It simply has never been worth it before to retool things in practice.
It's also electrically a lot more noisy. The limit of what is considered low volt is like 60v or something.
Motors, bulbs, relays.
Your motors and actuators operate directly on 48V (in fact, most actuators would prefer a higher voltage like 96V). That's really significant.
Microelectronics is effectively a "don't care" since everything is behind a regulator or a PHY.
Yeah, 48V tolerant switching regulators are going to be a bit more expensive until the volume gets rolling, but that problem solves itself while ethernet and CAN PHY chips are already 48V tolerant.
Or was there a pathological lock-in between suppliers and manufacturers that prevented even the most obvious innovations from happening if they required any amount of coordination?
Classic example of legacy automakers happy to maintain the status-quo because it is easiest, while the new kid on the block is pointing out that things don't have to be the way they are just because they are that way.
Why would they upgrade a system when they don’t need to? Tesla can be a front runner and companies like Bosch will start creating 48v parts as EVs become more popular for legacy auto to use.
Tesla worked closely with a supplier, did lots of its own research and development and continue to work with a supplier to make the technology better over time.
This was considered stupid at first. This is now very widely copied. First automakers from China copied it. And now others are copying it as well. Even Toyota (who everybody believes are some kind of gods of manufacturing) are copying this now.
If you have been doing something one way for long enough, going into a radically different direction is hard. Your whole workforce knows about welding steel sheets together, and nobody knows anything about casting. Your whole workforce knows about V12 and the ecosystem for that.
Its hard to fundamentally change how you do things when you are producing millions of vehicles a year.
>> S&P Global Mobility forecasts 15% to 20% of traditional body-in-white (BIW) stampings in 2030 may be at risk from these gigacastings. Underbody components typically comprise about 50% of a vehicle's BIW shell, and this soft underbelly is the target of gigacasting's focus.
Source: https://www.spglobal.com/mobility/en/research-analysis/gigac...
Obviously not a technical journal, nor did I double check anything.
What the numbers say 15-20% might be replaced by gigastampings (sidenote: I ak surprised Elon didn't rename Twitter to Giga-X). Which, ok, is a thing. But nowhere near the revolution people seem to believe. First, it has to happen. Second, others have to consider it better, read overall cheaper, and continue doing it. Then we can properly judge. Until then, it might as well be in the category of Teslas fully automated factories that ended with Elon doing pyjama parties with his workers on the shopfloor and ad-hoc tents.
And Toyota, as per the linked article, is not copying it, they are "eyeing" it. And believe itbor not, manufacturers are eyeing new production tech all the time.
And Toyotas reputation in car manufacturing is well earned.
The majority of cars sold then will be existing models with minor updates.
And car companies only upgrade their major platforms every 5-7 years. Many companies are currently planning their second (or first real) EV platform.
But by 2030 even the waste number of EV will be on first generation EV platforms still.
Companies have 100s of million invested in their current body lines. So unless you build a totally new factory you are not gone adopt this technology.
When Toyota came out with their production system it took decades for all the practices to become established.
Also, 20% of castings still means a lot of welding before a car body is made.
One thing regarding Toyotas production system so: it is all about management and processes and not about the actual machinery. And it share a lot of principles ranging as far back as Venice's Arsenal in the 16/17th century and, especially, WW2 mass manifavturing of, e.g., planes.
Anothet major difference: TPS is a proven methodology for longer than Tesla, let alone any production tech coming out of Tesla, exists as a company.
Using a higher voltage is not an innovation (it's an obvious change, and we've gone from 6V to 12V to, in some cases, 24V already) - rather it's just a slight efficiency improvement in largely non-critical systems, with not a lot of incentive to take on the cost of transition.
In a a personal ICE vehicle, the only real significant power was to the starter motor and from the generator, and the distance there was short so the copper didn't really matter and thus no one cared for 24V - unlike industry where you might have significant aux systems. With EVs, you have heat pumps and brake boosters on the auxillary power, so you now have a stronger driver for conversion.
