TMC is an independent, primarily volunteer organization that relies on ad revenue to cover its operating costs. Please consider whitelisting TMC on your ad blocker or making a Paypal contribution here: paypal.me/SupportTMC

Garbage Article at MIT Technology Review

Discussion in 'Model S: Battery & Charging' started by Donkasan, Aug 17, 2013.

  1. Donkasan

    Donkasan Member

    Joined:
    May 18, 2013
    Messages:
    14
    Location:
    CA
    There is a garbage article at MIT Technology Review today that claimed that the Tesla Model S will overload the local grid. It said one model S is the equivalent of one house's electrical load. This is nonsense.

    I have written a reply in response...
     
  2. FlasherZ

    FlasherZ Sig Model S + Sig Model X + Model 3 Resv

    Joined:
    Jun 21, 2012
    Messages:
    7,019
    I disagree, it's not a garbage article, it's not nonsense, and it's largely true. A Tesla Model S can increase a home's instant electric load by up to 1000% or so in some cases. Many homes in California draw less than 10A @ 240V at any given point in the day or night. A single-charger model S will increase that instant load by 400%. A dual-charger Model S will increase it by 800%.

    In neighborhoods like mine, it isn't a big deal - there is one transformer for every 2 houses, and they're pretty beefy (37.5 kVA). But in certain areas of the country, a single transformer bank may serve 8 or 12 homes with the rating of 1960's requirements. That is a big concern and only a couple of Model S's charging at once can send the oil boiling out of a transformer. Power companies are permitted to oversubscribe transformers because you never use the full rated capacity of the service entering your home. Prior to my upgrade, the transformer serving my home (I'm the only one on it) was rated at 15 kVA (about 60A or so) while I had a 200A service panel. Moreover, even prior to the Model S the power company recorded some cases where I was drawing over 120A. So imagine if I hadn't called the power company. I'm drawing 120A on a transformer rated for 60A (which usually works without a hitch), and now I add another 80A via HPWC... I can pretty much guarantee that transformer is going to start suffering, if not venting its oil and burning up.

    Additionally, they're correct when they say that we really need to look at active management of demand as well as supply. Today, power companies actively manage their supply only, dynamically purchasing additional power as uncontrolled demand increases. In the future, having control of both supply _AND_ demand (or at least that demand which is arbitration-enabled) will help ease the difficulty of managing supply and reduce the likelihood of rolling blackouts in areas.

    rolosrevenge knows a hell of a lot more about this than I do, and I can let him address some of the more intricate details.
     
  3. Cottonwood

    Cottonwood Roadster#433, Model S#S37

    Joined:
    Feb 27, 2009
    Messages:
    5,062
    Location:
    Colorado
    The numbers are real, but you and the MIT article, pick worst-case, low probability points to get attention. There are lots of big loads that people add. For example, a new steam unit in the shower. Steamist makes units up to 15kW and you can put multiple of these in for a large shower. Turn on all the burners in a new electric range for that big Thanksgiving meal and you can draw 15kW.

    Most neighborhoods have just enough capacity in the local transformer because that saves the utility money, and in the end, keeps rates low. As the article points out, planning is what is needed. Offering incentives for the EV owner, like Time of Use (TOU), EV rates are a great start! This is a win-win. The utility knows more load is coming, and, as a part of their normal grid upgrade plan, put larger transformers in the right spots. The customer gets better rates for charging at low demand times, and thats better for the utility, also.

    What I had not seen suggested before, is the concept of the utility controlling the EV demand with smart grid signaling. I have seen others suggest that the batteries in the EV be used to source as well as take power. Given the price of EV batteries and the concern over battery lifetime this seems to be a bad idea. However, allowing the utility to control the demand side by telling the car when to charge or not, seems like a great idea. For those times when the customer needs a charge NOW, you just need a priority button somewhere to force a max rate charge now. I have TOU electric water heaters at my house; on the side of each, there is a nice little red button to force heating of the water at peak rate times. I use this a couple of times a year when I have several guests over and they all shower after a day of skiing, biking, or hiking.
     
