This is mahonj's Typepad Profile.
Join Typepad and start following mahonj's activity
Join Now!
Already a member? Sign In
Recent Activity
They are claiming 40% efficiency for the turbine which is as good as any gas and most diesel engines, but smaller and cleaner. It looks a bit overpowered for normal cars, but could be good for buses or vans. A good range extender would be a great boon to all electric cars, then you could size the batteries to the typical driving range, rather than some hypothetical "maximum range".
a: I wonder what is sounds like. b: If the turbine is in the middle, it might not be so noisy, but I wonder where they exhaust the gases.
@Change: Sometimes big companies just mess things up if they try to do it themselves and have to buy the small company who has done it successfully. Look at google buying youtube. Google had google video, but couldn't get it right and just bought youtube in the end. Big companies have too many committees and policies and people sticking their noses into things which can really mess things up. Small companies do not have that and can move very fast and clearly. Hence they get ahead ane get bought by the bigger ones. I do not think that every car company will make their own AV system. I would expect them to buy them from 3rd parties, such as google (or Tesla) in the way that phone people use Android. Google will set a sensor spec and everyone (except Tesla) will follow. All the medium sized companies anyhow. If you are interested in this stuff, consider reading "The Innovator's dilemma" by Clayton Christensen (it is well written, anyhow)
Mobileye do the software as well as the hardware. Intel desperately want to get into Automotive electronics and already know a fair bit about machine vision, so I would see it as a good fit. They get right in and can start to bring more features from software into hardware. The fact that they have sales of $0.5B in a hot field like this means they are a good bet. Intel might have overpaid, but that is the nature of these deals, especially if the target is a real player.
It should be very clear who is at fault in every AEV crash. There will be telemetry logs and image streams from the AEV, so it should be clear who was at fault. As long as it does not cost more to be killed by a machine than by a human, there should be no problem. Nice cash for Mobileye shareholders, I hope Intel make it work, their shares dropped 2.3% on the news, so the markets may not agree. I suppose they need to get off the desktop in a hurry, and missed the mobile phone boat.
If people were able to keep their diesels off in city traffic, that would be good - but a mild hybrid won;t be enough to do this. You would need a full hybrid, which would be too expensive [IMO]. Alternately, they could just implement proper NOX control.
If the mild hybrid can reduce the gms/km by 25, and if this pans out across the whole VW+AUdi range, this would be quite something. Like the beginning of the end of consumer diesel.
Strikes me as a good/essential idea. Imagine 2 cars: a Ford Focus with autonomous driving and a Merc E-Class without - which would you want ? Which would be more use ? Better get with the program ASAP or you will be delivering low value products before you know it.
Note: A petrol pump can dispense 38L / minute. 6 L of diesel can drive 100 km (easy). Thus, you can dispense > 600 km miles of travel / minute with diesel, a bit less (say 450 km / minute) for gasoline. If you could fill a PHEV or range extender, you could solve all these problems and still reduce gasoline usage 50-80% in normal use.
or PHEVs or EVs with gasoline or diesel range extenders. Then you can just refuel at the usual gas stations when you need to.
It is a bit like world peace - a very good idea that everyone can understand, just a little hard to implement. Trucks and buses could use overhead catenaries, but private cars couldn't reach. Cars could reach down to the road, or across to some side mounted rail, but both would be very expensive to fit and maintain. On the other hand, if you have fast charging and a battery (which you will have), you do not need continuous charging connections. One possible idea would be to have a truck height catenary with frequent charging zones every 5Km or so for cars; like mini service stations with a cafe and a bunch of charge points. Thus, you could charge every 1.5 or 2 hours of travel without having to make a big deal of it, or calculate of you could go the next 25Km to the next main charger station.
I hope Alphabet sues their asses off. This is clear industrial espionage. I wonder how much of this data "other nations" have ...
It all depends on whether you have access to another car. many families have two cars, and if one was an EV and the other not, the problem is solved. Similarly, you might be able to to have a swap arrangement with a neighbor/relative, or some deal with the company who sold you the EV.