Even within 12V, you'd get a larger weight reduction from not carrying an aux battery, and just feeding through a converter from the HV system.
If the aux supply is near or in the battery, leaving that connected would while the rest of HV is interrupted would probably not cause any notable increase in risk.
At the same time, vehicle fires have been caused by a shorted auxillary battery (I have personally experienced an entire industrial building burn down because the 12V battery in a parked ICE car shorted and went up in flames), so I imagine only having one battery to worry about is a risk reduction.
As far as i know, 60 V is the limit for alternating current. For DC it is 24 V.
https://paultan.org/2016/10/31/mercedes-benz-reveals-first-d...
This dynamic, where Tesla "announces" something that the rest of the industry has been doing for a while anyway, and a bunch of star-struck enthusiasts and stock manipulators tout it as an example of Tesla' "super smart innovation", is getting tiresome.
Except that BMW 5-series and 7-series from 2010 onward have had this option... They've done it in a smarter way using a plantery gear system to keep a physical steering wheel connection as a backup instead of going the cheap and less safe drive by wire way that Tesla put on the CyberTruck.
This is like when SpaceX landed a first stage and everybody was like 'DC-X did it SpaceX did nothing new'.
There is a reason lots of people, including experts are exited of what Tesla did here.
Previous 48V systems were only partial for good reason as well - traditional manufacturers have been using "mild hybrid" 48V systems since ~2001. Many of the electrical components on ICE vehicles are parasitic engine loads since they need power than can be provided with 12V, so e.g. the water pump and AC compressor have separate belts that always are 'robbing' the engine power regardless if they're needed or not. Adding that 48V system allows for the engine to be freed from the draw that those components require and adding some light regenerative braking is sufficient to keep the batteries supplying that system charged.
That these 48v systems aren't universal should provide some color on how successful / important the manufacturers had found them to be.
All sorts of people have all sorts of random ideas that never go anywhere until an industry "golden child" says it's the way forward. Without that effect the change can take much longer to happen.
48 volt has been 'on the radar' in automotive spaces for a couple decades.
If something draws 48 Watts of power, you can either supply it with 4 Amps @ 12 Volt, or 1 Amp @ 48 Volt.
It's the amps that determine how thick the wiring needs to be, so by lowering the amperage, you lower the amount of wiring needed.
I believe the only reason cars are 12 V is because that's a practical voltage to build lead-acid batteries at (which is actually closer to 14 V in practice). Early car electric systems were only 6 V back when the only accessories were the front lights and the horn, but as more stuff was added on cars moved to 12 V.
Power dissipated in the wire as heat is proportional to the square of the current. It's equal to the voltage drop times current (P = I * V). Voltage drop is current times resistance (V = I * R). So P = I^2 * R.
I know the older Honda CT110 postie bikes came in 6v in the old days. That was a kick start.
What exactly is magical about higher voltages that makes them suddenly able to carry more power across the exact same wire? I know you just said it's "lower amperage" but I don't understand how amperage can be the only thing dictating the required size of a wire.
For the other half, "why do amps determine wire size?" Amps are coulombs per second -- how much electrical stuff moves through a cross section in a second. The amount of heat produced is directly proportional to amperage (every bit of movement has a chance to whack into something and produce heat, and more movement per second or more seconds generates more heat). Wire size is ultimately constrained by being able to release that heat while still acting like a wire (not melting, not sublimating, still conducting, not setting anything else on fire, ...). The amount of heat released is proportional to surface area (proportional to the square root of diameter, but that difference doesn't matter a ton right now), and the amount of heat produced is proportional to amperage and approximately nothing else. The point where those two terms are equal is the limit for the system (engineers add in huge safety factors to account for conduit and insulation and other imperfections), above which your wire melts and below which it behaves reasonably. Higher amperages require higher surface area to release the extra heat (so much higher thicknesses), so a given amount of power (stuff you want your electrical system to do) requires lower amps or thicker wires to work appropriately, because you can achieve the same power by increasing the voltage on the same amperage, thereby solving your power problems without extra heat.
(RPM and voltage are in fact related in EVs as well: generally speaking a higher voltage motor will be able to produce more power for the same amount of space and materials, but mostly by spinning faster, not producting more torque. You then need to gear it down to be practical in a car. But the trend is that things that make higher voltages and higher RPMs possible (the technology in transistors, insulation, bearings) are already cheap or getting cheaper, while things that make higher currents and torques possible (generally more raw material like copper and steel) are staying the same or getting more expensive. So EVs in general are pushing to higher voltages for lots of things)
If the pressure coming from the hose is high, then you don't actually need a lot of water, and so you can use a narrow hose. On the other hand, if the pressure is low, you'll need to use a lot more water at once, so a wider hose.
It's the same thing for electrical components. If you'll excuse the mixed analogy, if you need to push an object with 480 watts of water to get it to move the speed you want it to, you can either do it with 48 volts of pressure with a hose 10 amps wide, or do it with 12 volts of pressure 40 amps wide.
It's very similar with electricy: water pressure is voltage and the rate of water delivery is power.
48V still lets you touch both conductors with dry hands and it's still very unlikely for you to be able to get any current to flow through you, you can't even feel it (although with safe working practices you'd only touch either positive or negative at one time, not both). Obviously it's very much not the case to be able to touch the 800V conductors safely in the traction systems in EVs, that side is extremely dangerous and requires extreme caution and safety procedures!
As long as the battery pack is still sealed and the interlock systems are undamaged and working correctly, it's no big deal. You follow the manufacturer's instructions to make it safe, then work on it like there's no voltage present.
If the battery pack is broken or a prototype or whatever, yeah, then you need to think things through carefully before doing them. And have a plan for what you're going to do when things go wrong.
Source: I work in electric aviation and work on battery packs.
If you design 100v computers, monitor, headlights, seat adjustment motors, window/windshield motors, pumps, etc. you could cut these 48v wires in half again but if you end up stepping everything down anyway it loses its utility.
The safety rules about voltage and the human body are often poorly overstated, as is the addage that 'its the amps that kill you, not the volts.' in reality it takes a whole lot of both [0].
As further research, here[1] is styropyro touching 2 contacts which have more current than the largest bolts of lightning (enough to almost instantly vaporize a crowbar) with his bare hands, but because it is 12V it is not able to pass through him.
60 volts is nothing UNLESS you are completely wet or sweaty.
"It is estimated at 150 ohms for completely wet skin (in water), 1000 ohms for sweaty skin, and 100,000 ohms to 500,000 ohms for dry skin."
Assuming a worst-case scenario with dry skin providing a resistance of 100,000 ohms, fatality becomes a possibility if the current exceeds 50 mAmp.
Therefore, the lethal voltage would be above 0.05 (50 mAmp)×100,000=5000 Volts. [1]
So if you are wet or sweaty, it could be 7.5V to 50V that gets dangerous.
So it makes sense why 60 volts is a safety threshold, especially for those that live in Florida or Arizona.
[1] https://www.scienceabc.com/humans/how-many-volts-amps-kill-y...
Ha, it’s always easy to tell when people have experience with these things.
Doorbell wiring is 48V. Go hold those in each hand and tell me how impossible it is to feel if your hands are dry.
You might be confused with POTS (landline phone service) which is 48V DC with all phones not in use.
From experience I'm not touching 60v, sweaty or dry as sand - that voltage hurts! And 48v is seriously uncomfortable. People have died from less DC voltage in industrial settings.
Following EN61010, the max safe DC voltage in laboratory equipment is 35v for wet locations. For a car, we ought to assume that being wet is a possibility.
---
The equation for power delivered by an ideal system is P = IV, or power (P) = current (I) * voltage (V). Power is measured in watts, and that is generally the overall number that matters, in terms of what you can run off of your system at the same time.