  4. deonb

    deonb Active Member

    Joined:
    Mar 4, 2013
    Messages:
    3,020
    Location:
    Redmond, WA
    Well, the Model S is great in this way, because it allows the utilities to make the grid much smarter very cheaply.

    All Tesla needs to do is to upload the information about your charge schedule to a central location and allow the utilities to access it. The utilities will then have a way to exactly plan what power will be needed when.

    Tesla can also go further by having users instead of saying when to start charging, say when they need the charge by (some users will anyway prefer this). Utilities can then control which cars should charge when, to stagger the impact on the grid. It's their dream smart grid scenario, all doable in just a few lines of code.

    I actually volunteered to my local utility (Puget Sound Energy) to create a pilot program where a Model S owner can sign up, and they can control and monitor the car via the REST API. However, PSE has so much extra capacity that they don't see the need for this for the next few decades.
     
  5. FlasherZ

    FlasherZ Sig Model S + Sig Model X + Model 3 Resv

    Joined:
    Jun 21, 2012
    Messages:
    7,019
    #6 FlasherZ, Aug 17, 2013
    Last edited: Aug 17, 2013
    I'm going to be honest with you - if you think the average home has a 60A shower steamer (a specialty appliance), then we're never going to agree. Many homes (I'd guess 70% of homes built before 1970 and declining as the home age reduces) can't even add a 60A shower steamer without a service upgrade. A 40A load for car charging can barely be squeezed in to a good chunk of older homes without an upgrade.

    We're not talking about the average Roadster and Model S owner with the 3,000+ or more sq ft home. We're talking about the *average* home, which is a much different -- the ones who stretch to buy $40k cars instead of paying cash for $120k cars; the ones who have 1,500 sq ft instead of 3,000 or 6,000; the ones whose largest electric loads are A/C units. These homes have 100A panels - or maybe 200A with undersized service conductors like my home was, built in the early 1990's.

    Electric ranges are typically specified with a 40A or 50A breaker, and rarely draw more than 40A total. Even when "all the burners" are on, each "burner" is cycled as an intermittent load and the total aggregate - even for that thanksgiving dinner - is 30-40A at the absolute worst. Finally, keep in mind that these are intermittent loads, compared with continuous loads - A/C compressors, etc. can surge for a few moments without harming electrical components, but electric car charging draws a heavy load for a long, long time.

    The reality, though, is that "average" is nowhere near what you suggest and is much closer to my examples. I've been through hundreds of homes to see their electrical infrastructure, and I can tell you that the "average home" in any given area of the country generally has less than 20-30A instantaneous load. Adding a 40A charging load increases that by more than 100%. My father's home was built in 1884 and has a 100A service panel with all gas appliances in a ~2,000 sq ft home - the heaviest electrical load they have is when my mom runs a hair dryer while dad is running a table saw in the garage, and that just MAY get them to 15A across the 240V.

    For most power companies, there is no "normal grid upgrade plan". The majority of transformers out there stay on the poles until they cause some type of problem, then they're replaced to the new standard based on their new calculations. The deployment of significant numbers of continuous loads will accelerate that normal aging process (read: transformers blowing up) quite significantly. There are transformers on poles near me that have been rusting away since the 1980's.

    As for TOU rates, I had a lengthy discussion with my co-op president on the topic... outside of some low-generation, high-demand states the difference in the wholesale cost of power doesn't justify operating a TOU based scheme. In fact, my co-op used to have a TOU program but stripped it out back to a flat 8.5c/kWh rate.

    I stand by my original statement: it is not a garbage article, and it is not sensationalized. It is absolutely true that the current residential grid is not sized for significant numbers of electric cars charging at 10 kW, and the current model for "grid refresh" relies upon failure before capacity is increased.
     
  6. ItsNotAboutTheMoney

    ItsNotAboutTheMoney Active Member

    Joined:
    Jul 12, 2012
    Messages:
    4,503
    Location:
    Maine
    Actually several people (including me) have raised this idea of electric vehicles as a controllable demand sink. I call it G2V (grid to vehicle), which literally just means charging, but as a back-formation from V2G it's referring to the smart control. People have a "charge by" time so I think you'd end up with a lot of available power and capacity, especially if the long-range BEV wins out. While it'd take some significant hardware and software to crunch the numbers, since the rate the car can charge at depends on battery state, I think that fact the demand control is already used means it would be economically beneficial and therefore would work. This demand control could be combined with output and grid balance control to handle the communications lag.

