Posts by John Robson

SpaceX's Starship explodes again ... while still on the ground

Re: Return Journey

Why would they bring the robots back... a starship I can see the value of, but the robots can stay and carry on doing useful stuff...

AST just got a small boost in its D2C battle against Elon Musk's Starlink

Re: Cellular connectivity is a niche

Doesn't even have to be particularly rural US.

If you're best fixed line option is <1Mbit up, and maybe 15-20 down... then starlink starts to look really attractive!

Ship abandoned off Alaska after electric cars on board catch fire

Re: Its not popular for a reason

Because that's not what the standard was designed for.

But yes, it's a bit daft that it's a one way PWM signal, rather than an actual comms protocol

"The state of charge could be kept low, say 50%, but they are probably made and tested at over 80%, so that'd be quite a faff."

Not really, you just charge the batteries about to be tested using energy previously stored in those about to be transported...

"The reason no-one uses hydrogen unless they can connect generation and consumption with a pipe and a pump."

There are a couple of use cases... First stage of a rocket and second stage of a rocket....

Re: Li+ batteries

Not a false assumption - just an observation of grid demand and generation capacity.

The fact that one sector is a 24/7 operation doesn't mean that all demand is flat.

Re: Li+ batteries

Oh dear.. you seem to have completely failed to do any of the required calculations or reading.

Even if all the energy used to power my car came from burning coal - it would still be less polluting than an ICE vehicle.

And of course that's not actually how power is generated - but yes, we really do take that into account when we look at emissions.

You appear to have added a zero onto the "years before the embedded emissions are overtaken by the emissions of an ICE"... it's regularly analysed and comes out to a little under two years, not twenty.

Re: Li+ batteries

Well, mine is already doing that. Because smart tariffs exist, and they already do this.

Re: Li+ batteries

That's a smart grid...

Doesn't take more than dynamic pricing driving usage.

In terms of the communications... it's already settled in the most part. An EVSE can tell the vehicle connected to it the current capacity of the circuit it's connected to. That's enough for a charger to manipulate the charge rates throughout the day to use energy when it's appropriate. In my case, that's as determined by my electricity supplier, and just on my tariff... they had over 1GW of dispatchable demand a year ago, so I imagine it's quite a bit more now.

The alternative is to have the car do the communication, and that's entirely reasonable, but it's more difficult to guarantee connectivity.

I'd like to see an addition to the EVSE comms to allow something like the VIN to be transferred to enable billing integration (and obviously that needs appropriate cryptographic protection)...

Re: Li+ batteries

Absolutely - my point was that most cars are parked for ~23hrs a day.

The peak periods obviously have fewer cars plugged in, but there will still be plenty of them.

I'm absolutely with you on AC charging everywhere...

Re: Li+ batteries

"Which is more or less irrelevant because transfer loss is a significant part of consumption. 15% for that alone is more or less realistic and the losses to/from battery come on top of that."

That makes no sense. Overall grid transmission losses are at about 8%, and one of benefits of distributed generation/storage is that it's used locally. Even when I'm exporting energy to "the grid" it's actually being used by my neighbours.

Not only does that come with absolutely negligible transmission losses, but is also reduces losses that would have been incurred had their electricity been imported from a distant power station.

Grid storage batteries are being colocated with generation/import facilities, so that the energy doesn't have to navigate the grid multiple times. You could say that Dinorwig power has to go over the grid twice, and could suggest that means it would get double transmission losses - but even that's on the high voltage side, so the losses are more like 1.5% each way (i.e. you're adding 1.5% to the RTE losses when importing energy)

Re: Li+ batteries

"The other point EV fans keep making is that you have to stop anyway - well yes, but I prefer to stop at places where that is interesting, not because I have to play the mandatory game of find-where-they-placed-the-chargers-and-hopefully-one-is-free"

That's not been a thing for a long time.

