: Thanks for your response. I was just curious if you thought it was common
: for 1.5V devices, for example the Rio MP3 player, to use the MOSFET
: technology you described or rather a charge pump or similar technology?
: The root of my questioning in this matter is that my formal education is in
: Chemical engineering and Computer Science, electronics is just a hobby to
: me, and I suspect I will never have the training or practice to consider
: myself truly competent in it. Perhaps one day I will achieve the level of
: 'hack', but time will tell
: I had always thought, when I further explored the hobby (to me) I would
: discover a series of off the shelf chips (Similar to jumping from TTL to
: CMOS) that would make the existence of these 1.5 volt devices evident, but I
: never did.
Devices like Rio MP3 players are highly integrated, and may
contain as little as a single IC on the board. Regardless, all
contain a CPU/microcontroller (which may be integrated with other things
like audio CODECS/amplifiers, etc, or it may not be) on which I will focus
my attention in this post. This CPU is is almost certainly fabricated
using a low-voltage CMOS process, rather than a charge-pump, followed by a
"high" voltage CPU/microcontroller.
One reason is simply efficiency, which translates to battery life.
No charge pump converter is 100% efficient (I don't think that most are
even close) so that doing things that way would result in a
significantly shorter battery life.
On the economic side of things, like I may have mentioned, CMOS
devices in modern processes cannot tolerate higher voltages (because of
their thinner gate oxides) so if a process with a smaller geometry is used
(which will result in a smaller chip area = lower cost for the
manufacturer = desirable for the manufacturer) lower supply voltages MUST
be used.
Therefore, for this type of application (low-voltage digital CMOS
design) the economic and the technical aspects of the design re-inforce
each other, making it VERY likely for that the be the prevailing method of
design.
Does that make sense? Things ARE more complicated than that, but
this is a "basics" newsgroup, and those are pretty much the basics of what
drives/helps drive modern low-voltage IC design, which has made (and
will continue to make) lots of new low-voltage products available
recently.
Take it easy,
Joe
: :> : I was curious, I have seen a number of devices which simply took a
: single
:> : AAA battery as a power source(Rio MP3 player, some IR remotes etc..).
:>
:> : What is the technology behind these low voltage devices? I thought even
: HCT
:> : CMOS needed 2 volts or higher?
:>
:> : Thanks
:>
:> GTR,
:>
:> Every CMOS technology/fab is different, therefore, it is not
:> correct to generalize the characteristics of HCT (which IS CMOS-based
:> logic) to all CMOS fabs. It's also likely that the different fabs of
:> different manufacturers of HCT devices are different, and thus, the
:> performance of even those two different brands is very different (although
:> it probably wouldn't be speced differently.)
:>
:> That being said, assuming all devices operate in strong inversion,
:> the only constraint that the technology places on the voltages is that,
:> roughly speaking, the supply is greater than the threshold voltage of the
:> devices (Vt.) Practically speaking, some additional headroom would be
:> required, but let's not worry about that for now.
:>
:> Modern CMOS processes (0.18 -> 0.13 -> 0.09 microns ->) typically
:> have Vt values of approximately 0.5V for "normal" devices, but they also
:> have "low-vt" devices with Vt values as low as 0.2V, or even native
:> devices with Vt of 0V (or even negative -- yes, for an NMOS device,
:> meaning that the device is always on.) In fact, many of these devices
:> can be damaged by any voltage significantly higher that 1.5V, due to their
:> very thin gate oxide! All of these different "flavors" of MOSFET devices
:> are useful in low-voltage CMOS design, but they have their problems.
:> Lower Vt means larger leakage current, those native devices that I
:> mentioned above basically can't be turned off, etc, all of which are
:> additional challenges that need to be dealt with by the designer. My
:> point is that modern processes give a designer lots of options that have
:> made the design of very low voltage circuits practical.
:>
:> I am an IC designer with an interest in low-voltage/low-power (not
:> the same thing!) IC design, and I can tell you that a previous design
:> that I worked on, in a modern process, is able to run comfortably at 0.8V.
:>
:> So, 1.5V (which is usally speced down to 1.2V or less to deal with
:> the inevitable droop of the power supply as the battery drains) operation
:> is a piece of cake!
:>
:> Take it easy,
:>
:> Joe
:>
:>