    The article's main point about local demand is a valid one and it's why the distributor wants to know when somebody gets an electric vehicle with fast charging. They'll upgrade transformers as required. In fact, some utilities have already been upgrading local transformers based on predicted demand. PEV charging is unlike most other loads, with a large draw sustained for several hours. It's not any kind of catastrophic problem, just something that needs to be taken into account.
     
  7. Cottonwood

    Cottonwood Roadster#433, Model S#S37

    Joined:
    Feb 27, 2009
    Messages:
    5,062
    Location:
    Colorado
    I agree completely!

    Not only should EV owners who install an EVSE larger than X kW notify the local utility, there should be a procedure that the electrician should use to measure voltage drop with the new load, a good measure of how much strain the new EVSE is adding to the house and the neighborhood. We can debate whether the threshold value of X should be smaller than 15kW to include all those HPWC's, smaller than 10kW to include all those 14-50's, or smaller than 7.2 kW to include all those 30A J1772's, but that depends on how bad the problem is.

    Here is a start of a procedure for the installer and new EV owner:

    1. Pick a threshold, maybe 5%, as the maximum acceptable voltage drop between no EV load and full charge load.
    2. If we use 5% and 240V, this means that if the Voltage drop as displayed in the car, drops more than 12V between 0A and full Amperage, then look for where the drop is happening.
    3. Cut the problem in half, and check the voltage with charging off and on at the input side to the main panel or the service entrance to the house.
    4. If the big drop is on the house side, look for where the big drop is coming from in the house and/or check for poor connections and warm wires to verify that there are no potential fire hazards and that the EVSE was installed correctly.
    5. If the big drop is on the utility side, call your utility, or have your electrician call the utility and have them check the capacity of the local, neighborhood distribution.
    This or some variant should be standard procedure for an EVSE install. Notifying the utility will help with the multiple EV neighborhood, and the simple voltage drop test will alert EV owners and their utilities to the "you better fix this quick" problems.
     
  8. FlasherZ

    FlasherZ Sig Model S + Sig Model X + Model 3 Resv

    Joined:
    Jun 21, 2012
    Messages:
    7,019
    Y'all thinking like a bunch of engineers again - because we're all (mostly) a bunch of technology-lovers. Remember: AVERAGE CONSUMER as adoption increases.

    My mother doesn't give any attention whatsoever to voltage drop; she won't be pulling the cover off the panel to measure service voltage drop vs. circuit voltage drop, and she won't be wanting to deal with any of that crap. She's going to call an electrician (or, likely me) to install something she can charge from, and she's going to follow the instructions.

    Even if the Tesla would warn her about voltage drop, an electrician would get the call. If the car charges and nothing gets hot, blows circuit breakers, or starts on fire, the consumer may ignore it, saving the service call cost.

    I'm going to be the one to look over the infrastructure and likely I'm going to be the one calling the power company. In some areas of the country, they don't want to be bothered until something breaks (e.g., Incoming utility cable capacity - A cautionary tale - Page 2 )... in others, they'll come out for the curiosity factor as mine did.

    You're giving a lot of organizations and people very set in their ways far too much credit... :)
     
  9. AmpedRealtor

    AmpedRealtor Active Member

    Joined:
    Jun 30, 2013
    Messages:
    4,711
    Location:
    Buckeye, AZ
    A Model S 85 kWh battery is sufficient to power my home for one day. I use approximately 28,000 kWh per year, which is just shy of 80 kWh per day. So from that perspective, if you need to fully charge an 85 kWh battery, it takes as much power as my house consumes in a 24 hour period.
     
