Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

How electric trains work and why they make interesting sounds [video]

youtube.com

Post reply on HN
Print view View original

61–70 of 79 posts

Re: How electric trains work and why they make interesting sounds [video]

#61

Post by leoedin »

Animats wrote: post #58
rwmj wrote: post #27

Earlier quoted context omitted.

Photo on Wikipedia: https://en.wikipedia.org/wiki/Insulated-gate_bipolar_transis... Are they manufactured using regular silicon lithography? You'd think when making something so large you could just mix up the chemicals instead.

It's amazing how small those things are. That's a transistor which can switch four megawatts . A manual switch with the same rating is bigger. Here's a data sheet for a somewhat larger one.[1] The control signal is only 6 volts at 5 mA. It's incredible that this is physically possible. [1] https://www.infineon.com/dgdl/Infineon-FZ1500R45KL3_B5-DataS...

As an embedded engineer sometimes working on PWM, it's interesting to think that the origin of the gate drive signals is still just some sort of microcontroller timer running at 3.3V and a few mA. I'm sure there's a few layers of level translation and isolation between the micro and the 4 MW power transistor - but it's still indistinguishable at the digital level from a circuit driving a tiny motor on my desk.

Re: How electric trains work and why they make interesting sounds [video]

#62

Post by jillesvangurp »

jimmaswell wrote: post #18

I still find it amusing that most fossil fuel trains only burn it to power a generator, with the wheels run by electric motors. It makes perfect sense with the massive startup torque a train requires, something electric motors are much better at.

That's true for a lot of heavy equipment. A lot of ships use diesel generators to power their electric engines. Same for a lot of mining and construction equipment. Likewise, most hydrogen vehicles are basically a fuel cell powering a tiny battery powering an electric motor. Fuel cells aren't great at variable output so it's easier to just put a battery in between.

That also makes fully electrifying these things less off a technical challenge and more of a battery cost challenge. Mostly size and battery density isn't that big of a deal even. There's plenty of room in a large heavy vehicle to put some batteries. Even if they weigh a few tons it's not necessarily that big of a deal. Except of course if you need to go half way across the globe with a ship. But shorter range of up to 100-200 nautical miles if very doable now. A big mining truck has plenty of space and they are big and heavy anyway.

Re: How electric trains work and why they make interesting sounds [video]

#63

Post by rcxdude »

rwmj wrote: post #27
theideaofcoffee wrote: post #13

As much as I love the futuristic and sophisticated sounds that electric traction drives make as compared to their fossil-fueled counterparts, I do think that the driving and control systems that actually generate the PWM waveforms that are sent to the motors are even more interesting. Huge, massive IGBTs weighing a kilo+ each chopping up hundreds of thousands of joules/s of energy without destroying themselves, curre…

Photo on Wikipedia: https://en.wikipedia.org/wiki/Insulated-gate_bipolar_transis... Are they manufactured using regular silicon lithography? You'd think when making something so large you could just mix up the chemicals instead.

It's definitely lithography on silicon wafers, but a lot of the details are very different. For example, it's pretty common for power devices to be manufactured to be formed from the full thickness of the wafer, as opposed to just on the surface. And a single device tends to be formed of many repeating blocks, since you do need fine details still to get the performance (And a lot of the challenge is making sure each of those blocks is as similar as possible, so you don't get hot-spots).

(silicon lithography is more of a broad range of processes: digital vs analog vs power vs MEMS all have very different needs, and so a process designed for one will look very different to another, and one designed to allow multiple will be making compromises)

Re: How electric trains work and why they make interesting sounds [video]

#64

Post by elihu »

amluto wrote: post #47
elihu wrote: post #40

Earlier quoted context omitted.

Continuing to burn fossil fuels forever isn't really an option. At some point we'll effectively run out, but before then we'll have caused catastrophic climate change due to our CO2 emissions. Synthetic or plant-based fuels are plausible options, but synthetic fuels can't compare with battery EVs in terms of energy efficiency, and plant based fuels need crop land that's probably better used to grow food. Electric veh…

> synthetic fuels can't compare with battery EVs in terms of energy efficiency, This is almost meaningless. If you are turning renewable energy sources into either synthetic fuel or grid power, then efficiency is irrelevant. What matters is cost (of the whole system, including distribution of the energy) and emissions (burning even carbon-neutral methane or hydrogen isn’t quite zero emission). There are startups work…

Of course efficiency matters, because electrical energy is a bottleneck resource. Renewables may be cheap, but they aren't free and we have a lot of other things we need that energy for. If you can make a car go 4 miles with 1kwh of electrical input, that's much better than a car that can got 2 or 3 miles with the same input. Multiply that by the ~1.5 billion cars/trucks in the world presently and that's a huge difference.

Burning liquid fuel in a heat engine typically wastes about 2/3 of the chemical energy as heat. Hybrids do better, but there's only so much you can do.

