GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Medium- to long-term reliability of the electrical grid

DCcontrarian | Posted in General Questions on

This is a split-off from another thread about backup heat. A poster wrote: 

Our national plans to convert vehicles to EV over the next few years will increase the demand on our grid tremendously, not including the plans to try to convert most fossil fuel heating systems to various forms of electric heat pumps. And our grid is very susceptible to various forms of attack, including the one recently in NC. So I’m concerned about the possibility of more frequent outages in the future. I’m not planning to go off grid, just find a way to be prepared for more unreliable power.”

In the short term power may be more unreliable, but if you look at almost any commodity, in the long run increased demand makes the supply more reliable, not less. Look at how much more reliable the internet is than it was 20 years ago. If you look at both sides of the cost-benefit equation, increased demand makes increased investment in reliability more worthwhile: more users means more people to spread the cost of infrastructure over, and more usage makes the cost of unreliability higher. 

On a policy level, if we’re trying to electrify the economy, we really don’t want every household to have a backup generator and fuel depot in the back yard. 

Your thoughts?

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. plumb_bob | | #1

    I have a backup generator in the yard so I can keep the sump pump and freezers going for a short while if need be. I feel our infrastructure is extremely fragile and could fail on a large scale from a variety of challenges. We hear much talk of going all electric but less talk of required grid upgrades.

  2. Expert Member
    BILL WICHERS | | #2

    I'm not so sure I'd say the Internet is "more reliable" today, but there is certainly a lot more access to high speed connections, and the capacity of the networks in general has increased significantly. For anyone curious, big drivers of this have been two fiber optic technologies, mainly DWDM and the EDFA. I don't want to go into detail here since it's very off topic, but that is my industry and it's cool stuff for anyone curious about why capacity has increased so much and costs have dropped so much in the past two decades or so. Much/most of the actual "backbone" -- the intercity fiber optic cables themselves -- were installed in the 90s, with the capacity increases due to those technologies I mentioned. Nifty stuff.

    Anyway, the backup generator issue is a seperate one. One major thing that comes up is that it's easier to run a gas (or oil, or propane) fired furnace on generator power than it is to run a large heat pump, because you need a smaller generator to do it. This means if you want backup, an all-electric house is harder to fully backup than one using something like a conventional natural gas furnace, because if you have natural gas available, you have a second energy source not entirely dependent on the operation of the first. If you're all electric, you obviously lose this option. This is a bigger concern for areas with frequent power outages, and many of those areas are also more difficult (read: more expensive, in most cases) to improve in ways that will significantly reduce outages. Even with lots of effort, there is still the possibility for infrequent major disasters.

    BTW, "put the lines all underground" isn't really an option. When you do a major undergrounding (utility term) project, the costs are very high, but there are also some practical issues that limit how far you can run electricity that way due to cable reactance. The solution to this is to run DC instead of AC, but then you need converters, so either way there are issues. The cost is also very much higher (for telecom, underground is about 5-6 times more expensive than aerial runs, and while I don't have exact numbers for power utilities, I expect their cost different is significantly higher, since they need special cables for underground installations while the same fiber optic cables can be run aerially or underground). Underground lines get damaged too, so regardless of what you do, stuff can still break.

    I don't think we're ever going to have backup generators everywhere. We don't now, and I doubt going all electric would change that. Remeber that a gas fired furnace won't run without electricity, either, so having an all electric heat pump system would be down just the same as a gas fired furnace would be without electricity. I've heard people argue that heat pumps make you more dependent on electricity, but that's not really fair, because gas furnaces need electricity to run too. It's important to be honest with this kind of thing I think.

    The big issues as I see them are a combination of needing more transmission (the BIG power lines, usually on metal pylons), and more generation. There is a lot of resistance to building additional transmission lines due to people either being scared of them, or not wanting to see them (the "NIMBY" issue). The thing is, if you want to run more things on electricity, you need to move that electricity around, which means more transmission capacity is needed. Sometimes higher voltages can be used (max in common use is 765kV, and it's not all that common), but sometimes additional circuits (new routes, more wire and pylons) is needed. To replace all gasoline vehicles and natural gas heating, we'd probably need AT LEAST double the transmission, everywhere. Remember that what "electrification" here really means is replacing one energy source at the point of use (car, furnace, etc.) with another, so you are REPLACING the energy provided by that other source with electricity, and that electricity has to come from somewhere. Transmission lines move bulk power, they are the backone of the grid. We're going to need a lot more of them. Smart technologies helps -- and I love seeing people advocate for those since I help build the communications networks that support them :-D -- but smart technologies alone are not enough to sufficiently increase the capacity available to move all this power.

