrelated to the smart meters. I myself (because I have a solar system) have a time-of-day rate schedule -- my electricity costs three times as much at 3pm on a weekday as it does at 9pm. I like it -- if I pay attention, I can save money. This type of schedule has long been in place for commercial users, but is unfamiliar to most residential users. We'll need to get used to it, as I'll attempt to explain further on.

7. General distrust of PG&E and the corporate elite
Yup, capitalism isn't working very well, especially to avert climate disaster. Yup, PG&E has acted badly, and arrogantly. But emphatically no, smart meters were not PG&E's idea and CPUC did not approve them lightly -- they are a legitimate and widely-agreed-upon piece in the puzzle, that is emerging more clearly, as to what a post-carbon society will actually look like.

Smart Meters in context
So, (taking a breath), may I try to describe what smart meters and the smart grid are supposed to do? I am not the expert on this, but have been paying attention since before this controversy. What a Smart Meter might be defined as is simply an electric meter that can communicate some aspects of its load back to the grid. The meters PG&E is installing are only a part of the necessary smart grid, so this section describes a vision that goes beyond the meters themselves. I think we need to be aware of this vision, and I think we can take hope from that fact that there is, in fact, the outline emerging of a realistic way to solve our most absolutely pressing environmental problems.

One of the tasks of a utility, and the California ISO (whom we heard a lot about during the rolling blackouts, the group that makes adjustments every 4 seconds to keep the grid stable), is to match supply with demand. At present there is little or no storage in the system, so electricity that we use needs to be generated at exactly that same moment. We require that the voltage and frequency of our power stay within strict limits. So, one thing the smart grid is supposed to do is to use demand management to match up supply and demand more efficiently.

At present we actually have “spinning reserves” and “peaker plants” -- power plants that are actually running, burning natural gas or other fossil fuels, but not actually connected to the grid, or plants that are less efficient but able to be deployed relatively quickly. These are needed because if demand suddenly increases, the ISO needs to be able to quickly switch in one of these plants to keep the grid up. Obviously this is wasteful.

An alternative to this would be to shed some of the least critical loads, in response to a spike in demand. This is where the smart meter might “talk” (editor - using a 3rd party service you agreed to) to your refrigerator, telling the fridge to turn its thermostat up a few degrees for a little while. We'd keep the grid stable without burning the substantial fuel required by spinning reserves.

Also, the utility having detailed usage information allows them to more accurately predict the aggregate load. Prediction is always a part of what they do, again in order to know what generation they'll need to keep the grid stable.

As I said, electrical rates that vary by time-of-day are common in the commercial sector and are in place for those of us who have rooftop solar, and they will be widespread soon. The reason for these is that the cost of power during peak hours is far greater than the cost during off-peak times -- cost in fuel use and cost in capital equipment for new power plants (having a power plant that is only used a few times a year is really, really wasteful). For both environmental reasons and for keeping our electricity rates low, we want to shift load away from peak times. Giving consumers an incentive to run our washing machine or dishwasher at night, for example, is a no-brainer as we design a sustainable society.

We're about to see (this year, we hope!) entry of the electric vehicle (EV) and plug-in hybrid electric vehicle (PHEV) into the mainstream market. This requires collaboration among utilities, automakers, consumers, and others, and much is going on to get that in place. The benefits are enormous and the smart grid is how we achieve them.

First of all, charging a car is a big electrical load. Studies have shown that we could support widespread adoption of electric vehicles with no new grid infrastructure needed if the charging were done mostly at night. Without incentives, i.e. a time-of-day rate structure, why would people bother to set a timer on their charger at all?

Chargers for cars, particularly PHEVs, would be the ideal load to be shed to manage demand. It goes much farther though. Here's a big idea, which is a little later on the time horizon but which has many people very excited. Electric vehicles can actually provide power back into the grid, referred to as Vehicle-to-Grid (V2G).

