Most (all?) hybrids and BEVs have a 12V (or 16V) battery. Why?

[VectorBear]

For standby (car powered down), why not have a low quiescent power traction voltage to 12V converter to power standby peripherals such as RFID key fob detector, security system, door locks, computer when the car initially powers up, etc. Once the car does power up, immediately the traction contactor closes and turns on the larger (100-200W?) converter to not only power all of the running 12V system but charge that 12V battery.
Usual arguments are silly: no isolation of traction voltages (converter can be before contactor, with its own protection), drain on traction pack (50-100kWh can handle a few W for a very long time), safety if traction pack fails (but what if 12V system fails in a collision? in the same boat), status quo AKA legacy (time for a new status quo).
So why not replace the 12V battery with a low power, low quiescent power traction voltage to 12V converter to run standby systems?

 

[3DDave]

It seems like they do; they use a converter to power the 12V system when the car is in use and periodically when the car is parked to recharge the 12V battery.

The main thing that separation is for is in the event of a crash, so that the 400-800V is isolated from the rest of the car while the 12V can still operate things like lights and emergency flashers, maybe a phone to contact emergency services all without the chance of any rescuers finding an always-connected electrical system has failed and energized parts of the car to lethal voltages.

 

[VectorBear]

It seems like they do; they use a converter to power the 12V system when the car is in use and periodically when the car is parked to recharge the 12V battery.

The main thing that separation is for is in the event of a crash, so that the 400-800V is isolated from the rest of the car while the 12V can still operate things like lights and emergency flashers, maybe a phone to contact emergency services all without the chance of any rescuers finding an always-connected electrical system has failed and energized parts of the car to lethal voltages.

In a collision, the traction contactor can and should turn off, so no HV outside of the pack. Quite safe, simple and reasonable. However the HV pack still exists intact or you have other major problems such as a mushroom cloud forming over your car. Why can't the traction pack still power those emergency ancillaries, safely stealing power before the contactor? This assumes that the 12V wiring is intact, reasonable assumption, circuit protections pop if shorted. See the attached sketch.
Not previously mentioned, the batteryless 12V buss can still be jumped from another car, but I don't know what use that would be.
Side note, years ago I was a witness to a collision where a 12V car battery passed us on the road after the cars abruptly stopped. Impaired driver hit another car, I was a witness in court, convicted, also the bartender who didn't know when to stop serving.

 

[BrianPetersen]

The high voltage to 12V DC-to-DC converter is outside the propulsion battery enclosure. Breaking the safety circuit by any means, including unplugging the high-voltage disconnect or unplugging the control wiring to the battery or cutting the firefighter emergency loop, opens the high-voltage contactors and completely stops ALL high voltage in the car external to the propulsion battery enclosure.

"So then put the DC-to-DC converter inside the propulsion battery enclosure and power it from before the main power contactors" then you're not able to completely isolate and lock out high-voltage power because you would have to maintain it to the DC-to-DC converter, even though it's inside the enclosure ... a technical failure (e.g. +400V shorted to +12V through the converter) might not be possible to be completely excluded. And if something ever went wrong with that gizmo, you would have to remove the propulsion battery from the vehicle and open its enclosure in order to service it.

There is a process called DFMEA that has to be gone through in order to minimise risk. You can be certain that for all production vehicles produced by safety-conscious and reputation-conscious manufacturers, it has been gone through.

 

[VectorBear]