  10. FlasherZ

    FlasherZ Sig Model S + Sig Model X + Model 3 Resv

    Joined:
    Jun 21, 2012
    Messages:
    7,019
    #11 FlasherZ, Aug 17, 2013
    Last edited: Aug 17, 2013
    Yes, you must manage for average load and peak load, and the Tesla's load is not spread throughout the day. Very roughly, as a 40A charge takes 8 hours (1/3 of a day), that says if it consumes the same amount of energy you normally take in 24 hours, that while it's charging the car is drawing 3x your nominal power load, add those together and you're using 4x the power you normally do while the car is charging... Do it at 80A, and you'll draw 8x as much as your nominal load for the 4 hours the car is charging on the HPWC.

    Now, power consumption isn't evenly spaced, it's why we have TOU rates... Make some really rough consumption estimates - that your consumption during the day is 2:1 compared with night. So let's just say for the ease of argument that you use 53 kWh during a 12H period (during day) and 27 kWh at night (other 12H). Daytime average would be 4.4 kW (53/12), and adding the Tesla's 10 kW to the night-time average of 2.25 kW (27/12) means that you've experienced an increase of 178% versus your daytime average (12.25 kW is a 178% increase over 4.4 kW).

    Even if you boost to 4:1, you're still seeing a hefty increase: daytime average will be 5.33 kW, nighttime will be 1.33 kW. Add the 10 kW charging, and the 10.33 kW at night is still a 94% instantaneous increase over your average daytime load.

    Now I recognize we're mixing average and instantaneous, and that electricity "super-suckers" will have to define the boundaries... but as I mentioned, the article is not cherry-picking stats. Imagine just half the homes on an 8-12 home distribution transformer bank boosting their maximum by 100%, and they all fire up after work (non-TOU) or after midnight (TOU) for charging...
     
  11. mitch672

    mitch672 Active Member

    Joined:
    Jul 1, 2012
    Messages:
    1,868
    Location:
    Stoughton, MA
    I'll give you a real life example. I had my service upgraded from 150A (what the house was built with in 1996) to 200A.
    I built a 75A OpenEVSE, for use with my Model S with dual chargers, I am already "not the average consumer"...

    When I started charging at 75A, my voltage sag was huge, it would drop from 238V to 212-214V
    I called my utility, multiple times to try and talk to someone about the issue, I finally received a call back from their engineering department months later. I explained I have an Electric Vehicle, it can present a 16-20 KW continuous load, and that I have a voltage sag issue.. They eventually sent out a lineman, who took a voltage measurement at the meter while I was charging, they finally believed me that either their transformer was undersized, or the conductor to the house where undersized.
    They at first partially replaced the drop to my house (there are intermediate poles), this didn't help.
    2 weeks later, they set a new pole next to the existing pole, I was wondering what they where up to (the high voltage cables in my town are scheduled to be replaced this fall, I figured they where replacing old worn out poles ahead of that)

    That is partially the case, but one day on my way to work there where 3-4 trucks parked by my driveway, I pulled up with the Model S and of course they where all curious to see it, anyway, they tell me they are setting a new transformer today, just for my house, and my voltage issue should be solved... They ended up setting a 50KVA transformer, and it's only connected to my house (secondary wires only connect to my drop directly). I still am getting some voltage drop, from 238 down to 225V on the 75A load, but it's better than it was.. If they replaced that last 100' of undersized drop wire, sure it would be fine, but I don't want to push them too much, they have done quite a bit already.

    The point is, the electric utilities will respond and fix the issues, but it takes them a long time, and they are only reactive to problems, not proactive upgrading things.
     
  12. Norbert

    Norbert TSLA will win

    Joined:
    Oct 12, 2009
    Messages:
    4,424
    Location:
    San Francisco, CA
    What sometimes gets a bit lost, it seems to me, is that these are also additional sales for the utility, more money, not just additional loads as charity donations. It is very common for additional business to require additional investments. The utility should be happy about it. And not every new EV requires upgrades, only some percentage (not specified in the article, the article just mentions the more extreme cases).

    - - - Updated - - -

    Model S doesn't usually require 10 kW all night, though. For average driving, a much smaller charge rate is sufficient (and/or charging at different times, with multiple EVs in the same neighborhood). I'm sure that as EV numbers increase, software support (with and without grid communication) will get more and more refined, in this direction.
     
