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

Tesla shares 48V architecture with other automakers

electrek.co

Post reply on HN
Print view View original

121–130 of 504 posts

Re: Tesla shares 48V architecture with other automakers

#121

Post by TheLoafOfBread »

KaiserPro wrote: post #111
yinser wrote: post #39

I liked the interview with auto engineer Sandy Munro discussing the change https://youtu.be/ADwGGEj8sqQ?si=qp6akvy1yyWPTYNe - moving the voltage up means you can drop current - increase the data rate by using ethernet and PoE - using ethernet and PoE means you don’t have to run one off wires to each device, they can share a bus which results in half the copper being used in a lower voltage car - moving the voltage up…

> using ethernet and PoE means you don’t have to run one off wires to each device, they can share a bus which results in half the copper being used in a lower voltage car Ethernet isn't a bus, its point to point. PoE over cat5/6 uses 4 pairs of UTP. so it might be used to join aggregate things together, but it won't be a bus. Yes, you can increase the datarate, but ethernet is fundamentally unreliable. So you'll need…

Ethernet can be used as a bus (see CSMA/CD), but if there are more than 2 nodes, performance of whole bus will go to complete shit and there is no guarantee that an ECU will transmit a single packet during its run, because that CD has no automatic arbitration, it is just random disconnection and try again. Not good for critical things like ABS. That's also whole reason why FlexRay was spawned, because even that FR is inflexible abomination of a protocol it actually guarantees that every ECU on the network will get a time window to transmit its own data.

Re: Tesla shares 48V architecture with other automakers

#122

Post by Kirby64 »

dragontamer wrote: post #40
firebat45 wrote: post #27

Earlier quoted context omitted.

How exactly do you define a negative voltage unless you are using some other voltage as a reference?

Label the power pin+ GND and the power pin- becomes -48V Voltages are all relative. It's like saying 'How do you get a height difference of 10 feet by digging?' Well, you dig and then label the initial level as +10 feet, and redefine the bottom of your hole to be ground.

No. In telco, the -48V is referenced against ground, like the physical ground. If you're isolated, you can do this. but they would still need to be referencing the 'ground' to something ... likely the negative side of the main battery pack.

The reason why -48V is used is because it is provided as a bias voltage to give wiring cathodic protection, to prevent corrosion of telecom infrastructure. If you used 48V, it would not work. You need a negative voltage referenced against ground.

Re: Tesla shares 48V architecture with other automakers

#123

Post by speedgoose »

stonogo wrote: post #94
contingencies wrote: post #82

Earlier quoted context omitted.

Compared to ICE vehicles, EVs are expensive and heavy Expensive maybe, IMHO not really, at least in China. Heavy .. this doesn't sound fair. Are you comparing a cherry-picked, heavy, full battery back EV with an empty tank ICE? Noting the EV has far more torque, and that the same tech is used in UAVs and in a ground vehicle you can arguably move the weight around (lower it) easier in an EV, this casual observer (not…

Vehicle weight also affects how much wear the roadways experience. I'm not sure "A Corolla weighs less than a truck" is relevant here, especially considering that the F-150 is the most popular vehicle in the US by sales number. Comparing things to the market leader is generally a useful metric.

Heavy trucks damage roads much more than cars. It depends on the weight but it’s exponential. The weight difference between an EV and an ICE of the same category is not a big concern to have in terms of road damage.

https://en.m.wikipedia.org/wiki/Fourth_power_law

Re: Tesla shares 48V architecture with other automakers

#124

Post by smileysteve »

Kirby64 wrote: post #100
bryanlarsen wrote: post #63

Earlier quoted context omitted.

No benefit? I've seen estimates of $1000 in reduced cost due to reduced copper wiring, and more importantly the labor required to string that wiring. Modern phones charge on 48V these days, so 48V parts are extremely common & cheap.

What modern phone charges at 48V? I'm not aware of any that charge at even 20V, outside of a couple of gimmick devices. No Samsung or Apple phone charges at anywhere close to that, that's for sure.

