Flutterby™! : EV Data

EV Data

2013-09-25 14:26:00.850027+00 by meuon 15 comments

I was on a "webcast" conference yesterday that included some interesting data from San Diego electric that hurt my head: Most (not all) of the electric vehicle owners with smart meters are apparently plugging in and starting a recharge cycle on their cars when they get home, even though the cars can be set to actually start charging later. The problem is this turns enough 5kW peak loads into a 9-12kW peak loads that extra generation and infrastructure is required. Ala natural gas generators (turbines), instead of the fixed load generation that would be running with some extra capacity after midnight.

These "green" minded individuals behavior is actually much less eco friendly than they think.

Got an electric vehicle? Charge it some random time after midnight, please.

[ related topics: Sports Pedal Power Conferences Bicycling ]

comments in ascending chronological order (reverse):

#Comment Re: made: 2013-09-25 15:00:47.436562+00 by: ebradway

# crontab -e

23 3 * * * ~/charge_electric_car.sh 0 8 * * * ~/set_random_charge_time_in_crontab.sh

#Comment Re: made: 2013-09-25 16:32:58.233455+00 by: Dan Lyke

Hmmm... I see a Tesla plugged in around here during the day. I hadn't thought about the obvious, that plugging in during peak usage was a bad idea...

#Comment Re: made: 2013-09-25 18:17:14.251667+00 by: ebradway

In theory, large numbers of electric vehicles plugged into the grid can act as a ballast/battery. But that's theory.

#Comment Re: made: 2013-09-25 19:14:57.838519+00 by: meuon [edit history]

Eric, I hear that a lot. that's the craziest thought on the planet EXCEPT for emergencies/disasters.

Why would I buy an expensive depreciating asset (miles and hours count on some EV) and use it to provide electricity to the grid at a cost that is probably less than my total cost of operation / batter cycle life, mileage, hours?
In most systems, it takes a second meter to measure what is fed back into the distribution grid, a phase synced inverter (expensive big inverter), and pole and feeder transformers that are efficient bi-directionally. Not all step down transformers work well as a step up transformer, on purpose. Switchgear, such as the Intellirupters EPB use have to be configured to allow flow both ways.
In emergencies, using an EV to power local lines can cause safety issues for field teams. Like the generator on my house, I have a transfer switch that drops my house and genset off the lines so that I don't cause issues upstream.
and I can rant for a while longer.. but it comes down to an idea that tree-huggering E.V. fanbois love and the engineers and technicians that actually make things work, hate. For example a lot of solar power generated on houses never actually goes anywhere, it just offsets your local usage. The infrastructure to use it effectively isn't a good ROI for the utility and people only do it for tax credits and/or subsidized high rates.

I would like to be able to use an EV for a backup power source for my house in emergencies.. if I have disconnected from the incoming main utility lines properly.

#Comment Re: made: 2013-09-25 19:32:45.051983+00 by: Larry Burton

And, unless there was one car per house on the local grid you might ride out a five minute power outage with all that capacity. Your alarm clock might not turn off but you won't be at work on time the next morning because your car will be discharged.

#Comment Re: made: 2013-09-25 21:39:36.50025+00 by: Dan Lyke

Yeah, it's gonna take a huge advance in batteries. Assuming you get inverters which can do the right thing (I've talked to an engineer from Enphase a bit, this is not an easy thing to do, operators are pushing back hard against house-based solar arrays using increasingly tight specs on how the inverters match, and to do decent load smoothing you have to have some pretty fine-grained low-latency control of all of those inverters), you're still dealing with batteries that have lifespans measured in cycles. And are expensive.

So the grid operator gets to cycle my batteries, and potentially leave me high and dry in the morning, in exchange for load smoothing? Not seeing it.

And then we start to think about what load smoothing means in a network of a few hundred thousand potential sources with varying latencies and ability to respond to commands, and... I'm seeing a nightmare of not just non-linear control systems, but the security nightmare of trying to protect that non-linear control system from bad actors.

#Comment Re: made: 2013-09-25 23:13:02.457763+00 by: meuon [edit history]

Re: "Bad Actors" that's what the "intelliruptors" are for, they switch/drop misbehaving segments of the distribution grid. Uusally the "bad actor" is a squirrel across the lines, or at least, was a squirrel. But more importantly, I think you understand the fallacy (or at least issues) of small local inputs affecting the local (neighborhood level) grid and upstream.

#Comment Re: made: 2013-09-25 23:40:37.212252+00 by: Dan Lyke

Yeah, I think my concern is that if you have a whole bunch of networked grid inputs that can be remotely controlled to load balance, screwing en-masse with that remote control is where the liability lies.

The net was a simple thing when we could call up the guys at the UTC operations center and say "hey, some dweeb in your computer room is trying to telnet in to arrakis, you wanna step in there and smack 'em?" and the ops there would say "sure!".

When that attack is coming from tens of thousands of exploited Windows boxes globally, it's a different scale, and you can't really react to it in the same way.

#Comment Re: made: 2013-09-26 20:58:34.058043+00 by: petronius

Howzabout a chip on the plug that cries out "Yo, you are plugging in your Tesla during peak power time. Set the timer and charge up later!"?

#Comment Re: made: 2013-09-26 21:00:52.273806+00 by: Dan Lyke

We need that for climate control and electric dryers, too. I suspect the stuff Meuon's been working with will be delivering that shortly for certain markets that... uh... aren't ours.

#Comment Re: made: 2013-09-27 01:24:43.030254+00 by: meuon

It's funny, in the developing world their use of electricity is so much more "self-controlled", and I'll warrant if you were paying a lot more per kWh at peak times of the day, you would be more self-controlling as well.

Rough numbers: residental rates in the USA are .10 per kWh, and are 5+ times that in many other places.

#Comment Re: made: 2013-09-30 02:23:08.685102+00 by: Dan Lyke

Brad Templeton has some thoughts on charging, including the note that the chargers at De Anza college are effectively $.55/kWh(!).

#Comment Re: made: 2013-09-30 14:06:02.771295+00 by: meuon [edit history]

You are paying for the kWh, the equipment, distribution, land area it occupies and the convienence: sounds like a deal to me.

#Comment Re: made: 2013-09-30 19:18:03.226863+00 by: Dan Lyke

True, but if that's the direction that fuel for EVs costs, it starts to change some of the economic calculation. It may very well be that delivering electricity to EVs costs 5x what it does to deliver it to homes, which means that EVs have to be even that much more efficient...

#Comment Re: made: 2013-09-30 19:56:32.893638+00 by: Larry Burton

When I moved to the Atlanta area, in 2002, I was surprised at all of the charging stations that were in various shopping area's parking lots. I never saw anyone charging their car or golf cart at one though and after a couple of years I no longer saw them. As best as I could tell these were all free charging stations. I'm beginning to see them pop back up again and, again, as best as I can tell, the one's I'm seeing appear to be free. I have yet to see anyone parked at one using it but I am seeing a number of Nissan Leafs and a few Teslas running around the roads these days.