I think we should give Hyundai some credit for their efficient electric approach. Electricity is not CO2 or pollution free, so you can't squander it as if it was. Obviously, many people would like more range occasionally, so the solution could be a fast charge network (as discussed), or a flexible car swap system - as follows. Lets say you want to drive 250 miles and have a 110 mile battery. You tell this to an app. It gives you 3 options: a: A guy comes to your house with a hybrid or ICE and swaps for you (you might pay for this). b: You go the garage and swap it yourself. c: You start the journey, drive up to (say) 90 miles in the EV and then pull into a garage and swap it for an ICE or a hybrid. You swap it back on the way back. (They would have to charge it for you so you got it back charged). The swap car could be another ioniq, or it could be any car. A would be the most convenient, but you might pay $20 for it. B: and C: would be very cheap, or free. You might pay a small amount / day of ICE usage, or you might have so many days free / year from Hyundai. or, if you expect high mileage, just get the hybrid at the start (58mpg is very good, IMO).
Most cameras run at about 25 frames / second. This is the approx update rate of the human eye. So you do not need to go at nanosecond update rates, unless you are monitoring atomic bombs or stuff like that. At 100 kph, or 27 m/sec, you move 27 mm every millisescond. That should be good enough for anyone. Even at 25 Hz, you are only moving 1 m / frame. If you up that to 100 Hz, you have 25 cms / frame, that should be enough for anyone to track typical road behavior (both good and bad).
OK, I am behind the times on that one. On the other hand, there are zero superchargers in Ireland (where I live).
@Change, I agree that wired is faster and more efficient. I am not sure you will get to 480Kw for cars, whatever about buses and trams - the cables and heat might be a problem. However, I do see a role for wireless charging for many home situations, not just for disabled people. It is so much handier, and most people don't care about energy efficiency, once something works.
It all looks good to me. The range of the EV is a bit short (by current standards) and the price of the PHEV is unknown, but let's see. WHat they should do for the EV is to offer to swap it for any other (used) Hyundai in the dealership for (say) up to 14 days / year, and offer a "valet swap" where a guy from the garage brings the Hybrid to your house, gives you the key and takes the EV back to the garage. Most people would never use this feature, but it would relieve people of range anxiety. + the fast charge is good. They could also have charge boost trucks which would have a large battery and a fast charger and could revive a dead battery on the side of the road. (The trucks could be diesel). This would be like the AA (or AAA) for the electric age. Either way, it is a decent try and we can only expect them to get better. The Koreans are very dogged and take the long view - they will do what is required to get it right and will absorb loses in the short term to get the product right.
@DSL You can see what Ford are pushing at. You can do 90% of your driving with a 10 KwH battery, 98% with a 100kwH battery, so the sweet spot is about 10KwH and some kind of petrol / diesel engine for the long runs. IMO, if you can cut your fuel usage 90%, you are more or less there, and you can do this with 1/6 - 1/10 the battery capacity of a BEV. The cost is complexity, and having two engines, but you could use a modified gas engine that really only has to act as a generator and make it simpler and lower emissions by restricting the revs range. But with complexity, once you solve it, it stays solved, so you can just use it - existing ICE engines and microprocessors are very complicated, but they are 10 a penny, and have been for decades.
It seems like a good idea that everyone should use data taken in a similar manner. This manner seems to be designed to promote PHEVs - Most of the daily drives can be done with a 30 mile range, but you get an upper tail on the distribution that can best be dealt with by a PHEV, or rapid charging. IMO, PHEVs are the way to go, but are expensive (requiring two full power engines) and complicated control software. A battery EV would be much simpler. as long as you can charge it quickly, or swap it for another vehicle. [ A neighbor of my mother's had a Nissan Leaf, and you could see ICE cars parked all around it. All he needed was the ability (agreement) to swap with an ICE for the occasional long run, and he would have been fine. ]
@lad, you may be right on the oil price, but I do not see "most of the World will be embracing clean energy and EVs" by 2020. We may well see a lot of solar and more wind, but we'll only see Evs in the more developed parts of the world, and even then (2020) in small numbers, certainly < 10% market share, probably < 5%. We might also see some in China due to worries over local pollution.
So, Volume in end 2018, might see cars in 2019 / 2020. As I have said before, the h/w cost of AVs eon't be too high once they get going. Much like the cameras and GPSs, gyros etc. in phones.
@All Here are some electric aircraft about to fly ... The 350 w/m^2 solar cells look like a breakthrough, as long as they are affordable, but i would expect multijunction solar cells to be expensive. Also, they are using 18650 batteries, so you could expect an improvement with the 2170 cells when they become available.