In background, many of the renewable energy sources we want to see deployed (solar and wind for example) are intermittent -- i.e. they produce abundant energy some times, but they don't produce it other times and this is not under our control. In Europe, where some countries are way ahead of us in deploying renewables, they tend to get to about 25% renewables and then beyond that, more renewables don't make much sense. Why? Because they still need enough back-up fossil power plants, many of them “spinning”, in order to keep the grid stable when the renewables drop off in their production. So our dream of a 100% renewable society runs into this obstacle. This is real. Bottom line, we need storage on the grid.

It turns out that the batteries in electric vehicles, which most of the time would be plugged in, are a huge resource for such storage. So allowing the grid to actually draw power when needed from these batteries (and pay the car owner for this service) would be the key that unlocks our ability to really get to a fully renewable, distributed power network. And it also provides a much-needed incentive to electric car buyers, providing us an income that would offset the increased expense of these cool vehicles.

There's another reason for smart meters, which we hear more about during the present debate in part because PG&E has chosen to highlight it. This is that if we consumers have better access to our electric and gas usage data (including being able to see our usage on close to a real-time basis), it is likely that we will be able to better understand our household and conserve. There's a lot to the puzzle of how to get people to conserve, including putting a price on carbon, education, and just the general distractions of our busy lives, but having good information to act on is definitely part of it. Also, conservation can be advanced by research on consumer behavior, and by targeted communication to the largest consumers, which should lead to care in drafting the specifics of any proposed privacy policy. We will need everything at our disposal if we really expect to leave a decent planet for our descendants.

So, in summary, I'd like to see this conversation consider some ways out of this impasse.

On the technical level
1. The meters may not need to send minute-by-minute transmissions unless you are actually watching a web page showing your usage. This alone could vastly decrease the wireless traffic.

2. Wired connections should be considered, either at roll out or as a retrofit in selected cases.

3. If wireless, the power and duty cycle of transmission should be minimized and adapted to conditions, like a cell phone already does.

On a regulatory and communication level
1. CPUC needs to require all information about these concerns to be shared with the public. There is precedent for CPUC directing PG&E to be better communicators.

2. Though the meters are already designed, new firmware can surely be downloaded into them to implement improvements. If that is not possible, it's a poor design.

3. There should be a committee of citizen representatives involved with these design decisions, to include engineers familiar with embedded firmware, RF communication and protocols, and communications over power lines or other methods.

4. Taking this opportunity to set down the rules and expectations for access to utility data, we should involve those qualified and experienced in conservation research to enable access to this vital data for the public good.

And a final recommendation regarding the EMF issue
EMF is very difficult to measure reliably; hence there is little credence being given to citizen complaints by CPUC or PG&E. PG&E already contracted an expert to measure the meters’ output under laboratory conditions and extrapolate to real-world conditions. I have read those reports and they appear professional and accurate, and show that the radiated energy is far below FCC and other standards using conservative assumptions. However, we hear reports that some meters may be emitting far above these levels. I believe the solution would be, through some mechanism such as a CPUC study, to get an independent measurement expert with really good equipment (I’d consider Richard Tell Associates, who PG&E already used and therefore understands the system) to go to several sites where these meters are deployed, including some where people are complaining or have measured high EMF using whatever measurement tools they had available, and get more reliable and comparable measurements. In my ideal scenario, Mr. Tell would be accompanied by a PG&E witness if they wanted, and with a community witness (someone with enough background to know more or less what Tell was doing). Let community activists choose several places to make measurements, and let PG&E choose some too if they want, and pick a few at random. Then see if those measurements indicate that there are real-world conditions that were not taken into account in the original certification, and if so get to the bottom of what that's about, in an engineering sense.

I hope we can all act to calm this argument and try not to throw the baby out with the bathwater!

If you would like to comment on this article, please use the "Contact us" form above - and we will have you message sent to Bruce for publishing and comment. Alternatively, please register for an account at open4energy, and you will be able to add your experience to the discussion.


Open4Energy - Get Smart about Energy™