• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Why are PSU's so much larger than laptop Power bricks?

S

Shalmanese

Platinum Member
Is it a cost issue or is a PSU more sophisticated than a power brick or what?

They seem to serve the same function yet one is so much larger.

On a side note, would it be possible to mod a laptop power brick to use is a set top box or other very-small computer?
 
well, for one thing, laptop bricks provide MUCH less power. and they only supply one voltage, IIRC, as opposed to -12, -5, 3.3, 5, 12, etc. I think laptops also have more stuff inside to regulate power than just the external brick.
 
laptops only use like 50watt max power. so they dont need all that crap in the power supply for heating and supplying clean power. most of the inside of a power supply is just air and large capacitors for giving a stable charge supply. also big heatsinks . so dont need those with 50 watts .
 
For one thing, laptop power supplies are custom designed to provide just sufficient power for specific models of laptops so they can easily calculate the power necessary to run the computer. ATX and AT PSU's are built to be capable of set output like 300W and is designed to work in wide variety of setups. I think it is possible to build PSU's smaller today than five years ago, but it will be built into same size since there already is a standard for AT and ATX power supply dimensions.
 
I think laptop power supplies may be a little more advanced than PSU since in PSU's the overriding concern is cost and notebook bricks care about cost & portability. Perhaps some size savings can be gained by using more compact capacitors. There are many capacitor types but I think PSU's just use cheap electrolytic ones.
 


<< For one thing, laptop power supplies are custom designed to provide just sufficient power for specific models of laptops so they can easily calculate the power necessary to run the computer. ATX and AT PSU's are built to be capable of set output like 300W and is designed to work in wide variety of setups. I think it is possible to build PSU's smaller today than five years ago, but it will be built into same size since there already is a standard for AT and ATX power supply dimensions. >>



well, the PSU in my IBM PS/2 is giant, and the one in the IBM PC 6150 (from 1985?) is almost the size of a mini tower case (longer and not as tall). neither supplies all that much more than a modern AT/ATX PSU. so they definitely are shrinking.
 
To be honest, I haven't found that PSUs produce a great amount of heat. I unplugged the fan in my computers PSU to get my computer quieter, and it got hot but there were no stability issues. Don't know whether newer power supplies are getting hotter or colder - better efficiency presumably means less loss of power = less heat output, while increasing power demands mean greater wattages so the same % efficiency means a greater output in watts of heat energy.

Still, wouldn't it be possible to say leave the back plate of the PSU where it screws into the case the same, so it would still fit into a standard case, but to halve the depth of the power supply to give a bit of extra room? That would at least help small form factor machines, and might also decrease the necessary depth of a standard ATX case.
 
And to add one more thing, given that the PSU was fine without a fan (although I've now installed a silent 80mm Papst case fan) it should be fine using a little 60mmx10mm fan which would take up considerably less volume than the huge 80x25 fan in there now.
 


<< And to add one more thing, given that the PSU was fine without a fan (although I've now installed a silent 80mm Papst case fan) it should be fine using a little 60mmx10mm fan which would take up considerably less volume than the huge 80x25 fan in there now. >>



well, the PSUs i've played with get REALLY hot without a fan. I haven't measured the voltages, but even if they are good, it can't be good for the life of the PSU.
 
Just to sort of synthesize what everybody's said here:

To add to what Ctho said, I think that a laptop brick is really not much more sophisticated than the bricks that you find with printers, phones, answering machines, etc. Perhaps it does a bit more filtering to give cleaner/more consistent voltage but it's really just a transformer, converting AC wall power to DC (I think mine is 19V.).

The insides of a laptop (i.e. RAM, processor, hdd, etc.) obviously require multiple voltages and the hardware for doing that is just built into the laptop, I guess. As Jerboy said, it's kind of a closed system, so the components can be specifically designed and therefore can be smaller. Also, as Hans said, the overall power requirements are less with a laptop.

I guess that the external brick is like the largest part of a PSU (the toroidial transformer) (well I guess the heatsinks could be larger) and all the other filters, capacitors etc. that divide out the different voltages in a PSU are just built into the motherboard of the laptop. I guess the main benefit of having an external transformer is the heat savings... those things can get pretty hot.
 


<<

<< transformer >>


Don't you mean inverter? 😀 >>


Don't you mean rectifier? 😀

(actually, transformer to lower voltage, then rectifier to convert to DC?)
 
Um, I guess... I have always called power bricks transformers b/c of the voltage conversion that goes on... I guess it's an inverter too, to switch from AC to DC? Is that what an inverter does?

Anybody care to explain how that works? I understand how the coil ratio of a transformer dictates the input/output voltage, but going from AC to DC seems like it'd be more complicated than that.