    Generation is the other issue. I know people don't like to hear it, but I don't see "wind and solar" as being capable -- ever -- of fully supporting an entirely electrified economy. This is due to numbers, math. The capacity just isn't there, and I don't think enough can be built. I DO think both can help though. Solar, for example, is especially helpful in offsetting heavy load for airconditioning in the summer, which tends to be highest on hot, sunny days. Solar is also fairly predictable on a day to day basis. But then there is night, which is also very, very reliable. Night is a problem. Wind comes and goes too, but I hear people say if you spread if over a large enough area, then it's "dispatchable", which is a fancy way to say it's predictable enough that you can plan on having a certain percentage of capacity available at any given time, reliably. I've been keeping track of that in my area (MISO) for the past year or so, taking somewhat random snapshots of the generation mix. I've included three pics to show the variation over just a few days.

    Some will say I cherry picked these times/days to make my point, and I did -- I intentially picked a time range that shows a very big swing, to make it easy to see what I'm talking about. For those people, I'll mention that out of the 100+ of these random samples I have over most of last year, the average level of wind output over most of the year tends to be closest to the pie graph showing wind at the lowest level of the three pics I attached. The summer tends to have very low wind production, the spring and fall slightly more, and the most I've seen has been during the past month or two of winter.

    What about spreading it out? MISO covers 900,000 square miles in the central part of the US and part of Canada, covering 15 states and one Canadian province. While they don't really have data as to exactly how the wind farms are spread out as far as I know, I think it's safe to assume that at least a good portion of the total available wind capacity is spread out over a number of wind farms spread across their service area -- it's not all in any one spot. BTW, for those following that "EV charging load vs reliability" in the other recent thread, MISO's peak demans in 2018 (a handy year I happen to have data for), was 121,563 MW. That's not enough to handle all that EV charging, although keep in mind that all those EVs wouldn't be likely to be only in MISO's territory, but it's a good perspective.

    The issue isn't that everything can run on electricity, the issue is that it is a multidecade process to get there. People tend not to be aware of the MASSIVE scale of the grid. A little solar here and a little wind there isn't going to support, for example, the conversion of all gasoline vehicles to electric. It is going to take a LOT more generation, and a LOT more transmission to support that, and it takes a good amount of time to build all of that. New technologies can help improve efficiency (smart grid stuff can help with load leveling -- spreading loads out throughout the day -- for example), but it can't replace the need for a lot more input energy to supply all this stuff.

    I kinda bet DC knew I'd have a lengthy response to his Q too :-)


    1. Patrick_OSullivan | | #3

      > I'm not so sure I'd say the Internet is "more reliable" today, but there is certainly a lot more access to high speed connections, and the capacity of the networks in general has increased significantly. For anyone curious, big drivers of this have been two fiber optic technologies, mainly DWDM and the EDFA.

      I'll go ahead and pick the least GBA-relevant thing to comment on, because why not? :-)

      I will say that though the economies of Internet access don't necessarily prioritize reliability over other factors, continued evolution of technology like coherent optics and tuneable lasers has made it so much cheaper to build and operate both higher capacity and higher reliability networks (depending on your priorities)--even just over the last 5 years.

    2. Expert Member
      DCcontrarian | | #4

      >I kinda bet DC knew I'd have a lengthy response to his Q too :-)

      Yeah, baiting you is like shooting fish in a barrel!

  3. plumb_bob | | #5

    I am in a mind frame that sees exceptional events as becoming more common, both natural and man made. Several years ago things like global pandemics, record inflation and trench warfare in Europe would be unthinkable.
    Is the grid surely going down? Of course not. Is there a chance that it could due to an extreme event? Of course. Weather, solar flare (is this a real worry? Who knows!), cyber warfare, physical warfare, etc etc.
    Because it is the one system that makes all of our other systems run I like to have some back up, even for a short time. I am not a prepper, but I read plenty of history, which shows unthinkable failures happening to us humans over and over.

    1. Expert Member
      BILL WICHERS | | #6

      I doubt we'd have a complete failure of the grid, but a large scale failure consisting of many large areas being simultaneously down wouldn't surprise me. In terms of overall damage, things would probably be pretty similar regardless.

      Solar flares ARE a real concern. There have been some relatively recently, such as:
      The march 1989 geomagnetic storm that shut down the Quebec Hydro grid in Quebec
      The BIG one we know about: the Carrington Event in September of 1859

      A solar storm like the Carrington event would have been very, very bad for today's modern world. NASA has several spacecraft that monitor for these sorts of storms, although the recent modern world has never seen one like that Carrington event, although we did come very close in July 2012 , where NASA's STEREO-A spacecraft was able to get measurements, but Earth wasn't in the line of fire, so to speak, from the sun at the time.


