Hacker News Reader

Comment by tomsto | Hacker News Reader

I don’t know whether on-street parking is as common in the US as it is in Europe, but the solution being quite widely adopted in the UK is to utilise the “leftover” current from when street lighting was converted from sodium to LED to retrofit EV chargers to the lamppost. It works really well up to about 7 kW I think. This is then supplemented with inner city rapid chargers at 22-50 kW.

Not as cheap as charging from the wall but (in London at least) other incentives like free on street parking and no congestion charge more than make up for it.

View on HN

Any actual detail?

A "normal " sodium street lamp is (was) usually 250 W while a "modern" LED lamp is usually 50-70 W, and typically posts are 25-30 meters apart.

So, you have (roughly) 200 W "excess" per post, you need 35 lamps to leds replacement to make 7 kW and this would mean 35 posts at 30 meters each = roughly 1 km in length.

How many (electric) cars will be parked along this 1 Km stretch of a street? I would say as much as 200 (once the conversion to electric will be 100%).

But even supposing that each lamp post can actually house several 7 kW outlets, enough to cover the 30 m stretch, let's say 6 outlets as each car will take 5 meters along the road 5x6=30, the cable for the single car (or at the most two) parked right in front of the post will be a "normal" length but to get to the farthest ones you will need a 20 m cable, not exactly handy to carry around or to store.

And anyway you are now at around 6x7=42 kW of power per post (as opposed to the original 250 W) and - per kilometer, one side of the street only - you are around 35x42= 1470 kW (as opposed to the "current" 35x0.25=8.75 kW.

You're right, I've done a bit more research into this. The company in question (Ubitricity) are uprating lampposts to 25A single phase at 240V, so 6 kW per post. Even though sodium (and earlier flourescent) lighting never used anything like that much current, I can't find a source to confirm whether or not the lampposts were originally rated at 25A or lower.

On your calculations of how many cars can charge at once, I would keep it simpler and say that a row of parallel parking typical in UK cities, there is one street lamp for every 5 cars. 50 kWh charge overnight, approx 300 km range for every fifth car.

Supposing 100% EV adoption - well, already not everyone drives themselves to work/kids to school in UK cities (where most on-street parking takes place) - but if they do, and have charging at work plus some nearby rapid charging, that should do the trick, assuming a typical commute by road isn't more than 50 km each way. It's certainly in the ballpark.

As I see it, as long as the amount of (electric) cars will be a fraction of the total, it can be managed, but as soon as the amount will increase this "lamp post" approach won't work and something different will be needed.

A practical example is the following:

1) you use 20% battery per day

2) you get home at 20:00, with battery at 60%

3) you find the "right spot" near the lamp post free

Now, do you "risk" postponing the recharge one or two days or you "fill it up to the brim" right now?

And, then, let's say that after 4 hour recharge you will be at 100%, will you at 24:00 unplug and move the car (to allow someone else to recharge)?

Or will it be possible (at 24:00) for the owner of the car parked right after yours to unplug your car and connect his/her own?

And why should he/she wait until 24:00, and not unplug yours when he/she arrives at 20:15?

Most probably, at least when the numbers are small, there may be some form of "rechargiquette" but I doubt it would last for long.

The question is whether it really is the right answer to invest enormous sums of money and infrastructure into enabling rich people to have electric vehicles, or does it make more sense to just... not. Don't upgrade onstreet parking, take the money and build a bus service, a train service etc. Cars are fundamentally a non-scalable solution for dense urban areas. They're just really inefficient and if you're living in a city where you can only get onstreet parking you're probably better of taking a bus or a train to get around anyway. It's also a massive bonus for everyone between you and where you want to commute that they don't have to deal with your traffic anymore.

London has already made this decision - they go out of their way to make it uneconomical to drive in London. Why suddenly unlearn everything about public transport.

That's fair. If self-driving cars pan out, then ideally, we'll soon be able to live in a world where we can all ride in electric cars without owning one. These self-driving electric taxis can then charge at fast chargers on lots somewhere out of sight.

Apart from that, why invest in charging infrastructure? Probably because, even though we've made it inconvenient to drive, some people insist on owning a car in the city. Maybe they are just being idiots, or maybe they need it for their work/business. If we don't make it more convenient to go electric, they will just stick with gas cars. Every little bit counts.

Trains, yes sure, but people love the convenience of a door-to-door service. I think Elon's idea of electric robotaxis and tunnels to alleviate traffic is not a bad one. It could work. We could also eventually have hyperloop "pods" that route themselves in a network tunnel underground. This would make it possible to avoid having to transfer from one train to another multiple times. Sounds sci-fi? Sure, but without a vision, there's no innovation.

As I understand it, the UK's original street lighting served the dual purpose of absorbing the night-time production of coal power plants, which are hard to throttle. Now that we barely use coal, and instead regularly throttle our gas plants, there isn't the "leftover" current that you think.

I interpreted it as having the infrastructure to deliver ~7kW to the kerb, which is a hangover from the previous lighting systems.

As parts and systems (cabling, transformers, etc) get replaced, I have no idea if the new parts will maintain this "excess current" (assuming OP really was talking about that), but even if they do, the power still has to some from somewhere.

Reply on news.ycombinator.com