I don't know what the state of the art for synthesizing liquid fuel is, but I assume there's some significant energy loss there too.

On the other hand, modern permanent-magnet electric motors can be around 95% efficient. Lithium ion batteries typically have coulombic efficiency better than 99%. Actual energy efficiency is a little less due to internal resistance (it takes a higher voltage to charge the battery than you get out when you drain it, so even if amp-hours in is almost equal to amp-hours out, watt hours might be a little different). Charging circuitry also tends to be pretty efficient. Battery-electric drive trains are already so close to optimal efficiency that there's very little that they can actually be improved on, and nothing else comes close.

The tire dust microplastics thing is a real problem, but it's not that much worse for EVs than other cars. Brake dust is much less of an issue on BEVs and hybrids, due to regenerative braking.

Personally I hope the idea of BEVs that haul around 800 pound batteries goes out of fashion (and it might if we could be bothered to electrify our major highways to make huge batteries largely unnecessary), and I also hope cars in general start to get smaller and less numerous. But I think cars are basically here to stay in some form, so they might as well be electric.

Re: How electric trains work and why they make interesting sounds [video]

#65

Post by fransje26 »

Animats wrote: post #12

Some OBB locomotives have their PWM systems tuned so that when they change frequency, they're always on a standard musical note. There's one example of that in the video. It's a cute feature, people like it, and it's probably all in software.

Explained with a bit of technical detail in an other comment on this post. [1]

[1] https://news.ycombinator.com/item?id=41807522

Re: How electric trains work and why they make interesting sounds [video]

#66

Post by rcxdude »

amelius wrote: post #15

Can't they filter out those sounds with some L's and C's?

Not very easily. Firstly, the power involved in these systems is massive, making any inductor or capacitor you put in very expensive (and bulky). Secondly, filtering it out may reduce the sound, but still will have some part of the circuit switching at that frequency, and with the power involved, some part of the inductor will see that and vibrate accordingly. Modern power electronics generally avoids this problem primarily by pushing the switching frequency higher than human hearing (which also has other advantages, like reducing the inductance and capacitance involved). Generally the higher-power the device, the physically bigger it is, and the slower you can switch it without melting it, so bigger systems are stuck at lower frequencies (this is something that's improving as technology develops: so there's the other factor that most train stock is quite old and the designs older still, meaning they switch even slower than your new laptop, even taking everything else into account).

Re: How electric trains work and why they make interesting sounds [video]

#67

Post by Angostura »

jkingsman wrote: post #10

Fascinating video! One of my favorite train vids is a Taurus starting up in frigid conditions. As the wheels slip and the PWM modulation on the individual motors varies to keep traction, it makes beautiful music. https://www.youtube.com/watch?v=Q_bF1Wb_pSE

Wow. Thanks for posting

Re: How electric trains work and why they make interesting sounds [video]

#69

Post by kgwgk »

rob74 wrote: post #59
asciii wrote: post #30

Earlier quoted context omitted.

That's really neat - it sounds like a violin. I can barely keep a tune so I'm wondering if anyone knows anything special about the melody

Detailed explanation (in German) here: https://de.wikipedia.org/wiki/Siemens_ES64U2#Ger%C3%A4uschen... > Google translation: When switching on from a standing start, a noise can be heard that is reminiscent of playing through a scale on a tenor saxophone. It is created in the three-phase motors by controlling the power converters. The noise is twice the clock frequency of the pulse inverters, which is gradually incre…

Made me think of https://en.wikipedia.org/wiki/Musical_road

Re: How electric trains work and why they make interesting sounds [video]

#70

Post by wiredfool »

gambiting wrote: post #24
Loughla wrote: post #23

Earlier quoted context omitted.

I genuinely believe hybrid cars will be the path once we get the itch for electric cars out of our system. You get all the good parts of each engine type, with less of the bad parts of the other type.

I have a PHEV and I genuienly believe this is the best combination of both, it's so practical with upsides of both drivetrains, but also let's not kid ourselves - PHEVs are vastly more complicated than either just plain ICE or EV cars, you have a lot more stuff that can go wrong. I deal with it by just buying extended warranty for as long as they let me lol, but it's definitely a concern with them.

I think a PHEV would be ideal for me -- Bimodal use, mostly a few km trips, and then a few 1000km trips/yr. The long trips are the ones where we wind up filling the car, so space inside/towing is important. (We've got a 7 seater, used to be a renault grand scenic, now a ford galaxy).

OTOH, one of the last times I rented a car, they 'upgraded' me to a PHEV, and I wasn't impressed. It was a tight fit with the kids + luggage, and we didn't even have the dog along. Fuel economy was well worse than plain old diesel, because it turns out when they give you a PHEV with 0km in the battery, it's not much of a hybrid. (This was a Volvo S60).

Post reply on HN