  4. Expert Member
    DCcontrarian | | #7

    Car battery charging is perhaps the ultimate load-leveling use. Think about it, a typical car is used about an hour a day. So you've got 23 hours to charge that battery, pick your time. There's even been talk of using parked cars as a source of peak demand power, by having them put electricity back into the grid from their batteries.

    But I do think that the model that seems to be most talked-about now -- basically replacing the gas tank with a battery with about the same capacity -- doesn't make sense in the long run. The problem is you have to carry the weight of that battery everywhere you go for the life of the car. What makes more sense to me is electrifying the interstates and other major highways, and equipping cars with a battery big enough to get around town, get onto the interstate, and get between interstates on longer trips. This is the only way that electric trucks are going to be feasible.

    I think it's going to be a long time before we stop burning fossil fuels altogether. The direction we will go rather quickly though is no longer burning fuel on an individual level, and doing it in centralized facilities where it's easier to collect and control the emissions. Just as we transitioned very quickly from burning gas, kerosene and candles for lighting to electric light, we will transition away from burning fuels in our homes fairly quickly. I wouldn't be surprised if in 30 years burning fuel yourself for heating, cooking or transportation is about as common as burning whale oil for lighting.

    Note that it didn't require government intervention to make the transition to electric light; once it was available and the superiority was established the transition happened on its own.

  5. maine_tyler | | #8

    As a rural resident in the twigs of the grid, it's hard to imagine not having some source of individualized backup (whether talking electric or heat).

    Of course rural folks in the twigs are a minority and will not prevent a majority swing towards zero localized combustion.

  6. Expert Member
    BILL WICHERS | | #9

    I would expect some availability of fuels as a backup power source to continue indefinetely, or at least until some superior technology comes along (not just a battery). There are industries that have to prepared for any kind of power outage, such as my industry (telecom), but also healthcare, broadcasters, and others. Fuel will be available for backup generators for those industries for a very long time, since those industries depend on continuous power to operate, and no matter what is done with the grid, there will always be some potential for outages. Rural people would likely also be able to access that fuel. I also think large farm equipment is unlikely to be electrified anytime soon, since the stuff can be huge, and there isn't a lot of it out there relative to cars.

    I do NOT NOT NOT like the idea of using car battery packs for grid load leveling. Scheduling charging with a smart system, to take advantage of idle grid capacity to charge at the "easy" times is. good. Using the power in that EV battery to feed back into the grid is what I don't like. The big reason is that batteries have cycle lives -- they degrade in part based on how many times you "cycle" (charge and discharge) them, so load leveling using EV packs will wear out the batteries in these EVs faster. Who pays for that? Do we do some kind of net metering? EV owners just have to suck it up and buy a new car sooner? Battery life is finite, so this isn't an unlimited resource. It's a neat idea, but there are some issues in practice. I'm not a huge fan of grid scale battery storage in general though, simple because batteries are NOT really "grid scale". You just don't get all that much energy density with batteries.

    The issue with charging EVs for longer drives is that charging is slow relative to refueling. That's gotten better, so maybe that can be solved, or at least improved enough that you can charge fully over lunch. Trying to electrify the roadways is probably not really practical on a large scale, but I admit I've not really considered that. I do think that would significantly increase maintenance costs on the remote sections of the interstates.

    We need to be careful the local burning of fuels. Running an electric resistance water heater on a grid with natural gas fired generation is less efficient than just burning the gas directly, for example. Heat pumps are the only way around that. I also think the push in some areas against gas cooking stoves is misguided and a little over the top. If someone wants to cook over gas, that's not a big deal.

    I absolutely agree with your last sentence DC. When technology improves, and it's been improving, sufficiently to make it a practical/economical/superior choice, people will naturally switch over since it's to their own benefit. We will get there with vehicles, it's only a matter of time. EVs are already much better about not needing lots of maintenance compared to ICE vehicles (electric drive trains tend to be more reliable), the fuel supply and range issue is really the only thing standing in the way of more acceptance in the market. Costs will come down as volume goes up too, so that's not going to be a problem forever, either. The reason for all the gov't intervention is that people are trying to push new technologies prematurely, before they're really ready for prime time. That's where trouble arises.


    1. Tim_O | | #10

      Depth of discharge has a huge impact on life usage of a battery. If EVs were able to sell back during peak demand, I would think you'd want to limit it to 10-20% of the battery, and the power companies would need to incentivize this to make it worthwhile.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |