Laptops need about 19v, desktops need 12v.

 

Do you have an exemple of a desktop that only need 12V? Usually only mobile PCs use 12V.

Highest voltage rail in a PSU is 12v.

 

Laptops need ~20V to charge the Battery,
which needs to be able to maintain a sufficient operating Voltage even when
running close to a fully dis-charged condition.
.
.
.

 

Highest voltage rail in a PC PSU is more than 300VDC.

What??
Where the heck would it use such a high voltage?
Do you have one that runs on tubes?

 

What??
Where the heck would it use such a high voltage?
Do you have one that runs on tubes?

Only on the primary side of the mains transformer. 220 RMS gets you 311 volts peak. The only way there would be a 300+ volt DC rail is if the desktop used an offline supply, but there is no real reason for doing that.

The brick that comes with a laptop takes in the mains voltage and outputs around +19V, so the laptop does not have to find a way to include a mains transformer. A laptop brick could conceivably use and offline DC-DC converter to reduce the transformer size and weight.

 

Laptops need about 19v, desktops need 12v.

No. Looking at it your way, Laptops need 19 Volts DC, Desktops 220 / 110 Volts AC.
The PSU in a Desktop has mainly +3.3, +5 and + 12 volts for its operation. These are derived from a ATX power supply unit placed the Desktop case.
The Laptop also requires such voltages that are internally derived from the 19 volts fed from a 110 / 220 volts power supply unit.

Putting it crudely, the PSU for a Desktop is internal, for a Laptop, external.

 

Also.... The monitor on a desktop is powered independently. The LCD contrast is usually @ 17v I suppose it just makes it easier to generate the correct voltages used.

 

Highest voltage rail in a PC PSU is more than 300VDC.

Given the TS's obvious lack of knowledge about small computer power systems, this statement is intentionally misleading, serves no useful purpose, and creates an unnecessary distraction in the thread.

ak

 

Given the TS's obvious lack of knowledge about small computer power systems, this statement is intentionally misleading, serves no useful purpose, and creates an unnecessary distraction in the thread.

ak

I disagree. In both cases a higher voltage is used to generate the voltages actually used, and there is good reason for this, using buck converters instead of boost.

Bob

 

A laptop power brick also has rectified AC mains, and therefore approx +300v, inside. Other than the ATX supply generates multiple outputs and the brick only one (typically 18 - 24v) they have more in common than might be supposed.

 

The question of, laptop power supply bricks generally convert mains voltage to 19vdc and send that (and only that) voltage to the laptop. This voltage is used because there are generally four lithium batteries in series (total voltage 4.2 x 4 = 16.8v at full charge) so you need a bit more than that to get decent current flow snd achieve full charge in a reasonable time. The battery voltage then declines from 16.8v over time to about 13v at lowest point. This range of battery voltages is converted to various lower voltages for each module in the laptop via dc/dc converters. I don't know of any peripheral on a laptop that uses 12VDC - a voltage typically used by video cards, fans, sound cards and other power-Hungary units on desktops. On the video card, additional dc/dc converters drop voltages as needed by various sub-units.

On a desktop, the ATX power supply has 3.3, 5 and 12VDC Outputs. There is also a fairly low current -12VDC tap. There is no 300vdc output or other high voltage on an ATX power supply. Most components on a desktop use these voltages (I think the -12VDC is used for an audio amplifier on some motherboards or audio cards).
To answer the core of the question, an off-unit ac/dc adapter is easier and more common to convert to a single voltage to charge the batteries and let the internals of the laptop do the rest.
for a desktop, standardization is key and the three standard output voltages of the ATX power supplies is hits done. Some on/board dc/dc conversions are done for modern chips (down to 1.2 volts or maybe even lower by now).

 

I disagree. In both cases a higher voltage is used to generate the voltages actually used, and there is good reason for this, using buck converters instead of boost.

desktops need 12v.

My read of post #1 is that the TS is asking about motherboard voltages, as if the circuits in a laptop run off of the 19 V supply directly, without additional voltage conversion. He already knows that something converts 240 Vac powerline to +12 Vdc, the largest circuit board operating voltage that he knows of. In that context, throwing around numbers like 300 V with zero explanation about how they differ from what he is asking about is showboating.

And wrong. As long as we're showing off, it is not 300 V. Contemporary ATX supplies are "power factor" corrected (actually, harmonic current corrected). For a universal input supply, the peak voltage into the boost converter can be as high as 338 V, and the DC voltage output from the boost stage is around 350-360 Vdc -ish.

ak

 

TS never mentioned anything about motherboards, in post #3 he talks about rails.


Not wrong at all. That's exactly why I say MORE than 300V in my post. Better than -ish.

Power supply Output rails - he specifically mentioned 19 and 12VDC. Stop justifying a 300v answer - read the OPs question - he never asked about "rails" in his original question or the internal workings of a mains to dc power supply - he only uses the word rails associated with a 12VDC output of a desktop supply when he tries to clarify to your ridiculous comment in post 2. And no, modern (less than 20 years) Mobile PCs don't use 12VDC anywhere as far as I've seen. All the dc input to charge batteries is 19 to 21v and internal modules use 5v, 3.3v or lower for some cpus.

 

Just out of curiosity what happens if you are only on a 120V mains? Will you still get the 300V or better?

Anyways...
Just to restate what others have said and add a little bit. 12V from your power supply is usually just used to power motors and fans, 5V and 3.3V are what the logic use, and many times they are stepped down even further (under 2V even) for processors and maybe memory. There are probably some motors and such that run off the 5V also, but that will vary depending on what is actually in the PC itself. The laptop 19V is just for charging batteries (and possibly some LCD signals). Everything else is stepped down to whatever voltage is required to run the the device.

 

Yes.

Interesting... I learned something today!! Unfortunately I'll never have a use for it other than another tidbit of trivial knowledge. Maybe one day I'll learn and understand how they work, but for now I'm just happy to know what comes out and what I can do with it.

 

Just out of curiosity what happens if you are only on a 120V mains? Will you still get the 300V or better?

Yes.

The most common topology for power factor correction (PFC) is one form of active PFC - a rectifier and regulated boost converter. The boost converter runs at something in the 50-250 kHz range. The output is a constant 350 Vdc (sometimes with significant ripple) over an input voltage range of 85-90 Vac to 240 Vac. The strange part is that it is not a DC/DC circuit.

The incoming AC is full-wave rectified, but *not* filtered. Thus, the input waveform into the boost stage is full-wave rectified AC (what you might call 100% ripple), with the half-sine peaks ranging from around 117 V to 338 V. There are two control loops. The long-term one holds the output voltage at the design point (350 Vdc). The short term one pulse-width modulates the drive to the switching FET such that the input current is proportional to the input voltage. In this way, the average input current waveform is almost exactly in phase with the input voltage waveform, the definition of unity power factor. Because the input current now is sinusoidal, there are no significant higher-order harmonic frequencies.

So, there is one switching FET driven by two control loops simultaneously. Inside the control chip this used to be handled by an analog multiplier. It's been a couple of decades since my last PFC project, so the latest generation controllers might have different magic inside.

ak

 

@AnalogKid thanks for the overview. It makes sense for the most part, but in the interest of keeping the thread on track I'll look into it further and maybe ask a question down the road.