  13. FlasherZ

    FlasherZ Sig Model S + Sig Model X + Model 3 Resv

    Joined:
    Jun 21, 2012
    Messages:
    7,019
    Understood, but I'll bet if you look at every TOU car that is set to charge at a certain time, it starts right around 12:00 am, whether 12:00, 12:05, or 12:30... I'll bet that with few exceptions, most people leave the default charge current at 40A, and Tesla won't change it because it doesn't want to be the first party responsible for someone's car not being charged fully in the morning. So whether you *can* charge at less than 10 kW, or you *can* charge at a different time, doesn't mean it will happen that way. it will aggregate and bunch at certain times.

    Back to the original topic, though -- the article is not a hit piece, and is absolutely the truth.
     
  14. Cottonwood

    Cottonwood Roadster#433, Model S#S37

    Joined:
    Feb 27, 2009
    Messages:
    5,062
    Location:
    Colorado
    #15 Cottonwood, Aug 17, 2013
    Last edited: Aug 17, 2013
    It all varies by the utility, what info you provide, and who you know. BTW, this is not a new problem. Electricity use has been going up for the last century or more and the utilities have been steadily increasing their generating capacity and distribution to match. The biggest additional load (maybe not in San Francisco) has been the growth of air conditioner use. Now the typical air conditioner is not as big a load as an HPWC or even a 14-50 for charging, but air conditioner loads of 5 kW are not atypical, and are certainly correlated on hot summer afternoons. The utilities have responded and upgraded their systems, even if this has been almost always behind the power curve. :wink:

    Look, I know that I am an atypical user and a geek engineer, but at least I can make sure my houses have enough capacity.

    There may be cases where multiple EVs hit one transformer, but I bet most of those are in the same garage like mine (Roadster and MS). For grins, I just went out to the garage and checked Voltage drop. I live in two places. It happens that both the Roadster, the MS, and I are at my second home here in the front range of Colorado, using Excel Electric. This house shares a 50 kVA transformer with one other house, has 400A service (100 kVA by itself...) with a 200A subpanel in the garage dedicated to EVs. The Roadster has a 70A wall charger and the MS has a 40A charge rate from a 14-50. The MS started with 244V at 0A , which went to 237V at 40A. Adding the Roadster at 70A, the Roadster started at 242V at 0A and dropped to 236V at 70A; with this draw, the MS dropped to 235V. This all looks reasonable to me. BTW, that 110A load from both cars was 240V*110A=26kVA, or over half of the rated transformer load.

    Now let's talk about my main house in Pagosa Springs, CO. I did a remodel/addition a couple of years ago. Because the house is over a mile from my nearest neighbor, I have my own transformer. The original house had a TOU Electro Thermal Storage (ETS) heater with the main heating elements on 4-60A circuits, drawing the full 80% each. That heater required 400A service, 2-200A panels, and drew 46kW by itself. The only other big load was a steam unit that drew about 10kW. This was served by a 50 kVA transformer. I have no air conditioning in Pagosa, nor do I need it! In doing the addition, I added capacity for an HPWC along with two 14-50 outlets in the three car garage. In addition, the new master shower has a 15kW steam unit. My electrician and I put together a spreadsheet with all of this for the local electric Co-op, La Plata Electric. The local office has one engineer and my electrician knows him. When the La Plata engineer saw the loads, he immediately authorized the upgrade of my transformer from 50 kVA to 100 kVA; this swap was done during the construction. I now have 600A service, the two original 200A panels in the old section, and a 400A panel in the 3-car garage of the new section. I don't have any problem with Voltage drops there. :cool:

    The funny thing that I noticed was the difference in the melt/sink hole in the snow over the transformer. By February, I usually have 2-4 feet of snow on the ground in the area of the transformer. With the old 50 kVA transformer feeding the main load of the ETS heater, I usually could see the green top of the transformer a week or two after each snow storm. Even with the higher load of the addition, the new 100 kVA transformer usually stays covered with snow because of the lower heat dissipation of the bigger unit.

    - - - Updated - - -

    But it is the same truth that happened when the utilities were behind the power curve reacting to air conditioner installs. Nothing new, and market penetration of EVs will happen at a rate that the utilities can react to. Even if that is a little slow, remember these are glacial electric utilities, not nimble high-tech startups.
     
