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

241–250 of 504 posts

Re: Tesla shares 48V architecture with other automakers

#241

Post by xjlin0 »

GuB-42 wrote: post #196
sitkack wrote: post #2

TIL that 42V fizzled out https://en.wikipedia.org/wiki/42-volt_electrical_system I welcome the move to 48V, as one gets lots of cross over synergies from the 48V telco standard. https://www.servertech.com/blog/48vdc-power-and-the-backbone... https://www.st.com/content/dam/AME/2019/developers-conferenc... https://www.maximintegrated.com/content/dam/files/products/p... https://www.infineon.com/cms/en/applications/infor…

Are you telling us that 42 wasn't the answer after all?

It is, just be patient and the battery will be drained to 42V eventually.

Re: Tesla shares 48V architecture with other automakers

#242

Post by biaachmonkie »

bagels wrote: post #203
CharlesW wrote: post #172

Earlier quoted context omitted.

> Everyone knew castings were the way to go… The jury's still out on whether gigacasting is courageous, or just moves costs from production (which will not be passed to consumers) to repairs (which will). > Everyone knew that 48V would be a big benefit, but noone did it. Audi did in 2016 with the SQ7 TDI, no?

Car A costs $5k less to manufacture than car B because it uses castings. Car A would be wiser to build because you can get better margins, or lower prices. It'll take a lot longer for buyers to catch on to the fact that the car costs a lot more to repair.

The question is really if the castings result in more cars being written off as total loss. Any collision severe enough to damage them would likely be severe enough to also total out a car without them as well.

Re: Tesla shares 48V architecture with other automakers

#243

Post by Trex_Egg »

klysm wrote: post #240
MrVitaliy wrote: post #236

PoE over CAN bus moves more complexity from EE world into software world. While Tesla currently has best auto software, this is still a very foreign and a different way of thinking/managing for traditional auto manufacturing. This won't be just using a different set of cables, it's going to flip their engineering departments upside down.

Wouldn’t it be PoC instead of PoE?

Power over Ethernet (PoE) is a technique for delivering DC power to devices over copper Ethernet cabling, eliminating the need for separate power supplies and outlets.

Source: https://www.google.com/url?q=https://www.cisco.com/c/en/us/s...

Re: Tesla shares 48V architecture with other automakers

#244

Post by dreamcompiler »

tzs wrote: post #133
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.

Hold on a second...4x the voltage means you only need 1/4th the current. The power loss in the wire is current^2 x the wire's resistance, so 1/4th the current does indeed mean 1/16th the losses. But 1/4th the current means you can use a higher gauge wire. Looking at a table of wire gauge current capacities it looks like if your maximum current is 1/4th you can switch to wire with 1/4th the cross section. And resistan…

Good point and you're not wrong. One mitigating factor is that resistance isn't constant: In copper it increases with temperature, and I^2R causes temperature to go up, which causes resistance to go up, and you get something of a positive feedback loop. Temperature rise is one of the main reasons why standard wire gauges are specified the way they are.

If you use 4x higher voltage, the temperature rise is much less significant (16x less significant to be precise) so it becomes more or less a non-issue and you can treat resistance as a constant.

Re: Tesla shares 48V architecture with other automakers

#245

Post by 1970-01-01 »

dreamcompiler wrote: post #235
gleenn wrote: post #212

Earlier quoted context omitted.

My understanding was that nearly every production vehicle on the road is 12V and that Tesla and perhaps Audi as you mentioned are doing things differently. Who is using 400-800V systems? That sounds extremely dangerous.

Tesla uses 800v only to run the Cybertruck's drive motors. Every other electric load in the truck uses 48v. Both of these are much higher in the Cybertruck than all other Tesla models. All other Tesla models use 400v [IIRC] for the drive motors and 12v for everything else.

Newer builds are using 16v lithium batteries for low voltage.

Re: Tesla shares 48V architecture with other automakers

#246

Post by mmaunder »

Many comments here seem to be confusing the accessories circuit voltage with the electrical vehicle motor and drive system. The vehicle motors have a combined power output of 450KW which requires a high voltage 800V system or the wiring would be so heavy it would not be feasible.

This spec appears to be the accessories circuit with Tesla pushing for a new 48V standard. That’s super smart because it immediately means you need roughly a quarter of the metal in wiring that you needed previously. It’s cheaper, lighter and more environmentally friendly. And it enables the use of higher power devices. If you’ve ever had to wire your 1000W sub woofer amp directly to your battery you know what I’m talking about.

It’s worth noting that newer GA aircraft like Cessna 172 use a 24 volt system. So non 12V systems for accessory power isn’t a totally new idea. And older 172s have 12 volt systems. So the transition was made in that space anyway.

Re: Tesla shares 48V architecture with other automakers

#247

Post by sephamorr »

KaiserPro wrote: post #101
panick21_ wrote: post #35

I think the more important change that Tesla made is the change in the databus. Much higher performance ethernet. They changed the whole architecture of the car where there are now very few point to point connection, and its all essentially routed with a few major modules in each part of the car. If you look at current cars there are sometimes huge cable bundles, lots of individual cables for everything. Its a nightm…

> Much higher performance ethernet. Depends what performance you are after. Ethernet isn't rated for safety critical stuff. It doesn't provide mechanisms for packet loss detection, and in most cases is pretty shit at flow control. Ethernet is also shit for small sensors/actuators. There are lots of low bandwidth devices that need power and comms, ethernet isn't designed for that. having to route 2 pairs of cables to…

There has been a lot of work developing standards for reliable ethernet for industrial and automotive applications, resolving most of the issues you mention which apply to 100/1000Base-T. Tesla isn't using your garden variety ethernet or abominations like ethercat.

100/1000Base-T1 is intended to be used with PoDL (802.3bg/cu) and TSN (various 802.1Q)to result in reliable links with guaranteed latency and bandwidth properties. PHY power is a few hundred mw though, and star topologies are limiting to replace CAN/LIN nodes, that's what 10Base-T1S is for though (cheaper, bussed, lower power).

Re: Tesla shares 48V architecture with other automakers

#248

Post by jondwillis »

CamperBob2 wrote: post #177
wnevets wrote: post #148

Earlier quoted context omitted.

I don't know if I would call a self proclaimed content marketer a Journalist.

(Shrug) He called himself one, although admittedly in a former role before moving to Electrek. The point stands. People who DGAF about their jobs are going to lose them, and that's not a bad thing.

In before the ex-content marketers are breaking into your house to steal things to fence on Amazon…

Re: Tesla shares 48V architecture with other automakers

#249

Post by dreamcompiler »

thot_experiment wrote: post #110
amluto wrote: post #79

Earlier quoted context omitted.

I find 48V a lot less scary when cars are involved than 12V. Neither is particularly likely to electrocute me, but 48V comes with fuses that will trip at 1/4 the current, giving 1/16 the resistive heating if something shorts, which is a lot less likely to melt or ignite things. Also, the wires are much smaller :)

> 1/4 the current, 1/16 the resistive heating Really underappreciated safety aspect. The currents required for your average doodad at 48V leave you with a MUCH lower chance of unscheduled welding. Remember folks, everything is a fuse if you put enough current through it, as a rule of thumb it's good to keep "enough" pretty low.

"The job of the transistors is to protect the delicate fuses."

--old EE lore

Post reply on HN