That is the definitive explanation, explains well. The reasons seem mostly safety related such as firefighters knowing the vehicle is as dead as possible, and for the worst case of HV reaching the 12V. Not to diminish the latter, there is already a HV to 12v converter and that one can fail anyway - water/corrosion, foreign object, etc. Regarding the proposed low power converter, it does not have to be inside the pack, it can be located adjacent to the HV contactor with its own overcurrent protection taken from before the contactor.
I certainly appreciate the quality and level of engineering in the automotive industry. Amazingly safe for incredibly low cost (I worked in low volumes, single unit to maybe 100's). I am waiting for level 5 truly self-driving cars where the organics don't drive, much safer.
Background: After sitting two weeks, our Prius' 12V battery had enough charge to power on the computer for a few seconds as the CPU crashed! The display flashed garbage, the HV came on for a second, things blinked until I quickly hit Power and it did shut off. Doesn't the central CPU have a watchdog and POR (Power On Reset) to prevent software crashes? Watchdogs are standard on any self-respecting processor, I always put in PORs to stop the CPU if the CPU supply voltage is too low.
FYI, my background is electronic design in heavy industry (steel, offshore oil, mining), telecom (long haul, high speed fibre) and research. The rule of thumb for bad things happening is that if it is possible then design to handle it, and if people are involved it definitely will happen. My work was [mostly] not for dangerous conditions such as cars, my failure analysis was most for life span, the "bathtub curve".

 

[Eufalconimorph]

The 12V system is already powered primarily by a DC-DC converter from the traction battery, there's just a 12V battery to provide power when the traction battery is disconnected or otherwise inoperable (discharged into lockout, overheated, etc). That allows powering things like the hazard lights in the event of a breakdown.

The better question is "why 12V?" and that's almost entirely because the OBDII connector *must* legally use 12V. So you'd need some sort of stable battery tech that works in all sorts of temperature ranges (LiFePO4 doesn't work well below freezing) and lead-acid is very cheap and easily available. 12V lead-acid batteries are cheaper than 24V or 48V ones, even though it'd be more efficient to use 24V or 48V. Some heavy-duty (non-OBDII) vehicles are already 24V even at the diagnostic connector (usually J1939 green 9-pin for US 24V vehicles), and a few vehicles like newer Tesla Cybertrucks use 48V internally for their low-voltage bus but convert to 12V for the diagnostic connector. Higher voltage allows for thinner wires to transfer the same power, which saves on copper and lets them be more flexible.

12V is mostly a "legacy" voltage these days, it's not high enough for good efficiency like 48V, it's not the single-cell voltage of any battery technology (lead-acid cells put out about 2V nominal, most lithium chemistries around 3.7-4.1V, etc.), it's not a nice voltage for transistors or MOSFETs, but it was a good voltage for vacuum tube heaters & mechanical telegraph keys, so it's still quite common. Just about everything uses a DC-DC converter to lower the 12V down to something useful, with the rise of LED lamps even the hazard lights & headlights don't run directly at 12V (they have to regulate the voltage down to not waste a ton of power as heat in a resistor).

 

[VectorBear]

I now fully agree with a seperate small battery for, if nothing else, a "UPS" for emergency braking, flashers, etc.

WRT the OBD2 requiring 12V because the lawyers say so, if the main supply is 48V, a small POL converter can efficiently supply that small load, no issues.

Way back when, British cars had -6V power(?). Soon after came +12V. As has been discussed in the electronics media for the past 20+ years, vehicles will soon use +48V. The 48V is reasonably shock safe and is below the 50V threshold for many country's requirement for compliance with their main electrical codes. For equal power transfer, current drops inversely with voltage increase. Therefore for similar wiring power losses, the wire cross sectional area drops by the square of voltage increase. So 12V to 48V needs 1/16th the copper. To accomodate 12V legacy devices there can be POL converters with a 12V and a 48V connector, use the one that fits.

LEDs can be driven two ways: low power (e.g. indicators) use resistors because of cost and don't care about wasting <<1W, high power (e.g. any light) use current drivers because LED brightness is directly proportional to current (within limits). Where a voltage converter converts voltage to constant voltage, LED drivers convert voltage to constant current. Select a device to run from whatever supply you have.

I tested a regular 18650 Li-ion over temperature. Empirically, it was okay down to ~0°C. At -16°C it was absolutely useless until it warmed up then still had 90% of its charge left.

 

[3DDave]

Also unlocking the doors in the event of a crash. People like the doors to unlock.

1/16th the copper = 1/16th the corrosion resistance and far more EMF driving a "poof" of the wire if there is a short rather than damaging a fuse as getting a very fast-acting fuse that is microscopic becomes a problem.