So to increase the wattage (power) of your system you can either increase the voltage or the current.
- Increasing the voltage of a system increases the amount of resistance it can "break through". In "danger to human" terms, our skin is generally not a great conductor, so voltages lower than 50V usually won't enter the body (read: vital organs) at all. Voltages above 50V will start to enter the body depending on conditions, which is when electricity becomes much more dangerous.
- However even if the voltage is high enough to enter the body, if the current (I) isn't very high it still won't be dangerous. Current is measured in Amperes (A), and the usual number at which a current inside the body becomes dangerous is above 30mA. 30mA can cause respiratory failure if it passes through the lungs, current as low as 100mA can cause cardiac arrest if it passes through the heart. In a car's electrical system, the currents we're operating with are definitely going to be over the 30mA threshold we just established, so we want to keep the voltage under 50V instead.
Anyway back to cars and ignoring danger to humans for a second. Resistance is the main thing you want to overcome when it comes to the efficiency of such a system. The equation for power loss is P = I^2 * R, or "power dissipated (as heat) is proportional to the square of the current times the resistance". So if you increase the current of your system (in order to deliver more watts) you will also increase the amount of power you lose as heat. You can decrease the loss by decreasing the Resistance.
The equation for resistance through a material is: R = ρ * (L/A), or Resistance = resistivity (an innate property of the material) * Length / Area. In other words, the longer your wire is the more current you will lose to resistance. But if you increase the cross-sectional area of your wire (A) by making the wires thicker, you decrease the resistance.
So in short: if you have a high current (amperage) system, you use thicker wires in order to ameliorate resistive heat loss. But you can alternatively just decrease the amperage to reduce your resistive heat loss, which means thinner (and therefore much lighter) wires. But then you need to increase the voltage of your system in order to offset the power you're losing by decreasing the current. If you increase the voltage by 4x, from 12V to to 48V, this keeps it under the human danger zone (of 50-60V) and means your wires can be up to 16x thinner, taking up less space and less weight. Having it be a nice multiple of the previous system (4x) should make upgrading the relevant circuits a little more straightforward as well.
Of course none of this really makes sense, for example you can't increase wire thickness to increase amps etc.
https://paultan.org/2016/10/31/mercedes-benz-reveals-first-d...
Elon is a serial confabulist and bullshit artist.
Nobody says this is the first time V48 has ever been used for anything. That literally just missing the point.
The point is that the whole architecture of the car will use PoE 48v.
And if you go and take apart these Mercedes cars you will see that the waste majority of all of the devices and the wiring use the exact same 12v architecture that all the ICE cars also use.
You're spreading bullshit with 2 seconds of effort.
Who? Because you can't mean Elon, not if your objection is to people spreading bullshit...
48V is a dangerous voltage. Better not touch any wire.
I work at an electronics store in Australia
This week I had a customer dumbfounded why his 3,500W inverter almost caught fire
He was using 8-gauge wire (~56A max)
He needed 00-gauge (almost 300A max)
If he’d have run the system at 48V instead of 12V he’d have saved $200+ per metre on cabling his inverter
My advice on voltage selection for a vehicle is once you’re needing more than a few metres of 0-gauge, you should possibly increase the voltage of the system
High current cabling means expensive ANL, mega or other high performance fuses or other expensive circuit protection methods
I strongly believe that 48V systems are inherently safer than 12V systems when it comes to the public doing DIY work
And the past 12 V systems are dead easy to work on. On older cars that is, as soon as you have electronics and bus systems DIY basically requires deep car electronics knowledge and skills.
I haven’t done a ton of work on newer cars, but at least adding wiring for a hitch on my Kona EV was just as easy as any wiring I’ve done on my old truck.
Is a job like adding amps and better speakers trickier?
Hence, one needs some good knowledge around that, what to connect where.
In the old days, well, just go directly to the main battery with your aftermarket wiring and Bob's your uncle.