I know before I stop that there is a charger free, and I stop at a place of my choosing... Generally it only applies to long journeys, so it's wherever is en route. There are hundreds along my most common route, but I tend to stick to just a small handful of those.

"You will also need a Godawful amount of chargers to support what you're proposing (being connected 23h)."

About 30-35 million... but remember these aren't expensive, or rare, they can literally be 13A sockets. There are *already* orders of magnitude more charge points than we could ever need... (though obviously not all the sockets in my house could be used at once for this, I'd be limited to two per ring circuit.)

Now, is it better in various places to have more than 2kW available - yes, of course... I don't expect every vehicle to be limited to 2kW all the time, but the suggestion that all chargers need to be 150+kW is also ridiculous.

"One of the key issues is that it takes time to charge a vehicle, and during that time the charger is not usable for anyone else. A petrol station can serve FAR more vehicles per hour than a charger."

But a petrol station also has to serve more vehicles per hour, because you can't fill up with petrol anywhere else, so they have to service all the needs of all the vehicles in the area.

Public DC charging is never going to be the dominant energy source for EVs.

The dominant source is going to be charging at home/work/stations/hotels/... and those can be supplied at 1.4 kW (the minimum) charging rate per space (in fact that's nearly twice what they'd need assuming uk averages etc)... the vehicle is going to be sat in those places for several hours at a time whilst the driver is off doing something else.

The secondary source will be shorter stops at retail/entertainment/leisure/hospitality/... where the vehicle will again be left alone for an extended period of time - but substantially shorter than the above. These would benefit from the 7-22kW range of AC charging, depending on the expected visit duration. Maybe hotels fit in here, with people obviously far enough from home to need a room overnight?

DC charging comes in a very distant third, and is primarily of interest for the second and subsequent legs of long journeys, which make up a very small fraction of journeys - so yes, we need much faster chargers and better grid connections on service stations on trunk roads, and whilst they do need to deal with peak vehicle throughput, it represents a small fraction of the energy needs of an EV.

The challenge is when we all decide to go on a long journey at the same time on the same day, but even on very busy days, I still don't play hunt the charger... because navigation software talks to the charger networks and knows which are free...

"What an EV takes away even more than mobility is choice"

Utter tosh, on both counts.

Re: 250kW, 320kW chargers

"No. Lower battery capacity means even more stops. It doesn't matter how short the stop is, it. is. a. stop."

Bladders mean stops as well.

Fatigue means stops.

Safe driving means stopping every so often - my car can easily outdrive me, and get charged faster than I can take a break.

All of your complaints there are about car safety tech, nothing to do with the vehicle being an EV, maybe you'd rather ride a bike?

Re: 250kW, 320kW chargers

Well if you want to count all the places a vehicle can charge then there are already tens of millions.

A 13A socket is all you actually need, at 10A that's well within it's thermal constraints and it supplies more than a couple of kW. Let's say it's down at 2kW charge rate, that's 2.5 hours of charging a day needed for the average car in the UK.

There are substantially more 13A sockets in the country than there are petrol stations.

Re: Li+ batteries

"I don't think there's now despatchable plant in the UK that could cover peak demand, we're reliant on wind, interconnectors, diesel farms and demand reduction. If sufficient interconnector capacity is unavailable, then you have to hope the wind keeps blowing. There have been some squeaky bum moments this winter eg 8 Jan, but we got through by paying around £20m in one afternoon to some industrial customers to cut energy use and last resort generators*, and having about 5-6GW of wind. If we'd had dunkelflaute around 8 Jan, then we'd probably have seen power cuts.

"

You'll hate to hear what we pay each year to turn down wind farms and fire up fossil fuel plants at the same time.

Regional pricing can't come soon enough - move some investment in power hungry industries (like computerised hallucinations) away from London, that would do the economy of the country no end of good.

Re: Li+ batteries

"For EV's, having TOU pricing in 15min intervals could help manage the grid. When prices dip due to lots of wind generation, EV's that can see the low price and are plugged in could switch on and take advantage of the low price."