I'm not really a physics guy... I just pretend to be one sometimes. Thanks for correcting me 🙂
 


<< Um, I guess... I have always called power bricks transformers b/c of the voltage conversion that goes on... I guess it's an inverter too, to switch from AC to DC? Is that what an inverter does?

Anybody care to explain how that works? I understand how the coil ratio of a transformer dictates the input/output voltage, but going from AC to DC seems like it'd be more complicated than that.

I'm not really a physics guy... I just pretend to be one sometimes. Thanks for correcting me 🙂 >>



There are a few different types of power supplies. Linear dropper, unregulated transformer, and switching power supply.


unregulated transformers, rectifier and smoother setups are usually used for something not too picky about voltage and small electronics with built in regulator.

Regulated transformers are called linear dropper. It has an unregulated DC power supply and an electronic circuit to adjust and regulate the voltage. The voltage regulator can't add voltage. It regulates it by cutting off the excess voltage through resitivity(using transistor) and difference is wasted as heat.

Switching power supply uses high frequency switcher.

Transformer setup is the cheapest for very small setup and extremely large setup(locomotive converter station and such), but otherwise they're expensive.

Transformers opeating at 60Hz must be very big and there is no easy of controlling the output voltage other than linear regulator or motorozed Variac(very expensive). A 350W transformer based PSU that outputs the samething as computer PSU will be as large as the whole computer and will need a forced air cooling of its own.

Swithing power supply first converts the mains line into pure DC. I wrote up an explanation, but it sucks and I don't feel like revising it so read all about it here.

Switching power supplies are controlled through many semiconductors that are placed under very high stress and by nature, they're more prone to failure than 60Hz iron core transformer based power supplies.


 
Wow, thanks, that's very interesting. The link was over my head, but your explanation was essentially everything I wanted to know.
 
Jerboy, just one correction to your well said comments-

Switching power supply components are indeed not under any more stress than a linear power supply, if anything they are under considerably less stress due to lower heat generation but the simple fact exists that for a switcher you need at least 50 components for a well regulated device (many of these are now included in a control IC) and a linear power supply needs merely three components - transformer, rectifier and capacitor. Add zener diode and pass transistor for a regulated linear power supply so 5 components versus 50 (usually a lot more) the probability of failure is ten times higher going by general component failure rates, not a stressed component failure rate.

I do repair laptops, in virtually all modern laptops the "brick" supplies a single voltage and the laptop has its own switchmode power supply (DC-DC converter) inside it. This power supply has a standby rail which is typically a linear power supply that feeds the standby controller power to allow the controller to fire up the motherboard and the rest of the laptop when you hit the power button, sort of like the soft power used with ATX systems. The remainder of the voltage rails are all switchmode, including the cold cathode flourescent backlight in the screen which the inverter generates many thousand volts to illuminate it.

Cheers
 


<< Jerboy, just one correction to your well said comments-

Switching power supply components are indeed not under any more stress than a linear power supply >>



I say some components notably the main choppers are placed under much stress, because it must directly chop the 170V DC derived from mains meaning it will be put under considerable inductive spikes and surge.

Linear supply usually only use semiconductor control on the secondary to drop the voltage and it usually doesn't involve stress caused by inductive spikes.

How often do LM7805 used in battery powered logic circuit fail compared to thyristors used in inverter to control high voltage or very high current to a transformer?

If they're not under more stress, what explains primary side of isolated switch mode power supply fails more often than secondary side? I used to troubleshoot surplus swithing power supplies for educational purpose and usually failures are in primary side MOSFETS or rectifier.
 
jerboy: If the common fault is the chopper then the supply is underengineered, plain and simple. Either a poorly made part, poorly designed circuit to produce excessive stress to the common failing component or simply an underrated component for the job.

There is no reason the supply (or any electronic device) should not keep working for 20+ years but usually the first thing that fails is electrolytic capacitors. Look at many old televisions, 95% of them use switchmode power supplies. Often the reason the chopper transistor blows is because of another component failing first, this is almost universally the situation. Only on rubbish quality products do the choppers fail first, usually because of an underrated part.
 


<< jerboy: If the common fault is the chopper then the supply is underengineered, plain and simple. Either a poorly made part, poorly designed circuit to produce excessive stress to the common failing component or simply an underrated component for the job. >>




You got a point.. Probably intentional too.

Theoretical peak voltage: 150V

Components available 160V($1 each) or 250V($1.10 each).

Engineer: 160V is barely scraching by. 250V components would give an adequate headroom and is preferred.

Sales: hey let's go with 160V so we'll make extra 10 cents on each PSU, who cares about stuff that breaks after warranty is over?

 
You must log in or register to reply here.

TRENDING THREADS

Back
Top