[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: Harddisk clicks -> System freezes/reboots

On Tue, 2007-01-23 at 17:13 +1030, Tim wrote:
> On Tue, 2007-01-23 at 01:12 -0500, Ric Moore wrote:
> > Tim, back to ohm's law, wouldn't low voltage also make it run hot???
> No.  Voltage and current are inter-related.  If a power supply is losing
> the ability to "supply," the current will go down.  Correspondingly, the
> voltage will reduce at the same time (due to lack of supply).
> It's not a see-saw arrangement where one goes up while the other goes
> down, so that the wattage stays the same.  Certainly not in this sort of
> power system.  These try and regulate the current to produce a constant
> voltage.
> Though, if you've got a supply going wonky, where it's starting to pass
> badly filtered AC (DC with lots of ripples), you might get a lower
> voltage reading on a meter, but still be producing enough current so
> that regulators are working harder than usual trying to regulate the
> current.  They'll get hot.
> > With all of the crappy capacitors blowing up and power supplies really
> > needing to be of higher capacity with all the Geez Whiz! things now on
> > the motherboards, I'd sure look hard at that power supply and what
> > it's putting out, from my own experiences. Ric
> Oh yes.  I'd be suspicious of power supplies, but wonky components is
> nothing new (it's been the bane of television sets for 40-50 years).
> Likewise, with inadequate ventilation, whether due to bad design or
> installation.  And it doesn't help when manufactures stick the most
> temperature sensitive components right next to the things that generate
> the most heat.  Guess what, guys, you can use both sides of a PCB:  Put
> the caps on one side, & the hot things with the heatsink on the other!
> I dislike towers, but where I use them, I'm apt to have a fan at the
> bottom front sucking in air, and a fan at the top rear exhausting air.
> It minimises recirculating the same air, and the front of the PC is less
> likely to be blocked.  Also, it avoids sucking air in through the
> drives, as happens with systems that only have an exhaust fan.
> For desktop case a similar strategy applies:  Take air in one side,
> exhaust out the opposite side.  
> I prefer desktop cases, they're less likely to be on the floor sucking
> in dirt, the drives are where you can reach them to change discs, the
> cables between things are shorter (or don't need longer ones to reach -
> we've had towers for years, but mice and keyboards still have too-short
> cables), and you can sit the monitor on top of them (getting up to a
> better height, as well as waste less desk space).  ;-)
> -- 
> (Currently running FC4, in case that's important to the thread)
> Don't send private replies to my address, the mailbox is ignored.
> I read messages from the public lists.
Hi, Tim,
	In switching mode supplies (which is all there is on PC's these days),
the problem with low voltage is that the switching transistors tend to
stay on longer.  At some point the cores can saturate, and in that case
low voltage = high current.  It is one of the big drawbacks to switch
mode supplies.

	As to capacitors, well, temp is not good for them, but again, switching
mode supplies bring their own set of problems, one of which is
transients.  If the supply runs too low a current, the switchers are
never static long enough to develop damping and the transient currents
increase, and a result is higher capacitor currents as the caps try to
compensate for the voltage swings that are occuring at high speed.
Speed = increasing power factor (just like on these new GH processors
soon to be THz processors.

	This situation makes switcher design a real quagmire for the designers.
They must have a large duty cycle range with minimal transients.
Capacitors must have high transient tolerance and yet have large
capacitance and minimal LR components.  At the same time, the cores used
in the transformers (generally torus types) must have high saturation
resistance, good coupling and tight coupling to enhance shielding.  The
transistors must have high switching speed and low on resistance, with
exceedingly good front to back ratios to help avoid leakage power and
transient power heating.

How's that for a design challenge?

Les H

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]