What modern phone charges at 12v?

Most phones charge at 5v. Modern USB-C chargers can charge between 5v and 20v based on configuration.

Buck converts and regulators are cheap and small these days.

Re: Tesla shares 48V architecture with other automakers

#125

Post by rfdonnelly »

contingencies wrote: post #84
jnsaff2 wrote: post #83

The "expert" in the article got it wrong. The relationship between the voltage and the losses in the wires is not proportional. They are squared. Going from 12V to 48V is not a saving of 4x rather 16x. For some applications you could also consider Power over Ethernet in the car, get both shielded comms and power. Or can and power over twisted pair.

The great thing about PoE standards is there's so many to choose from. https://en.wikipedia.org/wiki/Power_over_Ethernet#Standards_... Honestly, ethernet introduces a degree of non-determinism with respect to time in the link layer, plus increased bringup times, a potentially more costly core switching fabric, and the need for critical revision of latency assumptions on any potentially safety-related control concerns…

Regarding deterministic latency, the Time-Sensitive Networking (TSN) [1] set of IEEE standards address this. The IEEE P802.1DG project [2] in particular defines a TSN profile for automotive.

[1] https://en.wikipedia.org/wiki/Time-Sensitive_Networking [2] https://1.ieee802.org/tsn/802-1dg/

Re: Tesla shares 48V architecture with other automakers

#126

Post by MetaWhirledPeas »

simplypeter wrote: post #95

The thing is, unless the whole industry moves together to 48V, the cost of this change for a single OEM+Tier1 would be too big.

In cases where they're using a proprietary 48v part I wonder if Tesla would consider becoming a parts supplier to other manufacturers?

Re: Tesla shares 48V architecture with other automakers

#128

Post by hinkley »

magicalhippo wrote: post #53
firebat45 wrote: post #27

Earlier quoted context omitted.

How exactly do you define a negative voltage unless you are using some other voltage as a reference?

It's a matter of perspective. You tie one of the leads to earth (literally grounding it)[1], leaving the other non-grounded. Depending on if you tie the negative or the positive lead to ground, you get 48V or -48V with respect to ground. As long as the potential between the most positive lead and the least positive lead is 48V, the circuit itself doesn't care. As mentioned here[2], the reason for grounding the positi…

Apparently in cars it’s weirder. Wire it one way and the wiring corrodes. Go the other way and the body corrodes.

Re: Tesla shares 48V architecture with other automakers

#129

Post by cogman10 »

mcguire wrote: post #81
myself248 wrote: post #60

Earlier quoted context omitted.

Hi, automotive electrical is my job. There's quite a bit of very thick wiring in a car, not just the starter wire, but boring stuff like audio amplifiers, rear window defrosters, power seat motors. Those things don't draw a ton of power, like maybe just a few hundred watts, but at 12 volts even modest powers require extraordinarily thick wires, especially when you account for bundle derating. This requires large term…

The rumor I heard was that the higher voltage resulted in lower switch lifetimes. Any truth to that?

Yes, but not meaningfully. The higher the voltage you get, the more arching there is when a relay trips (also depends on if there's any sort of inductive load, think the sparks you see when you unplug a vacuum without turning it off).

But when you think about the impact that has on switches and relays, realize that in your own home you have 120V controlled by switches. Very cheap switches last decades (though admittedly not switched as often as something like a blinker).

Re: Tesla shares 48V architecture with other automakers

#130

Post by hinkley »

myself248 wrote: post #119
hinkley wrote: post #117

Earlier quoted context omitted.

Doubling again to 96 is not safe, however. I’m not sure why they settled on powers of two. Something perhaps to do with noise filtration, and fewer new tricks to learn?

It's convenient to build things in multiples and powers of 2 and 3. A nice two-tier or four-tier battery rack full of 2-volt cells just works out nicely.

Smaller transformers with whole numbers, right? But why 4 and not 3 or 6?
Post reply on HN