16
\$\begingroup\$

This question has two parts:

1) How inefficient is it to boost 12V to 120V and then back to 12V as in using a traditional car power inverter to power a laptop (i.e. the 12V car battery power is boosted to 120V by a power inverter and then back to 12V by the laptop's power supply)?

2) Is there any way to power a laptop directly from a 12V car battery? This would be useful not only for use in a car, but also for a solar-powered home that runs on 12V batteries. If there is a significant gain in not going through the boost/buck cycle of power inverters, then it would seem wise to power laptops and other 12V devices directly from battery power. I realize that laptops have different power supply ratings and some require more than 12V, but it seems rather wasteful to boost everything to 120V before bringing it back down.

\$\endgroup\$
  • 9
    \$\begingroup\$ Just an FYI, the battery terminals on your car are not 12V DC like you'll get from your laptop power supply. They're more like 13.8 when the alternator is running, they dip down to 10.5V or lower when starting the car, and they have all sorts of noise from the spark plugs, alternator, and various motors. PSU design for automobiles is tricky business. \$\endgroup\$ – Kevin Vermeer Feb 13 '11 at 20:52
  • 2
    \$\begingroup\$ Thanks for the tip. Actually I'm more interested in using car batteries with solar cells. \$\endgroup\$ – Reed G. Law Feb 14 '11 at 14:15
  • \$\begingroup\$ @Kevin, how do these voltages compare when the 12v car battery is in a fully electric car like the Nissan Leaf? Would you just get 12v flat? \$\endgroup\$ – Gabriel Fair Mar 22 '16 at 17:50
14
\$\begingroup\$

Yes, tons of power is wasted going from 12V to 110V, especially when all you do is to stick it into a psu which also loses some power turning it back into low voltage DC.

You can buy a DC/DC converter which will deliver a 9-20 V DC adjustable voltage when given 10-24 V DC input.

I've built a SEPIC style converter before, for just this sort of thing: http://dren.dk/carpower.html

\$\endgroup\$
  • \$\begingroup\$ What is the easiest way to step up DC voltage? \$\endgroup\$ – Johnes Thomas Jun 30 '11 at 22:24
  • 1
    \$\begingroup\$ @JohnesThomas, A boost converter, but easy is always relative. \$\endgroup\$ – Kortuk Jul 8 '11 at 6:51
7
\$\begingroup\$

Some laptops can run off variable power sources, usually older ones.

DC-DC adapters lose 20% in their basic conversion from 12V to 19V.(Tested myself with multimeter), vs 40+% or more to power 110V inverter to run AC adapter to output 19VDC.

I would like to find a newer already EFFICIENT laptop that has been built to take a 11-15.5V DC power input... One older computer I do have(too slow) is a NEC Daylite, 16V powersupply usually but it will run off 11-16V no problem. about 7-11Watts depending on usage.

\$\endgroup\$
  • 1
    \$\begingroup\$ Any updates on your search since you posted this? \$\endgroup\$ – Gabriel Fair Mar 22 '16 at 17:55
5
\$\begingroup\$

It's possible your laptop manufacturer already has an auto/airplane vehicle adapter accessory that you can buy. It will convert the battery voltage to the proper input voltage for the laptop directly (DC-DC). Perhaps it will serve double duty and be able to take AC (wall) input as well.

Here's an example from Dell:
http://accessories.dell.com/sna/productdetail.aspx?c=us&l=en&s=dhs&cs=19&sku=310-8814#Overview

Here's a generic version from Duracell that fits most major laptop brands:
http://www.amazon.com/DURACELL-Universal-Adapter-Interchangeable-DRACDC5101/dp/B003ICXALS

\$\endgroup\$
3
\$\begingroup\$

I have been doing exactly this at my home powered by solar panels. When building the house I had the electrician run DC wiring to certain points. Since mine is a 24 volt system I use a buck converter to get 19V to power both my Asus laptop as well as an LG external LED monitor (again 19V). There are many low price (but decent quality from what I can tell) buck converters available to do this. Note that in my case I had to use a bridge rectifier (BR68) before the buck converter to filter out low AC ripple from the DC line. I think the ripple was getting in due to AC/DC lines sharing some wall conduit. Here is one I found.

\$\endgroup\$
  • \$\begingroup\$ a bridge rectifier won't remove ripple. \$\endgroup\$ – Jasen May 4 '18 at 10:57
2
\$\begingroup\$

It's mostly a myth that it's much more efficient to power DC appliances like laptops on a full end-to-end DC system rather than using an inverter and then the existing AC-DC converter instead1.