We already have this in 30 minute intervals (since that's what the grid settlement period is).

Though actually - Octopus control my charging and it's rarely aligned with those 30 minute boundaries, it picks whatever times are best for them, whilst guaranteeing the charge I need.

I'd happily let them have the freedom to do less one night and more the next (or vice-versa), so that over a week or ten days I get the same charge, but "lumpier". As it is I use their "greener day" predictions to determine when to plug in.

This would benefit from having better communications between car and charger (see my other post earlier this evening).

Re: Li+ batteries

"If you can't charge at home, it's like all of the things you can't do to save money like you could if you were much more wealthy."

Early adopters of all new tech are inevitable the wealthier.

But the point is that you don't need to be able to charge at home to make it worthwhile, you just don't charge at DC chargers if you can possibly avoid it, because they are extortionately expensive, about as much as petrol...

Despite my feelings about the politics of a certain species of rat, his charging network provides the best value electrons - even to non members without that brand of vehicle.

Some of the other operators really are taking the mick... and particularly AC charging should be <45p (about to install a public charger at a commercial building, and that will be set at ~35p/kWh I reckon, possibly lower at off peak times if I can).

Home charging is a massive financial benefit, but all the other benefits of owning an EV still apply (and not having to go to petrol stations definitely counts in that list).

Re: Li+ batteries

Amazingly enough I have lived in places without dedicated parking, and indeed live on a road where there half the road is a terrace with on street parking.

Yes, these are the hardest people to have plugging in at home - but that's only one of the places to plug in, and they're in a minority. Moreover, that minority is more likely to not own car than the general population - whether through financial hardship or the fact that quite a bit of our densest housing is well served by public transport in cities.

You seem to ignore that I am continuously banging on about having AC chargers deployed everywhere.

Most journeys have one end at a car park of some sort - that's where you charge.

Ideally vehicles would be plugged in for as many of the 23 hours a day they aren't moving as is possible. Plug in at work, plug in at the shops, when you go out for a meal, when you go to the cinema, the theatre, whatever, wherever, whenever.

There are a few things we need to get right here...

- AC chargers should be cheaper than they are, they aren't particularly sophisticated bits of kit after all.

- EV's should be required to have bidirectional inverters.

- We need a robust integrated billing mechanism that's industry wide... Tesla do it, VAG do it with some networks, but it's not standard. Being able to just plug in, and have the car charge/discharge for profit is going to significantly increase participation in the market.

Let's assume this is a problem for 25% of the cars in the UK (which I suspect is generous). That's 71% away... it's an issue which we need to address, but it is absolutely not insurmountable.

You only need to "go somewhere special" on a long journey, when you're topping up en route. The rest of the time you charge at a destination - whatever shape that destination is.

Re: Li+ batteries

"To be able to not have to visit the petrol station most days of the year is great. I just got back from filling up my petrol tank..."

I agree - horrible smelly places petrol forecourts.

The advantage of basically never having to go anywhere special to top up is something that people who haven't made the move generally don't appreciate.

Re: 250kW, 320kW chargers

More and more manufacturers are realising that charge rate trumps battery capacity on long trips.

150kW+ is now pretty common - and if you're someone who does a lot of long trips, then it's obviously something you'd factor in when choosing a car.

350kW chargers are common - there are more chargers which supply 150kW or more than there are petrol stations (yes, I know one of those is chargers the other is locations) - that's actually more common than chargers between 50kW and 150kW (basically the rest of the DC chargers).

"Government should have made sure that there is sufficient charging capacity and convenience available before kicking off the aggressive EV mandates."

In the same way that petrol stations were mandated before cars needed them?

There is plenty of capacity at the moment, and as more EVs hit the roads that capacity is scaling (in fact its scaling faster than the EVs)

Re: Li+ batteries

70mph at 20kW is only 3.5m/kWh - have to get down to 2.8m/kWh to hit 25kW, which would still be over three hours (plenty of time between a break).

