Re: Very Hot Power Cord to Laptop
- From: Barry Watzman <WatzmanNOSPAM@xxxxxxxxxx>
- Date: Wed, 20 Jun 2007 20:10:49 -0400
That's a linear supply. And it does generate heat. First, the transformer creates some heat (how much varies widely) as the windings have resistance, and some current is induced and flows in the iron laminated core of the transformer. Second, as you noted, the regulator of a linear supply regulates by "wasting" the difference between what the transformer puts out and what the load uses as heat (the transformer is ALWAYS designed to put out at least 2.5 to 5 volts more than the load uses, and the [linear] regulator works by "wasting" the difference as heat).
A switching power supply works totally differently; the incoming AC is full wave rectified to DC (150 to 400 volts if it's a line operated supply), then the DC is switched on and off at a very high frequency (20 to 100 KHz) as a "square wave" passed to the primary of a pulse transformer. The output of the pulse transformer is full wave rectified and capacitor filtered (the frequency is so high that tiny capacitors produce perfect filtering, plus you are full wave rectifying a square wave ... it's virtually ripple free to begin with). Also, because of the frequency, the pulse transformer can be small, efficient, low loss.
Regulation is accomplished by comparing the actual output to the desired output and adjusting the duty cycle of the square wave switching on the primary side of the pulse transformer. Since the switching element is either always "on" (near zero voltage across it) or "off" (near zero current flowing through it), very little power is dissipated in a switching power supply.
While you may think otherwise, switching power supplies are virtually always FAR more efficient than linear power supplies. Linear power supplies require an absolute minimum of 3 volts to be dropped across the regulator at all times, and in many cases it's two to three times that much. Taking a 5-volt supply as an example, that means that the absolute best case efficiency is 5/8 (62.5%), but in fact it's often less than 50% (and this only takes into account the loss in the regulator and doesn't include the loss in the transformer, which may be that much all over again). Switching supplies, on the other hand, have efficiencies as high as over 90%, and never lower than the 60%'s (and that's total supply efficiency, not just regulator efficiency).
The efficiency, size, weight and cost ALL favor switching power supplies by a huge margin once you get above about 10-20 watts or so of total power required. That's why there hasn't been a laptop made .... EVER (even 2 decades ago) that used a linear supply. Or, for that matter, a desktop either since back in the S-100 days.
Ron Hardin wrote:
.
Back when I built power supplies, you put in a transformer that produced the
voltages you needed, and there was no heat to speak of generated. I'd tend to
call those linear supplies. If not, what are they called?
Anyway if it's getting very hot under load, I bet it's not a switching supply.
The current drawn is being dropped through some resistance, presumably to get the
right voltage, and there's your heat. A switching supply just increases or decreases
the portion of the cycle used, to keep the sought-after voltage.
- Follow-Ups:
- Re: Very Hot Power Cord to Laptop
- From: Ron Hardin
- Re: Very Hot Power Cord to Laptop
- References:
- Very Hot Power Cord to Laptop
- From: Von Fourche
- Re: Very Hot Power Cord to Laptop
- From: paulmd@xxxxxxx
- Re: Very Hot Power Cord to Laptop
- From: Ron Hardin
- Re: Very Hot Power Cord to Laptop
- From: Barry Watzman
- Re: Very Hot Power Cord to Laptop
- From: Ron Hardin
- Very Hot Power Cord to Laptop
- Prev by Date: Re: Connecting an old Parallel Port laser printer to a new Dimension E520
- Next by Date: Re: Connecting an old Parallel Port laser printer to a new Dimension E520
- Previous by thread: Re: Very Hot Power Cord to Laptop
- Next by thread: Re: Very Hot Power Cord to Laptop
- Index(es):
Relevant Pages
|
Loading