Let's take a look at your first question:

1) How inefficient is it to boost 12V to 120V and then back to 12V as in using a traditional car power inverter to power a laptop (i.e. the 12V car battery power is boosted to 120V by a power inverter and then back to 12V by the laptop's power supply)?

It depends on your hardware, but it's not too terrible. You have two primary conversions: the DC -> AC conversion in the inverter and the AC -> DC conversion in the power supply for the appliance.

Most modern quality inverters are over 90% efficient and many approach 95% efficiency over a large part of their operating range. Very cheap or small inverters may be worse, perhaps in the low 80s and even good inverters are often less efficient when operating at very low power relative to their rated power.

For the AC -> DC side you'll find more variance. Some quality converters e.g., those supplied with some name-brand laptops approach 90% efficiency, but many others are in the 70% to 80% range. Very small AC -> DC converters, such as those found in USB plugs tend to be slightly less efficient than converters will fewer space constraints.

Overall then, you're looking at a best-case loss of perhaps 15% (a 95% efficient inverter with a 90% efficient power supply) to a worst-case loss with a reasonable inverter of perhaps 40% (an inverter in the high 80s combined with a 70% power supply2.

Now consider also that the "end-to-end" DC path will generally also need a DC-DC conversion unless the device happens to operate exactly at the voltage (say 12V or 24V) of your DC system. This conversion is likely to be, at best, as efficient as one of the above conversions. At worst, if you buy one of the various adjustable buck/boost converters with wide input and output ranges, the efficiency might be considerably lower if it is operating outside of its ideal range. So ignoring all the other factors, it is even possible that the full DC route is already less efficient than AC!

Still, let's assume that the full DC path is theoretically somewhat more efficient than the DC-AC-DC path, by perhaps 10%. Here are downsides of a full DC path that might outweigh that small advantage:

  • Something like a home (or RV or whatever) as you mention in point (2) will already have existing 120V wiring: power appliances on a full DC system would require either locating those appliances very close to battery bank, or running a second DC wiring system at considerably effort (adding wiring to an existing house is a lot harder than doing it as it's being built - unless you don't mind ugly). Furthermore, you'll run into issues such as no standard outlet for DC power (the cigarette lighter is probably the closest widely supported thing, but unsuitable for many purposes).
  • Lower voltages are inherently less efficient than higher voltages for transmission: both because a given absolute voltage drop represents a higher relative fraction of the total voltage, and because proportionally more current is needed to deliver the same power. This effect is roughly quadratic: a 12V system suffers approximately 100 times the voltage drop as wires of the same gauge at 120V of the same gauge to deliver the same power. An example: over 10 feet 14 AWG household wiring, for a load of 120W, a 120V system needs 1 amp and suffers a voltage drop of 0.042% - basically a rounding error. A 12V appliance of the same power would need 10 amps, and suffer a voltage drop of 4.2% - so over 10 feet of 14 AWG you've already lost about as much power as you'd lose in a good inverter. In a house, you could easily have wiring runs of 50 or 100 feet, resulting in DC voltage drops that make the system unsustainable - even with a small 120W load. In practice, you'd need to use a significantly larger gauge of wire to counteract this: a significant cost which could instead just be spent on more solar panels or batteries.
  • AC is the default: almost every appliance you buy will by default come with an AC plug. There are all sorts of appliances where you can also buy a DC version, but often with a greatly reduced selections. Yes, you can buy a DC powered fridge, but you pretty much have to choose from the 1 or 2 weird models at your local solar/battery store. These are often twice the price of a fridge you'd buy anywhere else, and based on some old model that may inherently be less efficient. The same for DC powered fans, TVs, coffee makers, whatever. Yes, they exist, but the market is currently minuscule as the selection follows. You'll waste more money and be less happy with what you end up with than you'll ever save in "AC conversion losses". The one approach that does work here is getting things that run on AC but have an external AC-DC power brick: you can skip the brick and connect your DC system up directly (but again the voltages are usually weird things like 17V, 21V, etc, so you still end up needing a conversion).

So I'll be what seems like the lone voice here and say that any sort of large or medium sized "DC system" doesn't really make sense just to save on conversion losses when you are hooking up off-the-shelf appliances. 120V AC is actually a pretty reasonable method of power distribution, especially since it's the default input for almost everything you'd buy. The conversion losses are fairly small with modern equipment, and you usually can't avoid conversion losses entirely even with a full DC system.


1 I'll sometimes call this the DC-AC-DC approach.

2 Of course, you can push the worst case much further if you seek out a really inefficient inverter (but this is under your control) and find some device with a terrible (or just old) SMPS or linear regulator that is very inefficient.

\$\endgroup\$

protected by Chris Stratton Oct 10 '16 at 22:15

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

Not the answer you're looking for? Browse other questions tagged or ask your own question.