It'll be a while before I need to hit a motorway again to be able to monitor the usage, but the Enyaq is about as aerodynamic as a brick, so it might take a shade more, but I know it's not much more.

I also know that I can easily get 300 miles out of my car, though I have a fair bit of a-roads either side of the motorway on the main "long route" I do (though it does include going over all the hills in Cornwall and Devon).

More efficient shapes can easily do way better than that:

The various versions of the model 3 used between 206 and 265 Wh/m (3.8 - 4.9 m/kWh) on a bunch of 70mph tests (Out Of Spec Reviews: jumps straight to results, then a little method discussion, then the actual range). That'll take a hit in a storm, but there is plenty of wiggle room.

"I think most EVs would be lucky to make 160-180 miles."

That phrase rather suggests that it's not something you've actually any experience of.

Re: Li+ batteries

"However, because of charge/discharge losses are around 15% the grid needs to generate that much more energy than is time shifted"

Looking at my (home) battery efficiency I don't think the losses are that high, they are about 10% over reasonable time periods (in Q1 this year I put in 1.06MWh, and pulled out .94MWh, that's 12% losses in winter, last summer was under 5% losses)

"A further complication is that there's times when much of the car fleet will be in use or needing to charge up"

The vast majority of the car fleet won't need to charge very often... but there will be times when much of the fleet is in use, but with widespread AC chargers at places like train stations and other car parks along with the much higher proportion of people working at least partly from home that's not as much of an issue as it used to be. Of course there are also those who don't need to be out and about at peak hours, whether through retirement, shift work, more flexible work hours...

"so that the dispatchable capacity may be very low, and those periods of low availability are likely to coincide with times of high system demand"

That demand is what we shift to the times when generation is high... that's the point of having a smarter grid, we don't need to adapt generation to demand if that demand follows the generation.

Are you assuming that all cars are going to need a full charge every night? Because with a UK average milage of 20 miles a day many vehicles will easily do a couple of weeks between charges (note of course that what's more likely is repeated smaller charges as there is a surplus of energy on the grid, but on average they only need ~5kWh/day to operate as a car).

Re: Li+ batteries

Oh no, it's completely impossible to build new capacity. whatever will we do? < /s>

Built capacity has fallen along with demand, the likelihood is that demand is going to increase, and we'll be scaling up generation as that happens.

EVs are a great load for this, because they can act as a significant part of the grid balancing system.

Re: Li+ batteries

"Thus, an 8 hour journey can easily add two or more hours for charging, including the mandatory not-pleasure hunts for actual working devices. Personally, I think whoever came up with this without thinking it through should be hung by his testicles for a few days and then be convicted to never ever drive anything else but an EV for five years, because it's clear this was dreamt up by someone who either only uses a bicycle or is chauffeured to work and thus never had to face the consequences themselves for their ineptitude."

Well over five years of driving an EV exclusively and I've never had that much trouble.

I have had exactly one journey where I had to play hunt the charger, and that was in my first year with an EV, and entirely my fault. Additionally the charging infrastructure has been improving significantly over the last decade, and continues to improve at a very significant rate.

I've also never managed to add 2 hours onto an 8 hour journey. Let's be conservative and assume 20kW for 70mph... In 8 hours you'd use 160kWh. To replace *all* of that on a plain 150kW charger would take 1 hour, and you don't need to replace it all during the journey - and of course chargers go up to 350kW.

If you are doing long journeys then you probably want a vehicle with ~80kWh of battery, so that's only 80 you need to top up with en route, add another 20 for comfort, that's 100kWh of charging during your 8 hour drive.

But you're driving alot, so again you'll be getting a vehicle with a decent charge rate.. Various cars do over 250kW, the Taycan (fastest charge I'm currently aware of) does 320kW... CCS chargers are specced up to 350kW already.