And none of that, battery management, which equipment can be connected where, has anything to do with the voltage but rather with all the other stuff modern cars have (bus management, battery management, ECus,...) all of which might or might not accept voltage and amp fluctuations very well.
My 1982 Range Rover has all of three fuses, so electrically I'm fine! Except the mess previous owners did to the original wiring... No idea why it is so difficult to just make your own wiring for after market accesories instead of butchering the OEM one up beyond recognition. It works, so I don't touch it. I have no idea why it works so...
> I have no idea why it works so...
I know that feeling haha. Adding wiring is so much easier than ripping up existing stuff ...
So how is this done on cars? Dores each manufacturer have their own proprietary bus system (what i imagine), or is there a standard that everyone can use?
Protocols like CAN, LIN, SENT and XCP are all standardised point to point and bus protocols for automotive stuff
I personally think the standardisation is actually a bit limiting[1]
Just like SpaceX found massive cost savings in using networking tech over hard wired interconnects, I think there’s a good chance China or other manufacturers might catch on
[1]I think interconnect failure rates are so low out of factory it’s not unethical to begin making things more lean, obviously with sufficient human safety testing
But I’m not an expert, just a former engineer working in an electronics shop…
I'm sure a 48V bus would solve all those F1 errors too.
Not an employee, but a very happy reoccurring customer.
Didn’t think there’d be more than one of us on here
Such a wild west company to work for. I enjoy meeting the many very bright people who work for us before they inevitably move onto other things
In an industrial setting it can be done but you need an extremely well designed and made termination point for one system and an interconnect for another, I’ve seen customers using floating grounds (and negative voltages) to enforce it
If upgrading to 24V or 48V I recommend having a solid grasp on circuit protection first, and if you have concerns, pay the 30 minutes or so labour for an automotive electrician to inspect it
I love that they don’t keep releasing new products so they can properly support their existing (extremely well made) lineup
Look into data centre infrastructure if you want to find interesting brands and solutions
A lot of automotive/DIY marketed gear seems to be reinventing the wheel and charging a much higher premium for it
It’a “smart”, but there are probably reasons it hasn’t already changed everywhere else.
Yes, the legacy OEMs are constrained by their suppliers, who dictate what they can (not) do.
Basically it's just a more efficient way of doing things.
Elon has said it's not a revolution, or ground breaking in any way, it is simply a step improvement. Anyone could have done it. Tesla did it.
Audi has a mild-hybrid 48V (as do many others) that also runs some other things like active roll control and the power steering pump.
They still have a 12v system running infotainment, seat heaters, defrosters, powered seats, lights, etc. etc.
The move to 48V means much less power is lost (power loss = (IxI)xR ), cables can be much smaller, cables and terminals can be much lighter and much less copper can be used used.
Basically it's just a more efficient way of doing things.
The same thing that was wrong with vehicles that get 10mpg.
The same thing that was wrong with laptops that get so hot they burn your lap and the fan sounds like a jet engine.
The same thing that was wrong with incandescent bulbs getting hot instead of using that energy for light.
This is about making things more efficient.
In the long term it will safe money while while being more efficient. As cars get more and more electronics it becomes more and more relevant.
Its something everybody has understood for decades, but nobody had a long term enough look to make it happen. Car companies were struggling t show profit. And in the 2000s companies like GM literally went bust.
And yes its true, OEM can tell suppliers what to do. But if you want something that suppliers don't have at hand, your gone pay for development. And if you are say GM, to move to 48v you need to literally coordinate the work of 100ish suppliers to bring the product together.
And remember, these OEM since the 90s have outsourced the majority of all electronics devices and the waste majority of the software. Do they have the internal expertise to manage such a transition?
Look at over-the-air updates, its still not standard. And even those cars that can over-the-air update, that mostly only for some of the core main components. Lots of the supplier delivered parts can still not be upgraded like that.
And the car industry had plenty of troubles in the 2000s so its not surprising they didn't do stuff like that.
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.