  15. CalDreamin

    CalDreamin Member

    Joined:
    Jun 12, 2013
    Messages:
    438
    Location:
    SF Bay Area, CA
    When I called to request an EV TOU rate through PG&E, they asked how many amps my EV charger would draw (80A HPWC) and how many amps my electrical panel is rated for (200A). Obviously the utility wants to know about these loads. If several of my neighbors got EVs and we all charged during the 12 am - 7 am off peak period, I suppose we might overload the transformer. I actually charge at 40A, there's no need for 80A charging for my normal driving.
     
  16. Norbert

    Norbert TSLA will win

    Joined:
    Oct 12, 2009
    Messages:
    4,424
    Location:
    San Francisco, CA
    I'll be setting mine to something like 2:37 am... just so it won't be my fault. ;)

    And it's probably just utilities talking, making a big fuss about routine investments for rising demand, to pre-emptively keep us from asking for lower EV rates. ;)

    Did the internet providers complain about having to upgrade their routers when thousands of new customers were asking for internet service?
     
  17. yobigd20

    yobigd20 Well-Known Member

    Joined:
    Oct 28, 2012
    Messages:
    5,793
    Location:
    Skaneateles, NY
    Well, while I can't speak about residential grids, I can about the effect on the national grid.

    I went to a lecture a couple months ago given by one of the energy industry's experts who studied this intensely. I asked him specifically what would happen if every single car in North America (roughly 240million) was electric instead of gas powered. He cited multiple research articles on exactly this topic (electric car effect on the North American energy grid). The answer was not was I expected. And to my surprise it seems that all experts agreed on this. If every single car on the road in North America (this includes Canada) was powered by electricity instead of gas TODAY, the amount of increased load on the power grid would be less than 2% and would barely make a dent in terms of additional energy consumption and have absolutely no adverse effect on energy supply and distribution network.

    However, that is in terms of the national supply (a.k.a. will we be able to get the energy we need? without a doubt. no problemo) as well as in terms of national power distribution through the power grid. The MIT article seems to agree on that point too.

    The question at hand is local neighborhood grid and whether or not the transformer on the telephone pole outside my house is rated high enough to handle me and all my neighbors all went out and switched to EVs tomorrow and plugged in and all started charging at 80 amps in the middle of the day simultaneously while also running all of our central air conditioners. the answer to that is unquestionably no - we'd just blow the fuse on that transformer and then the power company would need to come out and put a higher rated one in and possible new lines of the old ones are enough to handle the load safely. That's it.
     
  18. FlasherZ

    FlasherZ Sig Model S + Sig Model X + Model 3 Resv

    Joined:
    Jun 21, 2012
    Messages:
    7,019
    ...and if you look at what most power companies did there, they were reactive. Air conditioning units were installed, transformers blew up, PoCo came out and replaced the transformer with the next largest model, and it was fixed (at the inconvenience of many hot homes). :)

    - - - Updated - - -

    Speaking as a veteran of said timeframe and industry, the answer is yes, absolutely, unequivocally. And they did the same thing - they let them fail, then updated when necessary. Now, in some cases they had it even better, because they could do things like install BGP filters that allowed them to stretch the life of a router. You can't exactly do that with an electricity bus.

    - - - Updated - - -

    You're correct, on an average basis across the entire grid, there is negligible impact. Data centers and such dwarf this exercise.

    However, I wouldn't downplay the distribution networks into residential areas, and it's not just the transformers - in dense residential areas, it can mean additional distribution lines or larger lines.

    My point, though, is that the existing grid isn't ready for everyone to run out, get an electric car, and plug in. It will get there - one transformer failure at a time - at a significant convenience cost to a lot of neighborhoods. :)
     
  19. Norbert

    Norbert TSLA will win

    Joined:
    Oct 12, 2009
    Messages:
    4,424
    Location:
    San Francisco, CA
    Well, Elon also sort of complains about having two CEO jobs. What can we do about it?

    Welcome to the third millennium...
     

Share This Page