All of those adaptations are great ideas and if they could save even 1 cent per car the automakers would do it. Since they don't do so, it is likely they increase cost directly or by introducing other failure modes that will affect warranty claims.

 

[IRstuff]

WRT the OBD2 requiring 12V because the lawyers say so, if the main supply is 48V, a small POL converter can efficiently supply that small load, no issues.

There's ALWAYS some issue, be it cost or complexity. Given that most automotive accessories are 12V, including replacement batteries, a manufacturer would have to have a serious need for a different voltage to not pick 12V. Note that car costs are absurdly sensitive; manufacturers have done stupid things because it added a few bucks to the base cost.

 

[BrianPetersen]

I don't think 48 volts for vehicle safety-lighting-comfort-convenience is going to become prevalent at any time in the foreseeable future. The prevalence of LED lighting has made electrical loads that are spread around the vehicle (lighting), quite a bit smaller (thus needing less wiring). The copper savings might be more significant for bigger loads, electric power steering was foreseen to be one of them, but it seems that those (with modern servo-motor controls) do fine on 12 volts - and their motors and drives are typically pretty close to where all the rest of the mechanical and electrical bits of the car are located anyhow, so there's not much cable length. Really big motor loads on EVs and hybrids, such as the air-con compressor, are driven at propulsion-battery voltage (usually in the 300 to 400 volt range).

The cost of copper is a factor. Often the answer to "Why is the charging socket located there" for an EV is "Because that's the closest spot to where all the electrical gubbins that make the car work, are located", i.e. that's the spot that has the shortest length of cable to get from the socket to the DC power distribution module and the AC charging module, which is in turn located close to the main power drive inverter and to where the high-voltage connection to the propulsion battery is. Tesla was originally designed as a rear drive vehicle (vehicles without all-wheel-drive, are rear-drive only), so that stuff is in the back beneath the trunk floor, and the charging socket is at the rear. My car (Chevy Bolt) is only available as front drive, so all that stuff is up front, and the charging socket is near the front.

 

[VectorBear]

Also unlocking the doors in the event of a crash. People like the doors to unlock.

1/16th the copper = 1/16th the corrosion resistance and far more EMF driving a "poof" of the wire if there is a short rather than damaging a fuse as getting a very fast-acting fuse that is microscopic becomes a problem.

All of those adaptations are great ideas and if they could save even 1 cent per car the automakers would do it. Since they don't do so, it is likely they increase cost directly or by introducing other failure modes that will affect warranty claims

Any corrosion above zero is bad, beginning of the end. The rule of thumb for marine is that air spaces breath, seals just slow it down, a lot.
Proper fuses are selected by their current vs time curves, but automotive fuses are often picked up at local stores. So yes, point taken about "fuseable wires".
I have intentionally designed in minimal size wires as last resort fail point, such as voltage monitor or high side current monitor. Choice is wire holds overcurrent then heats and melts adjacent insulation, or, flashes and burnt out. Replacement either way, just safer.
Yes, automotive is very cost concious, both production and post-sale.

 

[VectorBear]

I don't think 48 volts for vehicle safety-lighting-comfort-convenience is going to become prevalent at any time in the foreseeable future. The prevalence of LED lighting has made electrical loads that are spread around the vehicle (lighting), quite a bit smaller (thus needing less wiring). The copper savings might be more significant for bigger loads, electric power steering was foreseen to be one of them, but it seems that those (with modern servo-motor controls) do fine on 12 volts - and their motors and drives are typically pretty close to where all the rest of the mechanical and electrical bits of the car are located anyhow, so there's not much cable length. Really big motor loads on EVs and hybrids, such as the air-con compressor, are driven at propulsion-battery voltage (usually in the 300 to 400 volt range).

The cost of copper is a factor. Often the answer to "Why is the charging socket located there" for an EV is "Because that's the closest spot to where all the electrical gubbins that make the car work, are located", i.e. that's the spot that has the shortest length of cable to get from the socket to the DC power distribution module and the AC charging module, which is in turn located close to the main power drive inverter and to where the high-voltage connection to the propulsion battery is. Tesla was originally designed as a rear drive vehicle (vehicles without all-wheel-drive, are rear-drive only), so that stuff is in the back beneath the trunk floor, and the charging socket is at the rear. My car (Chevy Bolt) is only available as front drive, so all that stuff is up front, and the charging socket is near the front.