That's 20 minutes of charging for your 8 hour journey, not two hours. And in an eight hour journey I'm pretty sure I'd need the loo at least once, quite probably more.

And I'd be looking at the regulations for breaks for HGV drivers, which would mandate 45 minutes of break for 4.5 hours of driving, so the sensible break is already twice as long as the charge.

Or for taxi drivers, a 30 minute break after 5.5 hours, or 45 minutes off in an 8.5 hour shift. So again - twice as long as the charge is taking.

Re: Since the Felicity Ace

""Smoke was initially seen emanating from a deck carrying electric vehicles."" is a pretty good circumstancial indication of the probable cause."

Hope you never have to serve on a Jury.

That's not evidence of anything other than there were EVs on that deck... there may well have been EVs on every deck, I have no idea.

But you really need to consider that correlation isn't causation, and you haven't even established correlation.

Re: Since the Felicity Ace

You certainly didn't earn a downvote from me - just clarifying a few thoughts.

Re: Tesla that burned so hot, it melted part of the road

"It's so hot it will separate water molecules into hydrogen(fuel) and oxygen(oxidizer) just adding to the mess."

So it must take the same amount of energy released by burning those two to split them, unless you're suggesting that the reaction is actually phosphorus burning with the water and releasing H2 as a byproduct (which can then react with the oxygen in whatever air is around)?

Re: Tesla that burned so hot, it melted part of the road

For certain definitions of bulk...

Not really - partly because that just means you have to ship the batteries separately, and partly because it means two assembly plants.

Re: Looking at the map...

"standard Mercator projection, which simply flattens the globe out,"

Literally the definition of a projection...

Mercator preserves angles and is therefore useful for navigation, but bloody awful for most other things you might want a map to show.

https://en.wikipedia.org/wiki/Mercator_projection

It's really interesting to look at a transverse mercator projection and compare.

The Gall-Peterson projection preserves area, but not angles, making it useless for navigation, but much more honest about relative sizes of countries.

https://en.wikipedia.org/wiki/Gall–Peters_projection

Re: Li+ batteries

There is no evidence to suggest it was caused by an EV battery.

As you say, they are pretty well protected.

So assuming 25knots, a thousand miles is ~40 hours, so what would have triggered a fire two days into a long voyage?

The same question can be asked of the other vehicles on board - maybe a vapour buildup?

In the UK our grid is providing less energy year on year, and has been for many years... this is because of energy savings on all sorts of trivial things, like light bulbs, fridges, TVs...

The amount it supplied twenty years ago is sufficiently more than it provided last year to run every car on the road as an EV.

As for long journeys - they're no more of a pain than any other vehicle. I have to stop when the people in the car need a break for the loo, food, drink, or for the dog to have a short walk.

In that stop the car gets topped up, and is ready to carry on before the people are. Now obviously people will come in talking about driving a thousand miles each way every weekend uphill both ways whilst towing a supertanker, and managing to do that without ever stopping for the loo - but that's not how the vast majority of driving works.

The important thing here is to have a vehicle that can charge at high rates, less important than the capacity.

And of course home/destination charging - that's the real benefit. Simple plain AC chargers at any location you spend more than half an hour... They don't cost much to install/run...

Re: Since the Felicity Ace

You think it was one burning car?

Rather suspect it was more than one by the time anyone got to lay eyes on it.

And note also that I didn't say the crew should have investigated and conclusively determined the exact source and cause of the fire before they called it in - I said that the cause hasn't been reported.

Re: Since the Felicity Ace

It was certainly a fire that included EVs... the cause is as yet unreported, so of course there is a part of the population who will confidently proclaim that the bogey man of the day is to blame.

It must have been an illegal immigrant in a small boat who used a brick from their council house to cause a heatpump to draw more power than an oil refinery and the EV spontaneously combusted in sympathy. < /s>

Re: Tesla that burned so hot, it melted part of the road

Nitpick - I believe it was CTF, Chlorine Tri-Flouride, but I'd have to get my copy back from the last person I lent it to to verify.

