Post by AdamPost by PaulPost by mikeEven 20 years later, there are engineers still fucking up
designs. You find out after it's too late.
The charger industry is proud of abusing electronic
components to come up with cheaper and cheaper
solutions.
This is why my Black and Decker cordless screwdriver
reduced three battery packs to puddles of goo. The
charging solution has absolutely no merit at all
(no charge termination).
On the other hand, I like my car battery charger,
which uses only a transformer and selenium rectifiers
to make a "high impedance" charging circuit. The packaging
claims an amount of current will flow, which is never
achieved. So it's pretty hard to cook a battery (or
charge it quickly) with the charger. But in terms
of construction, they couldn't make it much cheaper -
removing the selenium rectifier thingy would leave
you with only an AC transformer.
If your laptop had NiCd batteries in it, I'm sure they
could have cut a few more corners.
It's the fact that Lithium Ion battery packs are
so dangerous (from a corporate liability point of view),
that a lot more care is put into them. If it wasn't
for Lithium Ion, we might never have seen precision
charging chips.
Paul
Thanks (Guru Paul, et al), for the clarification.
I think everyone is "right" but just saying
the same thing in their own unique ways.
Sorry, I should have been more clear.
Let me rephrase...
For laptops (or more valuable equipment nowadays),
if I stick with a compatible voltage (19V) AC adapter with
sufficient power (90+W), I should be fine since
more and more safety measures (like sensors) are
designed in to protect valuable equipment.
Safety measures designed in is directly proportional to
value of equipment (both increase/decrease together).
Any recommendations on resources (books, magazines, websites, etc.) on
power for newbies?
Most of what I've learned, was by analysing stuff (schematics),
or learning by making mistakes.
To illustrate, there are three kinds of adapters for
consumer electronics. These are general categories.
1) AC adapter (it's just a transformer)
2) Unregulated DC adapter (transformer, rectifiers, filter cap)
DC voltage varies with loading.
3) Regulated DC adapter (SMPS, similar to ATX power supply design,
isolated for shock protection, overcurrent detection with
various behaviors on overcurrent). Complete switch-off being
a common overcurrent behavior). Connect a 2 amp light bulb
to a 2 amp adapter, it will shut off. Why ? The bulb draws
4 amps when it is cold, trips the OC immediately, and the
adapter shuts off.
OK, I bought a label maker one day. It had provision for battery
operation. You were supposed to pour $$$ worth of dry cells into
the thing. An optional adapter was available, at $50 !!!
Well, naturally, being a cheap-skate, I wanted neither dry cells
nor a $50 adapter.
On the housing of the unit, next to the barrel connector for power,
it said "7VDC", and had the symbol for center-positive power. So
I bought a *regulated* 7VDC adapter and plugged it in. I verified
the adapter made exactly 7V, and it did do that. With a fairly high
current rating (so not likely to poop out on OC).
So I try to print a label, and... nothing.
So I eventually break down and buy the optional adapter for $50.
Turns out it is unregulated. At no load, the adapter makes *10V*
and at the instant the motor cranks the label through the print
area, the voltage drops down to 7V due to the increased current
draw.
So it really needed the elevated (unspecified) 10V voltage to make
the keyboard and display and control chip work.
That lesson taught me, that the specification printed on
the housing ("7VDC") could be regulated or unregulated, and
there is no way to know which is appropriate. They kinda got you
by the nuts.
And that's learning by making mistakes.
As for your laptop, I don't see a reason in what I've seen
in the schematic so far, for there to be a dependency on adapter
power rating. Your laptop is the 90W design. A 19V adapter
of 90W or 135W should work. The charging circuit has control
of what it is doing, and is not dependent on external impedance
characteristics. The charger chip has slow start (inrush limit),
so if the adapter is already plugged into the wall and you
shove the barrel into the laptop jack, the adapter doesn't
quit on OC. There is a clamp diode near the jack, so if the
barrel is removed while the adapter is powered, the inductive
kick-back is quenched. The design has a current sampling shunt,
which implies the chip can sense the current. And knows if
too much current is being drawn. And it has a fair number
of MOSFETs to control various things. As long as MOSFETs are
saturated (fully ON or fully OFF), they don't get all that
warm. And that's important. It's easy to burn out a MOSFET
with no heatsink, by turning it half-ON by design. This is
why SMPS circuits take turns with MOSFETs fully ON or fully OFF,
to achieve a desired result. The devices then get warm but
not hot.
There are adapters that have more than two wires, and that
immediately makes me suspicious. It implies some form of
control, or "adapter power rating checking" being done
by the laptop. That tends to happen with stuff above
65W. You're likely to find 65W designs to be pretty simple
and carefree. Anything above that, you should keep your
eyes open, and do a bit of Googling to learn of the
issues.
Paul