Stepping back and looking at the requirements, 12V was a bad choice. It was simply based on 6-cells of lead-acid (so why not 5 for 10V?). Those were the days of chunky designs, now engineering has far more finesse. Since now most ancilliaries need step down regulators anyway or designed for a specific voltage (e.g. small motors), 48V is safe and reduces copper, AKA cost. However, status quo will keep 12V around for a while longer.
Like your outlets' spatial explanation.

 

[BrianPetersen]

Plenty of design choices in the modern world might have been done differently if people a hundred years ago had known then what we know now, but they didn't.

 

[IRstuff]

Stepping back and looking at the requirements, 12V was a bad choice. It was simply based on 6-cells of lead-acid (so why not 5 for 10V?).

Plenty of design choices in the modern world might have been done differently if people a hundred years ago had known then what we know now, but they didn't.

Ah, but they might not have been able to make things work. 12V likely was chosen because that was what was required to run the two things that a car needed for operational, its starter motor and its ignition. A lower voltage might not have supplied sufficient power to get the starter motor to turn fast enough to crank the main engine and a lower voltage might not provided sufficient voltage for the spark plugs.

Of course, the other possibility is that 12V batteries were already readily available, and rather than spending money developing a new batttery for what was then a minor application would have be cost prohibitive. Backward compatibility is what ensures forward viability of marketing and sales

 

[3DDave]

Cars started (pun) with 6V. Probably increase electrical demands made 12V more desirable. 6V is plenty to start a car with, but the trade-off is increased plate size to produce the higher current.

I wonder what the trade-off is between increased voltage and increased weight/cost of insulation over decreased cost of copper wire and decreased damage resistance.

 

[BrianPetersen]

A pretty negligible amount of insulation is required to block low voltage like this. Starter motor cables need lots of copper at 6 volts.

The changeover from 6-volt to 12-volt happened through the 1950s and 1960s. Minor digging found that GM switched between 1953 and 1956 (some makes and models before others), Ford and Chrysler in 1956, Volkswagen switched in 1967.

My dad would have had two 6-volt cars - a 1941 Dodge, and a 1951 Chevrolet. He recalled exercises of neighbours getting their cars started to go to work in the morning in winter ... whoever managed to get their car started first, would use it to push-start their neighbours and they would repeat the process until everyone got going ... this was in the days of heavy steel bumpers that could do things like this, and automatic transmissions (if so equipped - few were) that had "rear pumps" capable of back-driving the engine for purposes of bump starting. Modern automatic transmissions won't do that.

On the topic of building wiring to the bare minimum in the interest of saving copper ... I have a 1990 Yamaha motorcycle. If stopped in neutral at a traffic signal with the turn signal flashing, the neutral indicator lamp dims and brightens in opposite-phase of the turn signal flash phase.

 

[3DDave]

I was more referring to the suggested move from 12V to 48V as a way to save copper. It's the sort of move that makes sense in aircraft and is laughable in the salt belt.

 

[VectorBear]

I was more referring to the suggested move from 12V to 48V as a way to save copper. It's the sort of move that makes sense in aircraft and is laughable in the salt belt.

I am in the road salt zone, not a problem since the '90's when cars vastly improved. Is salt spray an issue?

 

[3DDave]

Ah, the sample of one. Look at the other vehicles and see if there is any sign of corrosion on any of the cars around you.

I suppose what we are left with is that all the automakers aren't smart enough to see the potential that 48V or eliminating the secondary battery offers them.

 

[BrianPetersen]

Possibilities: A 48-volt battery is more expensive than a 12-volt battery by more than the amount of copper that would be saved, plus the additional headaches (= costs) of using abnormal not-off-the-shelf wiper motor, HVAC blower, lighting, sunroof motor, power window winder motors, etc etc etc.