NASA to silence Voyager's social media accounts

Re: No need to duplicate

That's great.. for those who actively seek it out, but how about spending some money bringing that joy of discovery to people... running a few antisocial media feeds has got to be one of the cheapest education programmes in history.

But only if they study the maga bible, not the one that says "be nice to each other"

Forked-off Xlibre tells Wayland display protocol to DEI in a fire

Re: Code talks

> Not so good as it will small-p politicise what should be a straight forward technical project and ironically discourage participation.

I would have to disagree about that. Equity policies in the main are explicitly discriminatory. I'm all about outreach to encourage participation by any who are technically proficient enough to contribute. Your skin colour should not be a consideration.

Great - and that's what DEI policies are about as well.

Because inclusion isn't discriminatory, it's the opposite.

Equity isn't discriminatory, it's the opposite.

Diversity isn't discriminatory, it's the opposite.

Since skin colour (to pick on the one you chose) isn't relevant to proficiency we should expect that the pool of talented people approximately reflects the general population - we should be expending effort to see why we aren't seeing all sections of society (however you define those sections), because we're clearly missing many of the very best people...

Need for speed? CityFibre punts 5.5 Gbps symmetrical broadband at ISPs

Re: I don't

Ah - two of my cable runs are down at 100Mbps speeds - that's not great... Probably an "old termination" issue...

Re: I don't

Not sure my wiring would be up to any more...

Laced the house with cat5 (Maybe cat5e, I can't remember) twenty years ago, and haven't upgraded any of it since... despite basically all of the cables running outside for some of their run, none are externally rated - but they now can't be replaced because the building has had alot of work done around them.

Yeah - distance can cause it's own issues, I suspect your DSL in Edinburgh is performing well above average, but would the fibre beat DSL in the same place (either rural or city)?

I'm not exactly in a city, but a fairly large town, and only just down the road from the green box... and fibre (both OR and CF) latency is substantially better than my DSL ever managed.

But I'm lucky enough to have lots of choices.

And fibre latency has been *so* much better than DSL for me, both in terms of absolute latency and in terms of jitter.

My limit would be the hardware in the house, that's only at GigE speeds, and that's just plenty...

I and getting an upgrade in a few days from 500Mbps symmetric to 900Mbps symmetric, mostly because pricing is weird and it's cheaper.

At least I can just about justify a connection that's the same speed as my internal network... any faster is truly pointless (and so is the 900 frankly, but as I said, it's cheaper for reasons best left to bean counters)

60 years ago the US took its first walk in space with Gemini 4

Re: That wasn't floating

I've got a hunch that there are a couple of people who actively search out my posts and downvote them irrespective of content.

It's really quite flattering.

Re: That wasn't floating

Those weren't orbiting, they were just hanging there.

That wasn't floating

It was falling, with style...

MIT boffins claim liquid sodium battery could one day power aircraft while sucking up CO2

All that dihydrogen monoxide, it's a big killer...

(Yes there are bits of other stuff in there as well, but that's the biggest direct killer)

Re: Huge quantities of molten sodium flying through the air?

The twin towers make a fine example of what a little avgas can do.

Large amounts of energy released badly do bad things, irrespective of the storage medium.

Ex-Meta exec: Copyright consent obligation = end of AI biz

Re: The article has misunderstood...

Of course opt in is reasonable...

If I don't mind people using my work in AI training I'll register. Don't worry, the registry won't need to be very big.

One is education even if you then pass on the book to someone else to do the same.

The other is redistribution, copying and passing on sections of the book to multiple people.

They are different behaviours. You can pass on the one book you bought, you can't copy it and pass it around/sell it as if you own it.

"Any fool can run a successful business if they don’t have to obey the law."

Well there is this one guy... who might not be able to... Managed to bankrupt casinos six times...